New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Old-Fashioned Audio Radio Transmission (Old-FART) #2

Open
Fastie opened this Issue Aug 13, 2012 · 52 comments

Comments

Projects
None yet
2 participants
@Fastie
Member

Fastie commented Aug 13, 2012

Image

Background

After a flurry of emails, Don is running with a plan for radio transmission of the sound made when an airborne camera takes a photo. The radio transmitter will be a hacked $10 walkie-talkie. Bubble-pack pairs of FRS/GMRS radios are sold everywhere and have a range of more than a mile. If he can extract the microphone, transmitter, antenna, and batteries (3 AAA) from one radio and configure them to ride on a KAP rig, then the other radio can be used on the ground to monitor the sound of the aerial rig. It should then be possible to know if the camera has stopped working.

This has always seemed like the simplest way to monitor a remote camera. How hard can it be to transmit audio 1000 feet? We don’t yet know whether an FRS/GMRS radio is the best way to do this. These radios are “push to talk” (PTT) which means they only transmit when you push the big PTT button. That button will have to be pushed permanently on the KAP rig, with unknown consequences for battery life. A solution would be to add a voice activation unit, so transmission only occurred when the camera made a noise.

Hardware inside the handheld units includes a receiver, amplifier, speaker, headphone jack, channel selector, volume control, keypad lock, low battery alert, call tone selector, paging alert, and channel scanner. None of that is needed on the KAP rig, but a lot of it is probably on a single PCB and can’t be removed. So it might be a challenge to reduce the weight enough to make it a feasible solution.

We also don’t know whether it is legal to modify an FCC approved radio device. The 14 FRS channels are free for anyone to use without a license, but the GMRS channels are not. Surprisingly, if you turn the dial on these ubiquitous radios to a GMRS channel and push PTT, you are breaking the law unless you have paid $80 for a no-exam license. There is a good 2005 article about this issue. And the GMRS service is described here.

Summary Results

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 14, 2012

Member

Chris -- your research notes are the most thorough, and your acronyms are the most entertaining.

  • Seems like it's quite difficult to transmit a non-voice signal 1000 feet legally; but I wonder if there's some provision in the law to do with how populated the surrounding area is? E.g., in Alaska, would you be forgiven?
  • Is there some PLOTS-approved/suggested Powershot model (cheap, easy-to-modify-for-IR, fully supported by CHDK) that you could point me to? I made some meager H2 balloons last night, and I think that with a little work, I could test the FRS audio at 1000 feet by trying to receive the sound of the Powershot "click" sound (is that what you'd want to be monitoring?)
  • The circuitry of the Motorola FV300, minus the plastic case, seems to be quite light. It requires 3 AAAs to operate, but I think it should be possible to power it with an external battery pack as well, extending the battery life.
  • Voice activation doesn't seem built-in on this model, but perhaps we could trigger the included "paging" button (visible as the button with the single musical note icon, on the front of the device) electrically. E.g., we could have an on-board Arduino monitor battery level, and send a "bleep" to the ground every time it the battery is drained?
Member

dwblair commented Aug 14, 2012

Chris -- your research notes are the most thorough, and your acronyms are the most entertaining.

  • Seems like it's quite difficult to transmit a non-voice signal 1000 feet legally; but I wonder if there's some provision in the law to do with how populated the surrounding area is? E.g., in Alaska, would you be forgiven?
  • Is there some PLOTS-approved/suggested Powershot model (cheap, easy-to-modify-for-IR, fully supported by CHDK) that you could point me to? I made some meager H2 balloons last night, and I think that with a little work, I could test the FRS audio at 1000 feet by trying to receive the sound of the Powershot "click" sound (is that what you'd want to be monitoring?)
  • The circuitry of the Motorola FV300, minus the plastic case, seems to be quite light. It requires 3 AAAs to operate, but I think it should be possible to power it with an external battery pack as well, extending the battery life.
  • Voice activation doesn't seem built-in on this model, but perhaps we could trigger the included "paging" button (visible as the button with the single musical note icon, on the front of the device) electrically. E.g., we could have an on-board Arduino monitor battery level, and send a "bleep" to the ground every time it the battery is drained?
@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 14, 2012

Member
  • I doubt that there is such a provision. But go just 1 km from anybody with a scanning radio and nobody will have any idea that you are transmitting at all. I think that is why there is so much illegal radio gear for sale.
  • The PLOTS IR camera is a Powershot A495 which can be found refurbished or used for $70. The shutter sound is a fairly loud chirp which will be easy to pick up. You could test with any number of beeper gadgets, the lighter the better since you have to make more H2 for every gram you add to the payload. But you clearly should have your own aerial camera, so consider a refurbished S95 for only $279. Not recommended for IR conversion, but an excellent KAP camera.
  • Three AAA could be plenty of power, especially if VOX (voice activation) is used. Also, 3 AAA is what I usually use to power the onboard RC receiver or controller, so it could be shared, saving much weight. It has to last only a couple of hours.
  • It looked like the microphone that came with the headset in one of your photos had a VOX mode. If that mic is used to pick up the camera sound, it could be totally tubular. I think you are correct that the paging feature could be used to send alerts of anything an Arduino can detect. But for this purpose, why add an Arduino when the walkie-talkie can transmit all the information you need?
Member

Fastie commented Aug 14, 2012

  • I doubt that there is such a provision. But go just 1 km from anybody with a scanning radio and nobody will have any idea that you are transmitting at all. I think that is why there is so much illegal radio gear for sale.
  • The PLOTS IR camera is a Powershot A495 which can be found refurbished or used for $70. The shutter sound is a fairly loud chirp which will be easy to pick up. You could test with any number of beeper gadgets, the lighter the better since you have to make more H2 for every gram you add to the payload. But you clearly should have your own aerial camera, so consider a refurbished S95 for only $279. Not recommended for IR conversion, but an excellent KAP camera.
  • Three AAA could be plenty of power, especially if VOX (voice activation) is used. Also, 3 AAA is what I usually use to power the onboard RC receiver or controller, so it could be shared, saving much weight. It has to last only a couple of hours.
  • It looked like the microphone that came with the headset in one of your photos had a VOX mode. If that mic is used to pick up the camera sound, it could be totally tubular. I think you are correct that the paging feature could be used to send alerts of anything an Arduino can detect. But for this purpose, why add an Arduino when the walkie-talkie can transmit all the information you need?
@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 15, 2012

Member

Bad news / good news re: VOX:

  • Unless I'm doing something wrong, I can't get VOX mode to work with the
    FV300; and the FV300 doesn't advertise it as a feature ... so I'm assuming
    it might just not work.
  • What does happen when one uses the VOX switch seems to be the
    equivalent of having the push-to-talk button constantly pressed. So this
    is a second-best (unfriendly-to-battery) option: having the mic constantly
    on. Come to think of it, unless there was a way to set the VOX threshold
    on the fly, it might have anyway been a bit tricky to make sure that only
    every camera click registered via VOX, rather than e.g. bursts of wind, or
    nothing at all ...
  • Still waiting to find a local friend willing to go and stand on the
    opposite end of a football field and try things out. Will post on
    Facebook, then on Craigslist Personals.

On Tue, Aug 14, 2012 at 2:31 PM, Fastie notifications@github.com wrote:

  • I doubt that there is such a provision. But go just 1 km from
    anybody with a scanning radio and nobody will have any idea that you are
    transmitting at all. I think that is why there is so much illegal radio
    gear for sale.

  • The PLOTS IR camera is a Powershot A495 which can be found
    refurbished or used for $70. The shutter sound is a fairly loud chirp which
    will be easy to pick up. You could test with any number of beeper gadgets,
    the lighter the better since you have to make more H2 for every gram you
    add to the payload. But you clearly should have your own aerial camera, so
    consider a refurbished S95 for only $279http://shop.usa.canon.com/webapp/wcs/stores/servlet/ProductListingViewAll_10051_10051_-1_12163.
    Not recommended for IR conversion, but an excellent KAP camera.

  • Three AAA could be plenty of power, especially if VOX (voice
    activation) is used. Also, 3 AAA is what I usually use to power the onboard
    RC receiver or controller, so it could be shared, saving much weight. It
    has to last only a couple of hours.

  • It looked like the microphone that came with the headset in one of
    your photos had a VOX mode. If that mic is used to pick up the camera
    sound, it could be totally tubular.


    Reply to this email directly or view it on GitHubhttps://github.com//issues/2#issuecomment-7735482.

voice / SMS: +1-651-252-4765
skype: dwingateb

Member

dwblair commented Aug 15, 2012

Bad news / good news re: VOX:

  • Unless I'm doing something wrong, I can't get VOX mode to work with the
    FV300; and the FV300 doesn't advertise it as a feature ... so I'm assuming
    it might just not work.
  • What does happen when one uses the VOX switch seems to be the
    equivalent of having the push-to-talk button constantly pressed. So this
    is a second-best (unfriendly-to-battery) option: having the mic constantly
    on. Come to think of it, unless there was a way to set the VOX threshold
    on the fly, it might have anyway been a bit tricky to make sure that only
    every camera click registered via VOX, rather than e.g. bursts of wind, or
    nothing at all ...
  • Still waiting to find a local friend willing to go and stand on the
    opposite end of a football field and try things out. Will post on
    Facebook, then on Craigslist Personals.

On Tue, Aug 14, 2012 at 2:31 PM, Fastie notifications@github.com wrote:

  • I doubt that there is such a provision. But go just 1 km from
    anybody with a scanning radio and nobody will have any idea that you are
    transmitting at all. I think that is why there is so much illegal radio
    gear for sale.

  • The PLOTS IR camera is a Powershot A495 which can be found
    refurbished or used for $70. The shutter sound is a fairly loud chirp which
    will be easy to pick up. You could test with any number of beeper gadgets,
    the lighter the better since you have to make more H2 for every gram you
    add to the payload. But you clearly should have your own aerial camera, so
    consider a refurbished S95 for only $279http://shop.usa.canon.com/webapp/wcs/stores/servlet/ProductListingViewAll_10051_10051_-1_12163.
    Not recommended for IR conversion, but an excellent KAP camera.

  • Three AAA could be plenty of power, especially if VOX (voice
    activation) is used. Also, 3 AAA is what I usually use to power the onboard
    RC receiver or controller, so it could be shared, saving much weight. It
    has to last only a couple of hours.

  • It looked like the microphone that came with the headset in one of
    your photos had a VOX mode. If that mic is used to pick up the camera
    sound, it could be totally tubular.


    Reply to this email directly or view it on GitHubhttps://github.com//issues/2#issuecomment-7735482.

voice / SMS: +1-651-252-4765
skype: dwingateb

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 15, 2012

Member

That's too bad about VOX. But it sounds suspicious. The VOX circuitry should be in the microphone that has the VOX switch.

From this document:

  1. If you selected VOX operation, set VOX/MIC for best operation as
    follows:

In a noisy environment, set VOX/MIC to L (low). This prevents
surrounding noise from keeping the transceiver in the transmit mode.

You must speak louder into the microphone at this setting to make the
transceiver transmit.

In a quieter area, set VOX/MIC to M (medium) or H (high). At these
settings, you do not have to speak as loudly to switch to transmit.

Member

Fastie commented Aug 15, 2012

That's too bad about VOX. But it sounds suspicious. The VOX circuitry should be in the microphone that has the VOX switch.

From this document:

  1. If you selected VOX operation, set VOX/MIC for best operation as
    follows:

In a noisy environment, set VOX/MIC to L (low). This prevents
surrounding noise from keeping the transceiver in the transmit mode.

You must speak louder into the microphone at this setting to make the
transceiver transmit.

In a quieter area, set VOX/MIC to M (medium) or H (high). At these
settings, you do not have to speak as loudly to switch to transmit.

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 15, 2012

Member

Hmm ... no apparent "Low" or "High" settings on the mic apparatus I've got
here. I'll have to check the part # on the package and see whether it
matches up with the doc you linked to ...

Aside: these walkie talkies are fantastic when shopping at Lowes
accompanied by a friend with a vastly different browsing agenda.

On Tue, Aug 14, 2012 at 9:54 PM, Fastie notifications@github.com wrote:

That's too bad about VOX. But it sounds suspicious. The VOX circuitry
should be in the microphone that has the VOX switch.

From this documenthttp://support.radioshack.com/support_electronics/doc16/16426.htm
:

If you selected VOX operation, set VOX/MIC for best operation as
follows:

In a noisy environment, set VOX/MIC to L (low). This prevents
surrounding noise from keeping the transceiver in the transmit mode.

You must speak louder into the microphone at this setting to make the
transceiver transmit.

In a quieter area, set VOX/MIC to M (medium) or H (high). At these
settings, you do not have to speak as loudly to switch to transmit.


Reply to this email directly or view it on GitHubhttps://github.com//issues/2#issuecomment-7746122.

voice / SMS: +1-651-252-4765
skype: dwingateb

Member

dwblair commented Aug 15, 2012

Hmm ... no apparent "Low" or "High" settings on the mic apparatus I've got
here. I'll have to check the part # on the package and see whether it
matches up with the doc you linked to ...

Aside: these walkie talkies are fantastic when shopping at Lowes
accompanied by a friend with a vastly different browsing agenda.

On Tue, Aug 14, 2012 at 9:54 PM, Fastie notifications@github.com wrote:

That's too bad about VOX. But it sounds suspicious. The VOX circuitry
should be in the microphone that has the VOX switch.

From this documenthttp://support.radioshack.com/support_electronics/doc16/16426.htm
:

If you selected VOX operation, set VOX/MIC for best operation as
follows:

In a noisy environment, set VOX/MIC to L (low). This prevents
surrounding noise from keeping the transceiver in the transmit mode.

You must speak louder into the microphone at this setting to make the
transceiver transmit.

In a quieter area, set VOX/MIC to M (medium) or H (high). At these
settings, you do not have to speak as loudly to switch to transmit.


Reply to this email directly or view it on GitHubhttps://github.com//issues/2#issuecomment-7746122.

voice / SMS: +1-651-252-4765
skype: dwingateb

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 15, 2012

Member

Oh well. If hacking was easy everybody would already have chirp activated audio from aerial cameras.

You were in a Lowes parking lot and didn't do a line of sight range test??????

Member

Fastie commented Aug 15, 2012

Oh well. If hacking was easy everybody would already have chirp activated audio from aerial cameras.

You were in a Lowes parking lot and didn't do a line of sight range test??????

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 15, 2012

Member

Egads! I suppose I was too taken with how my new shatterproof eyewear
looked in my rear view mirror to think of it -- total fail.

On Tue, Aug 14, 2012 at 10:07 PM, Fastie notifications@github.com wrote:

Oh well. If hacking was easy everybody would already have chirp activated
audio from aerial cameras.

You were in a Lowes parking lot and didn't do a line of sight range
test??????


Reply to this email directly or view it on GitHubhttps://github.com//issues/2#issuecomment-7746252.

voice / SMS: +1-651-252-4765
skype: dwingateb

Member

dwblair commented Aug 15, 2012

Egads! I suppose I was too taken with how my new shatterproof eyewear
looked in my rear view mirror to think of it -- total fail.

On Tue, Aug 14, 2012 at 10:07 PM, Fastie notifications@github.com wrote:

Oh well. If hacking was easy everybody would already have chirp activated
audio from aerial cameras.

You were in a Lowes parking lot and didn't do a line of sight range
test??????


Reply to this email directly or view it on GitHubhttps://github.com//issues/2#issuecomment-7746252.

voice / SMS: +1-651-252-4765
skype: dwingateb

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 16, 2012

Member

Attempt at redemption: today Marcelo and I tested the range of the FV300. We went to the Amherst High School athletic fields, and got into position 1300 feet apart. Detailed results, with map and pictures, are here, but the punchlines were (for a no-wind situation):

  • Audio transmission is still pretty good at 1300 feet, line-of-sight;
  • An iPhone camera "shutter click" sound, even at minimal volume, is easily discernible if the phone is held up right against the FV300 microphone;
  • The shutter click is still discernible at a hand's length away from the FV300 mic, if the iPhone is at maximum volume.
Member

dwblair commented Aug 16, 2012

Attempt at redemption: today Marcelo and I tested the range of the FV300. We went to the Amherst High School athletic fields, and got into position 1300 feet apart. Detailed results, with map and pictures, are here, but the punchlines were (for a no-wind situation):

  • Audio transmission is still pretty good at 1300 feet, line-of-sight;
  • An iPhone camera "shutter click" sound, even at minimal volume, is easily discernible if the phone is held up right against the FV300 microphone;
  • The shutter click is still discernible at a hand's length away from the FV300 mic, if the iPhone is at maximum volume.
@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 16, 2012

Member

Excellent results. And really good methods and discussion, too. So the $20 pair of radios is up to the task.

The only remaining issue is power.

  • How long will the battery last with PTT pushed?
  • Is it possible to implement VOX?
  • Can a transmission up to the airborne receiver turn on the airborne transmitter? When a handset is paged, is enough current produced to trip a relay? If so, it should be possible to toggle the airborne transmitter on and off by hitting the page button on the ground-based handset.

BTW, were you using an FRS or a GMRS channel during your tests?

Member

Fastie commented Aug 16, 2012

Excellent results. And really good methods and discussion, too. So the $20 pair of radios is up to the task.

The only remaining issue is power.

  • How long will the battery last with PTT pushed?
  • Is it possible to implement VOX?
  • Can a transmission up to the airborne receiver turn on the airborne transmitter? When a handset is paged, is enough current produced to trip a relay? If so, it should be possible to toggle the airborne transmitter on and off by hitting the page button on the ground-based handset.

BTW, were you using an FRS or a GMRS channel during your tests?

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 17, 2012

Member

Great idea re: triggering a relay. I'll look into it the paging mechanism.

Interesting twist: I set up a battery life experiment -- chose new batteries, starting an audio recorder right next to the receiving handset, and taped down the PTT button on the transmitter. After about 30 sec, the receiver emits a high pitched tone. This happens consistently. I imagine it's to prevent naughty people from using these devices for espionage (with the unfortunate side effect of preventing awesome people from using them as camera operation mointors).

Image

So, now we start to move into more interesting territory. I'll explore ways of triggering the paging button, and your question re: the possibility of tripping a relay.

I'll also check and see whether we were broadcasting on FRS or GMRS. Is there an expected difference in transmission power / range?

Member

dwblair commented Aug 17, 2012

Great idea re: triggering a relay. I'll look into it the paging mechanism.

Interesting twist: I set up a battery life experiment -- chose new batteries, starting an audio recorder right next to the receiving handset, and taped down the PTT button on the transmitter. After about 30 sec, the receiver emits a high pitched tone. This happens consistently. I imagine it's to prevent naughty people from using these devices for espionage (with the unfortunate side effect of preventing awesome people from using them as camera operation mointors).

Image

So, now we start to move into more interesting territory. I'll explore ways of triggering the paging button, and your question re: the possibility of tripping a relay.

I'll also check and see whether we were broadcasting on FRS or GMRS. Is there an expected difference in transmission power / range?

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 17, 2012

Member

Rats, foiled by the time-out timer. I think the TOT is intended to prevent battery drain if the PTT button gets held down accidentally e.g., in a pocket. Which suggests that the battery might not last long in continuous transmit mode, so that approach might not have worked anyway. Also, we don't really want to listen to the flying camera click every 10 seconds, we just want to tune in now and then, especially during the end of a flight, to make sure the camera is still working.

I assume that if you find the correct relay, it can be triggered by whatever current is travelling to the speaker when that handset is being paged by the other handset. So replace the speaker with the relay and wire the relay output to the PTT button leads. Our own Quindar tones.

GMRS is definitely capable of greater transmission range than FRS. The 10 mile range advertised for the FV300 is for the GMRS channels. FRS should be good for a few miles, so plenty for KAP work.

Member

Fastie commented Aug 17, 2012

Rats, foiled by the time-out timer. I think the TOT is intended to prevent battery drain if the PTT button gets held down accidentally e.g., in a pocket. Which suggests that the battery might not last long in continuous transmit mode, so that approach might not have worked anyway. Also, we don't really want to listen to the flying camera click every 10 seconds, we just want to tune in now and then, especially during the end of a flight, to make sure the camera is still working.

I assume that if you find the correct relay, it can be triggered by whatever current is travelling to the speaker when that handset is being paged by the other handset. So replace the speaker with the relay and wire the relay output to the PTT button leads. Our own Quindar tones.

GMRS is definitely capable of greater transmission range than FRS. The 10 mile range advertised for the FV300 is for the GMRS channels. FRS should be good for a few miles, so plenty for KAP work.

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 18, 2012

Member

Hi Chris!

Oh gosh -- you led to some very interesting reading re: Quindar tones. I guess we are launching satellites, after all; and figuring out how to ping them for information is very NASA-ish. Super cool.

  • Re: triggering the PTT, I'll start digging into this, and I'll also start soliciting help. The circuitry on the FV300 seems to me to be incredibly dense and tiny, at least compared to the open source circuit boards I've seen so far; I imagine that various parts of their design were optimized by separate teams of engineers over several iterations; figuring it out seems akin to reverse-engineering an evolved genetic network. But I share your intuition that signals to/from the PTT button and the speaker might be easy to intercept.
  • FRS vs GMRS: For our line-of-sight range test, we were using channel #1 -- the FV300 manual indicates that this channel is "GMRS/FRS", at a frequency of 462.5625 MHz; there are several other "GMRS/FRS" channels ranging up to 462.7125 MHz; and then there are "FRS" channels from 467.5625 to 467.7125 MHz, and "GMRS" channels from 462.5500 to 462.7250 MHz.
  • I was about to close this post with a question along the lines "what does it all mean?!" but then found a nice thread describing the differences / overlap in GMRS and FRS. Somewhat less confused now ...
Member

dwblair commented Aug 18, 2012

Hi Chris!

Oh gosh -- you led to some very interesting reading re: Quindar tones. I guess we are launching satellites, after all; and figuring out how to ping them for information is very NASA-ish. Super cool.

  • Re: triggering the PTT, I'll start digging into this, and I'll also start soliciting help. The circuitry on the FV300 seems to me to be incredibly dense and tiny, at least compared to the open source circuit boards I've seen so far; I imagine that various parts of their design were optimized by separate teams of engineers over several iterations; figuring it out seems akin to reverse-engineering an evolved genetic network. But I share your intuition that signals to/from the PTT button and the speaker might be easy to intercept.
  • FRS vs GMRS: For our line-of-sight range test, we were using channel #1 -- the FV300 manual indicates that this channel is "GMRS/FRS", at a frequency of 462.5625 MHz; there are several other "GMRS/FRS" channels ranging up to 462.7125 MHz; and then there are "FRS" channels from 467.5625 to 467.7125 MHz, and "GMRS" channels from 462.5500 to 462.7250 MHz.
  • I was about to close this post with a question along the lines "what does it all mean?!" but then found a nice thread describing the differences / overlap in GMRS and FRS. Somewhat less confused now ...
@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 19, 2012

Member

Good idea to solicit help. Everyone says you should get help.

And the relay part is tricky. Old fashioned relays are cheap and straightforward and you can even get a handy PCB kit. But solid state relays are also cheap and are smaller and require less control power. However, I think standard relays (old or SS) require constant power to keep the gate in one state (open or closed). When power switches off, the state changes. What we need is a latching relay which changes state whenever a short pulse of current is sent to the control side -- no current is needed to keep the gate in either state.

So we push the page button, current flows to the speaker in the remote handset, and that current causes the latching relay to either open or close the PTT circuit which remains in that state until another page is sent. But I don't know anything about solid state latching relays.

And there might be another problem with this approach. If we could use this method to remotely turn on the transmitter (like pushing PTT on the remote handset), will the remote receiver still be active? It needs to remain active so a subsequent paging signal can turn off PTT. So if you hold PTT down on one handset, what happens when you push page on the other handset?

Member

Fastie commented Aug 19, 2012

Good idea to solicit help. Everyone says you should get help.

And the relay part is tricky. Old fashioned relays are cheap and straightforward and you can even get a handy PCB kit. But solid state relays are also cheap and are smaller and require less control power. However, I think standard relays (old or SS) require constant power to keep the gate in one state (open or closed). When power switches off, the state changes. What we need is a latching relay which changes state whenever a short pulse of current is sent to the control side -- no current is needed to keep the gate in either state.

So we push the page button, current flows to the speaker in the remote handset, and that current causes the latching relay to either open or close the PTT circuit which remains in that state until another page is sent. But I don't know anything about solid state latching relays.

And there might be another problem with this approach. If we could use this method to remotely turn on the transmitter (like pushing PTT on the remote handset), will the remote receiver still be active? It needs to remain active so a subsequent paging signal can turn off PTT. So if you hold PTT down on one handset, what happens when you push page on the other handset?

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 19, 2012

Member

It looks like when PTT is depressed, the receiver on that handset is deactivated, or at least not sending anything to the speaker. That's the way walkie-talkies work, they can receive or (when PTT is pushed) they can transmit, but they can't do both at once. But hold the phone...

I just described this to Galen (age 13) who immediately said "Oh, so you just have to put a timer on it so the transmitter stays on for 20 seconds and then turns off." I said, "Wow, that's a really good idea," which was true.

So...

  1. You briefly push page on the ground-based handset
  2. the current intended to operate the remote speaker trips a switch (via a relay or MOSFET)
  3. which starts a timer circuit (e.g., based on a 555 IC)
  4. which closes a relay
  5. which closes the PTT circuit allowing transmission of audio of the shutter clicking
  6. until 20 seconds passes and the timer opens the relay which turns off the transmitter.
  7. But then the timer is still running...
  8. and the first relay is in the wrong state...
  9. So much for my electrical engineering skills.

So the circuits needs some fine tuning.

Or maybe dispense with the remote activation of the transmitter and let an onboard timer turn on the transmitter for 20 seconds every 5 minutes.

Member

Fastie commented Aug 19, 2012

It looks like when PTT is depressed, the receiver on that handset is deactivated, or at least not sending anything to the speaker. That's the way walkie-talkies work, they can receive or (when PTT is pushed) they can transmit, but they can't do both at once. But hold the phone...

I just described this to Galen (age 13) who immediately said "Oh, so you just have to put a timer on it so the transmitter stays on for 20 seconds and then turns off." I said, "Wow, that's a really good idea," which was true.

So...

  1. You briefly push page on the ground-based handset
  2. the current intended to operate the remote speaker trips a switch (via a relay or MOSFET)
  3. which starts a timer circuit (e.g., based on a 555 IC)
  4. which closes a relay
  5. which closes the PTT circuit allowing transmission of audio of the shutter clicking
  6. until 20 seconds passes and the timer opens the relay which turns off the transmitter.
  7. But then the timer is still running...
  8. and the first relay is in the wrong state...
  9. So much for my electrical engineering skills.

So the circuits needs some fine tuning.

Or maybe dispense with the remote activation of the transmitter and let an onboard timer turn on the transmitter for 20 seconds every 5 minutes.

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 19, 2012

Member
  • Clearly we can use Galen's help on the PVOS wiki -- great suggestions!
  • Re: relays and all that -- a friend who does a lot of electronics-y
    stuff suggested that an Arduino-based system is something to try first. In
    his opinion, the downside of moving to Arduino - that you have to buy an
    Arduino, learn how to program it, and store it on-board -- would likely be
    outweighed by the upside -- a lot of flexibility re: programmable control
    of any electronics that you can find a way to connect to it. An
    Arduino-based approach would also likely be extensible / hackable later,
    when moving onto the next step (live video, etc). In any case, I'm excited
    to dig into how relays work (in general, and how they might be useful in
    this particular context).
  • Speaking of live video ... you mentioned a friend who is looking
    into wireless Go Pro transmission. A few days I simply searched for "cheap
    alternative to Go Pro" online, and found this review of a tiny little $20
    key-fob camera that takes video and still photos, recording to SD storage.
    It looks like folks have been trying to hack it in order to transmit video
    wirelessly, too. Currently for sale here. I'll look
    and see how plausible the Arduino-control hacks seem. If we could get a
    microcontroller to talk to it, we should be able to transmit the images ...
Member

dwblair commented Aug 19, 2012

  • Clearly we can use Galen's help on the PVOS wiki -- great suggestions!
  • Re: relays and all that -- a friend who does a lot of electronics-y
    stuff suggested that an Arduino-based system is something to try first. In
    his opinion, the downside of moving to Arduino - that you have to buy an
    Arduino, learn how to program it, and store it on-board -- would likely be
    outweighed by the upside -- a lot of flexibility re: programmable control
    of any electronics that you can find a way to connect to it. An
    Arduino-based approach would also likely be extensible / hackable later,
    when moving onto the next step (live video, etc). In any case, I'm excited
    to dig into how relays work (in general, and how they might be useful in
    this particular context).
  • Speaking of live video ... you mentioned a friend who is looking
    into wireless Go Pro transmission. A few days I simply searched for "cheap
    alternative to Go Pro" online, and found this review of a tiny little $20
    key-fob camera that takes video and still photos, recording to SD storage.
    It looks like folks have been trying to hack it in order to transmit video
    wirelessly, too. Currently for sale here. I'll look
    and see how plausible the Arduino-control hacks seem. If we could get a
    microcontroller to talk to it, we should be able to transmit the images ...
@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 19, 2012

Member
Member

dwblair commented Aug 19, 2012

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 19, 2012

Member

Status of the Old-FART walkie-talkie exploration – is it possible to monitor remote camera operation by radio transmission of audio?

Pros:

  • A potentially simple solution for monitoring remote camera operation
  • Inexpensive
  • More than adequate range for all KAP work
  • Lightweight, small PCB
  • Can share a power supply widely used on KAP rigs (3 AAA)
  • FCC approved
  • Could be a widely adopted hack

Cons:

  • Continuous transmission of audio is unneeded and wasteful of power
  • FRS/GMRS radios are not intended for continuous transmission
  • Switching the transmitter on and off remotely may require: 1) a separate radio, or 2) a timer circuit to kill the remotely started transmitter after 20 seconds, or 3) serious hacking of the dense handset circuit
  • Switching the transmitter on and off via VOX requires a fancier model of handset

Current questions:

  • Continuing to investigate the potential to control the transmitter from the ground may be fruitful: Can the page feature turn on the transmitter, and can a timer turn it off 20 seconds later?
  • Is the simplest solution a 555 timer and relay that turn on the transmitter (push PTT) for 20 seconds every 5 minutes?
  • Investigating the potential to use VOX to transmit whenever the camera makes noise may be fruitful: Would it work with a more expensive handset?
Member

Fastie commented Aug 19, 2012

Status of the Old-FART walkie-talkie exploration – is it possible to monitor remote camera operation by radio transmission of audio?

Pros:

  • A potentially simple solution for monitoring remote camera operation
  • Inexpensive
  • More than adequate range for all KAP work
  • Lightweight, small PCB
  • Can share a power supply widely used on KAP rigs (3 AAA)
  • FCC approved
  • Could be a widely adopted hack

Cons:

  • Continuous transmission of audio is unneeded and wasteful of power
  • FRS/GMRS radios are not intended for continuous transmission
  • Switching the transmitter on and off remotely may require: 1) a separate radio, or 2) a timer circuit to kill the remotely started transmitter after 20 seconds, or 3) serious hacking of the dense handset circuit
  • Switching the transmitter on and off via VOX requires a fancier model of handset

Current questions:

  • Continuing to investigate the potential to control the transmitter from the ground may be fruitful: Can the page feature turn on the transmitter, and can a timer turn it off 20 seconds later?
  • Is the simplest solution a 555 timer and relay that turn on the transmitter (push PTT) for 20 seconds every 5 minutes?
  • Investigating the potential to use VOX to transmit whenever the camera makes noise may be fruitful: Would it work with a more expensive handset?
@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 20, 2012

Member

An Arduino is certainly up to this task. Would the approach be to add an onboard Arduino to the onboard walkie-talkie guts, or to replace the walkie-talkie with a ZigBee?

If ZigBee is part of the mix, the obvious ground unit would be a smart phone. This is clearly too much technology to throw at the task of making sure the camera has not stopped capturing (Camera Operation Detector), but there are many other tasks to which it could be applied. The same hardware required to solve the COD problem could also allow remote or autonomous control of pointing and shooting the camera, changing camera settings, and delivering data or images to the ground in real time.

This potential is alluring, but implementing just a few of these functions is a very big project -- maybe too big for PVOS? Some in the KAP community have had success with this, so collaborating with them to document and compile their implementations might be a good way to start and a useful contribution. There has been some haphazard discussion of using Arduinos for KAP on the Berkeley KAP forum, but maybe a dedicated repository for these experiments is needed.

I don’t yet have a good idea how much hardware would be required (Arduino, ZigBee, shields (?), connectors, cables, power, ground-based devices, etc), but I have a suspicion that it would be easy to spend $100 before a solution was near, not including the phone. Is that in the ballpark?

Member

Fastie commented Aug 20, 2012

An Arduino is certainly up to this task. Would the approach be to add an onboard Arduino to the onboard walkie-talkie guts, or to replace the walkie-talkie with a ZigBee?

If ZigBee is part of the mix, the obvious ground unit would be a smart phone. This is clearly too much technology to throw at the task of making sure the camera has not stopped capturing (Camera Operation Detector), but there are many other tasks to which it could be applied. The same hardware required to solve the COD problem could also allow remote or autonomous control of pointing and shooting the camera, changing camera settings, and delivering data or images to the ground in real time.

This potential is alluring, but implementing just a few of these functions is a very big project -- maybe too big for PVOS? Some in the KAP community have had success with this, so collaborating with them to document and compile their implementations might be a good way to start and a useful contribution. There has been some haphazard discussion of using Arduinos for KAP on the Berkeley KAP forum, but maybe a dedicated repository for these experiments is needed.

I don’t yet have a good idea how much hardware would be required (Arduino, ZigBee, shields (?), connectors, cables, power, ground-based devices, etc), but I have a suspicion that it would be easy to spend $100 before a solution was near, not including the phone. Is that in the ballpark?

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 20, 2012

Member

The 808 cameras are so small and lightweight that it seems silly NOT to have one on the rig every time a kite or balloon flies. But I am not sure what they add if there is already a nice camera with an A/V out port on the KAP rig. All Powershots can stream video and they all have a bigger sensor and better lens than an 808. So the video will be better and it will be exactly what the Powershot is seeing if you want to use it for framing still shots.

I have ordered components that should allow me to wirelessly stream the video and audio from a Powershot that is 1000 feet above ground. This is a $200 analog solution that is completely independent of any remote or autonomous control of the KAP rig. If an Arduino-based system can accomplish the video streaming, it would be the preferred solution because it could also replace the devices I currently use for remote or autonomous camera and rig control.

Member

Fastie commented Aug 20, 2012

The 808 cameras are so small and lightweight that it seems silly NOT to have one on the rig every time a kite or balloon flies. But I am not sure what they add if there is already a nice camera with an A/V out port on the KAP rig. All Powershots can stream video and they all have a bigger sensor and better lens than an 808. So the video will be better and it will be exactly what the Powershot is seeing if you want to use it for framing still shots.

I have ordered components that should allow me to wirelessly stream the video and audio from a Powershot that is 1000 feet above ground. This is a $200 analog solution that is completely independent of any remote or autonomous control of the KAP rig. If an Arduino-based system can accomplish the video streaming, it would be the preferred solution because it could also replace the devices I currently use for remote or autonomous camera and rig control.

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 20, 2012

Member

Hi Chris,

Great points -- and I really need to sit down and list the costs and complexity of the various Arduino-related solutions (that I can think of). I'll be able to do that a little later today; but briefly, I think you're right on your cost estimate -- might even be more like $160+. The typical Zigbee configuration seems to be to have a Zigbee module connected to an Arduino serving as a single node in the 2+ node communication network. An Arduino is going to be at least $20, a Zigbee module around $40, and then, as you pointed out, there will be bunch of connectors, cable, and power on top of that .... Matt was indicating in the workshop that it's possible to use a Zigbee itself, without the Arduino, as a simple node, as it has its own basic microprocessor; if this could work, then maybe the total cost could be under $100.

Replying to your earlier question: I was thinking of using the walkie-talkie as the radio, and an on-board Arduino to handle generating signals. The Arduino could monitor battery consumption and trigger the "paging" button on the walkie-talkie (and could also replace the 555-timer circuit in order to trigger the camera, as you suggested). The Arduino model I'm thinking of is around $20, could be powered by the 3 AAAs (I figure), and so the total cost of the solution -- Arduino + walkie-talkies -- would be around $40.

Up til now, I've been a little shy re: opening up the FV300 and poking around -- I've been afraid to break it. But it looks like we've explored (so far as I'm able on my end, anyway) the main options re: using the FV300 without fiddling with its internals ... so it's time to dig in.

I'm going to look into this more later today, but I found a promising link last night re: commandeering walkie talkie circuitry. The full write-up (via captions to flickr photos) is here. Let me know what you think?

Great to hear re: the $200 analog solution! Looking forward to hearing about / seeing the setup and results ...

More in a bit ...

Member

dwblair commented Aug 20, 2012

Hi Chris,

Great points -- and I really need to sit down and list the costs and complexity of the various Arduino-related solutions (that I can think of). I'll be able to do that a little later today; but briefly, I think you're right on your cost estimate -- might even be more like $160+. The typical Zigbee configuration seems to be to have a Zigbee module connected to an Arduino serving as a single node in the 2+ node communication network. An Arduino is going to be at least $20, a Zigbee module around $40, and then, as you pointed out, there will be bunch of connectors, cable, and power on top of that .... Matt was indicating in the workshop that it's possible to use a Zigbee itself, without the Arduino, as a simple node, as it has its own basic microprocessor; if this could work, then maybe the total cost could be under $100.

Replying to your earlier question: I was thinking of using the walkie-talkie as the radio, and an on-board Arduino to handle generating signals. The Arduino could monitor battery consumption and trigger the "paging" button on the walkie-talkie (and could also replace the 555-timer circuit in order to trigger the camera, as you suggested). The Arduino model I'm thinking of is around $20, could be powered by the 3 AAAs (I figure), and so the total cost of the solution -- Arduino + walkie-talkies -- would be around $40.

Up til now, I've been a little shy re: opening up the FV300 and poking around -- I've been afraid to break it. But it looks like we've explored (so far as I'm able on my end, anyway) the main options re: using the FV300 without fiddling with its internals ... so it's time to dig in.

I'm going to look into this more later today, but I found a promising link last night re: commandeering walkie talkie circuitry. The full write-up (via captions to flickr photos) is here. Let me know what you think?

Great to hear re: the $200 analog solution! Looking forward to hearing about / seeing the setup and results ...

More in a bit ...

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 20, 2012

Member

Is this a fair description of a possible way to apply an Arduino to the Old-FART approach?

  • A ground-based walkie-talkie sends a signal to the remote handset
  • This sends a current through the easily accessible wire to the speaker (the strength of this current depends on the volume setting on the remote handset, and whether PTT or page was used to send the signal)
  • An Arduino is somehow alerted by this pulse of current -- Is extra non-Arduino hardware needed to do this (to filter the current or protect the Arduino)?
  • The Arduino responds by sending a signal to a relay (or MOSFET or some switchy thing) that turns on the remote transmitter (this closes the PTT button circuit) -- Is extra non-Arduino hardware needed to do this (are there Arduino switches?)?
  • The ground-based handset can now hear the remote camera clicking
  • The Arduino sends a signal to the switch to turn off the transmitter (this has to happen within 30 seconds or the TOT will intervene)

In the above case the Arduino may be doing little more than counting to 20 and sending a pulse at the start and finish of each counting episode. A 555 timer circuit could do the same thing and include the relays, etc on the same PCB which would need a single power source.

If an Arduino was tasked with doing this on-demand 20 second timer job, could it also be sending a pulse every 8 seconds to release the camera shutter?

Member

Fastie commented Aug 20, 2012

Is this a fair description of a possible way to apply an Arduino to the Old-FART approach?

  • A ground-based walkie-talkie sends a signal to the remote handset
  • This sends a current through the easily accessible wire to the speaker (the strength of this current depends on the volume setting on the remote handset, and whether PTT or page was used to send the signal)
  • An Arduino is somehow alerted by this pulse of current -- Is extra non-Arduino hardware needed to do this (to filter the current or protect the Arduino)?
  • The Arduino responds by sending a signal to a relay (or MOSFET or some switchy thing) that turns on the remote transmitter (this closes the PTT button circuit) -- Is extra non-Arduino hardware needed to do this (are there Arduino switches?)?
  • The ground-based handset can now hear the remote camera clicking
  • The Arduino sends a signal to the switch to turn off the transmitter (this has to happen within 30 seconds or the TOT will intervene)

In the above case the Arduino may be doing little more than counting to 20 and sending a pulse at the start and finish of each counting episode. A 555 timer circuit could do the same thing and include the relays, etc on the same PCB which would need a single power source.

If an Arduino was tasked with doing this on-demand 20 second timer job, could it also be sending a pulse every 8 seconds to release the camera shutter?

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 21, 2012

Member

This seems like a reasonable scheme to me!

Quick question (sorry if the answer is obvious based on prior discussion :)): the rationale for the initial ground-based signal to initiate the eavesdropping process is that it conserves battery -- e.g., one could check in only at whatever intervals one wanted? That's neat!

Aside: the concept I had in re: the Arduino setup was to do with monitoring the camera battery level (as per an earlier suggestion of yours). It also would allow for checking in at intervals, but only with pre-set intervals; with the (possible) advantage that it could eliminate some complexity:

  1. Arduino on board the rig is controlling the release of the camera shutter -- e.g., every 8 seconds. This is simply replacing the functionality of the 555 timer.
  2. The same Arduino is monitoring the camera battery level (we found some circuit for doing this at some point earlier, can dig it up.) After every shutter release, the Arduino compares the current battery level to the battery level before the release. If it has decreased by a predetermined threshold (determined experimentally beforehand by measuring the battery drop after each camera release, and written as a threshold into the code), the Arduino considers that this camera shutter release was a "success". It then updates a variable -- cameraPulseSuccess -- to "TRUE"; if the battery level doesn't drop by the proper amount, cameraPulseSuccess=FALSE.
  3. At a pre-set interval (e.g., every 60 seconds? every 5 minutes?), the Arduino triggers a quick pulse to the "page" or "PTT" button on the walkie talkie, which sends a pulse to the ground. If cameraPulseSuccess==TRUE, send two quick bleeps in this manner; if cameraPulseSuccess==FALSE, send one quick bleep.

If it's more desirable to listen in on the camera rig than to monitor battery level, step 2. could involve turning on the PTT button for some time (less than the allowed 30 seconds), then turn it off, more in line with your scheme above; but this method seems more prone to being overwhelmed by wind noise & requiring some fine-tuning in-the-field.

Was this in line with what you were thinking earlier re: monitoring battery level? Or did we reject this idea in favor of the audio eavesdropping of camera clicks for a reason I'm not remembering? (This is why your summaries re: status quo are super-useful ... maybe we should even copy them out of this comment stream, and add them as "interim reports" in the Old-FART wiki page ...)

Member

dwblair commented Aug 21, 2012

This seems like a reasonable scheme to me!

Quick question (sorry if the answer is obvious based on prior discussion :)): the rationale for the initial ground-based signal to initiate the eavesdropping process is that it conserves battery -- e.g., one could check in only at whatever intervals one wanted? That's neat!

Aside: the concept I had in re: the Arduino setup was to do with monitoring the camera battery level (as per an earlier suggestion of yours). It also would allow for checking in at intervals, but only with pre-set intervals; with the (possible) advantage that it could eliminate some complexity:

  1. Arduino on board the rig is controlling the release of the camera shutter -- e.g., every 8 seconds. This is simply replacing the functionality of the 555 timer.
  2. The same Arduino is monitoring the camera battery level (we found some circuit for doing this at some point earlier, can dig it up.) After every shutter release, the Arduino compares the current battery level to the battery level before the release. If it has decreased by a predetermined threshold (determined experimentally beforehand by measuring the battery drop after each camera release, and written as a threshold into the code), the Arduino considers that this camera shutter release was a "success". It then updates a variable -- cameraPulseSuccess -- to "TRUE"; if the battery level doesn't drop by the proper amount, cameraPulseSuccess=FALSE.
  3. At a pre-set interval (e.g., every 60 seconds? every 5 minutes?), the Arduino triggers a quick pulse to the "page" or "PTT" button on the walkie talkie, which sends a pulse to the ground. If cameraPulseSuccess==TRUE, send two quick bleeps in this manner; if cameraPulseSuccess==FALSE, send one quick bleep.

If it's more desirable to listen in on the camera rig than to monitor battery level, step 2. could involve turning on the PTT button for some time (less than the allowed 30 seconds), then turn it off, more in line with your scheme above; but this method seems more prone to being overwhelmed by wind noise & requiring some fine-tuning in-the-field.

Was this in line with what you were thinking earlier re: monitoring battery level? Or did we reject this idea in favor of the audio eavesdropping of camera clicks for a reason I'm not remembering? (This is why your summaries re: status quo are super-useful ... maybe we should even copy them out of this comment stream, and add them as "interim reports" in the Old-FART wiki page ...)

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 21, 2012

Member

Yes, the rationale for my last scheme is:

    1. to check in manually whenever you wanted to confirm that the camera was still operating. This conserves the remote battery by keeping the transmitter off most of the time, and anyway the PTT handsets will not keep the transmitter on continuously.
    1. To monitor the sound of the camera clicking because that seemed to be the simplest way to ensure that the camera is still operating. The cameras make a very distinct noise, and putting a cheap mic right against the tiny camera speaker (or just a tube between the speaker and the handset mic) will eliminate any wind problem. This would probably work fine with VOX which might be the simplest solution of all because the handset already knows how to turn on the transmitter and turn it off when the noise stops.

Monitoring the battery level after each shutter release as you describe is very clever and might work, but it needs to be tested. The battery level could be affected by temperature (ambient air temperature changes, sun heats up camera, power consumption heats up camera), or by shot-to-shot variation in how hard the camera is working to focus or process (jpg compression) and save each image. Also, how much drain on the batteries will be caused by monitoring the levels every 10 seconds? Is this approach self defeating?

Sending signals to the ground could be done either 1) on demand, 2) at intervals, or 3) when the Arduino detects a problem. Any of these would work. On demand might be the most effective because the user might fail to notice an unexpected message, but this might be the hardest to implement.

If an Arduino is connected to the USB port of a Powershot in order to send pulses to release the shutter (either autonomously or by remote radio control), then the Arduino has the potential to receive real time data from the camera. A script running under CHDK can access many values (CCD temperature, camera body temperature, battery level, number of shots taken, whether the camera is currently shooting, etc) and presumably send messages to the USB port using PTP. Arduinos apparently know all about PTP. With this type of data, the Arduino could easily determine whether camera operation was normal, and could then send a signal to the ground. Different signals could code different messages (camera not shooting, battery level at 1/4, SD card will fill with 20 more shots). Now THAT would be exploiting the little ATmega168! (and also the camera's own sophisticated circuitry)

Member

Fastie commented Aug 21, 2012

Yes, the rationale for my last scheme is:

    1. to check in manually whenever you wanted to confirm that the camera was still operating. This conserves the remote battery by keeping the transmitter off most of the time, and anyway the PTT handsets will not keep the transmitter on continuously.
    1. To monitor the sound of the camera clicking because that seemed to be the simplest way to ensure that the camera is still operating. The cameras make a very distinct noise, and putting a cheap mic right against the tiny camera speaker (or just a tube between the speaker and the handset mic) will eliminate any wind problem. This would probably work fine with VOX which might be the simplest solution of all because the handset already knows how to turn on the transmitter and turn it off when the noise stops.

Monitoring the battery level after each shutter release as you describe is very clever and might work, but it needs to be tested. The battery level could be affected by temperature (ambient air temperature changes, sun heats up camera, power consumption heats up camera), or by shot-to-shot variation in how hard the camera is working to focus or process (jpg compression) and save each image. Also, how much drain on the batteries will be caused by monitoring the levels every 10 seconds? Is this approach self defeating?

Sending signals to the ground could be done either 1) on demand, 2) at intervals, or 3) when the Arduino detects a problem. Any of these would work. On demand might be the most effective because the user might fail to notice an unexpected message, but this might be the hardest to implement.

If an Arduino is connected to the USB port of a Powershot in order to send pulses to release the shutter (either autonomously or by remote radio control), then the Arduino has the potential to receive real time data from the camera. A script running under CHDK can access many values (CCD temperature, camera body temperature, battery level, number of shots taken, whether the camera is currently shooting, etc) and presumably send messages to the USB port using PTP. Arduinos apparently know all about PTP. With this type of data, the Arduino could easily determine whether camera operation was normal, and could then send a signal to the ground. Different signals could code different messages (camera not shooting, battery level at 1/4, SD card will fill with 20 more shots). Now THAT would be exploiting the little ATmega168! (and also the camera's own sophisticated circuitry)

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 21, 2012

Member

Hi Chris,

Great stuff. And, I still do like your VOX idea -- it seems like the simplest, and likely cheapest, solution. Too bad the FV300 (so far as I can tell) can't seem to implement it.

TODO find the least expensive walkie talkie with VOX capability out there.

As far as the other options go, the simplest and most accessible entry point (for me, anyway, given my meager electronics background) seems to be getting an Arduino to trigger the pager circuit on the walkie talkie at intervals, based on some measurement (CHDK / PTP). We know that folks have gotten an Arduino to control / monitor Powershots via CHDK, so it's just a matter of finding and following good tutorials. The only real hurdle for this approach (again, given my background -- wouldn't be a hurdle for a true supergeek hacker -- btw: hello serious hacker supergeeks out there! a little help?), is getting an Arduino to trigger the walkie talkie pager circuit.

Last night, I did a lot of searching online, and it seems that one of the easier methods for controlling an external circuit with an arduino is to use an optoisolator-- which is a really cool little $0.20 device I'd no idea existed. Basically, it would allow an Arduino to be connected to an external device that has a different voltage / current / whatnot and not worry about blowing up either device. There is a fantastic-looking, easy-to-follow-seeming tutorial on OpenMoCo that goes through setting one of these up. I'm not sure, but I think it'll be directly applicable to controlling the paging circuit on the walkie talkie.

The pager button is one of four metallic-looking discs atop a white background and underneath the lcd:

The PTT button is a similar style button located on the side of the circuit board:

At first I was a bit flummoxed by how these buttons work -- but it turns out that they are simply tin-like metal discs that buckle when you press; buckling makes their center bend down and touch an exposed bit of a particular circuit on the circuitboard immediately below, and the outer ring is also connected to the same circuit -- thus connecting that circuit. I figured this (not-so-earth-shattering-insight) out by ruining a $4 weight scale bought off amazon:

I think this button -- the cover of which is visible on the table next to the circuit board, and which originally covered the part of the circuit labeled S3 -- I was able to pull it off the circuit board with a gentle pliers-tug -- is the same design as that on the walkie - talkie.

So, my game plan for the next week is:

  • Order a few optoisolators from Newark (darn, the price has risen to $0.30 ea)
  • Follow the OpenMoCo tutorial on controlling an LED via an Arduino using an optoisolator
  • Use a similar circuit with minimal modification to try to tweak the walkie talkie pager button. Should be a matter of simply soldering wires to the inner and outer part of the button, if the walkie talkie button looks like what's in the last picture above.

Then we'll have Arduino-controlled transmission of a signal! (In case that might be useful. In any case, it'll be fun.)

Meanwhile, I'll also try to get my hands on a CHDK-compatible Powershot (thanks for your tips on doing this, above) for the next step -- controlling a camera and pinging a walkie talkie ...

Wheee!

Member

dwblair commented Aug 21, 2012

Hi Chris,

Great stuff. And, I still do like your VOX idea -- it seems like the simplest, and likely cheapest, solution. Too bad the FV300 (so far as I can tell) can't seem to implement it.

TODO find the least expensive walkie talkie with VOX capability out there.

As far as the other options go, the simplest and most accessible entry point (for me, anyway, given my meager electronics background) seems to be getting an Arduino to trigger the pager circuit on the walkie talkie at intervals, based on some measurement (CHDK / PTP). We know that folks have gotten an Arduino to control / monitor Powershots via CHDK, so it's just a matter of finding and following good tutorials. The only real hurdle for this approach (again, given my background -- wouldn't be a hurdle for a true supergeek hacker -- btw: hello serious hacker supergeeks out there! a little help?), is getting an Arduino to trigger the walkie talkie pager circuit.

Last night, I did a lot of searching online, and it seems that one of the easier methods for controlling an external circuit with an arduino is to use an optoisolator-- which is a really cool little $0.20 device I'd no idea existed. Basically, it would allow an Arduino to be connected to an external device that has a different voltage / current / whatnot and not worry about blowing up either device. There is a fantastic-looking, easy-to-follow-seeming tutorial on OpenMoCo that goes through setting one of these up. I'm not sure, but I think it'll be directly applicable to controlling the paging circuit on the walkie talkie.

The pager button is one of four metallic-looking discs atop a white background and underneath the lcd:

The PTT button is a similar style button located on the side of the circuit board:

At first I was a bit flummoxed by how these buttons work -- but it turns out that they are simply tin-like metal discs that buckle when you press; buckling makes their center bend down and touch an exposed bit of a particular circuit on the circuitboard immediately below, and the outer ring is also connected to the same circuit -- thus connecting that circuit. I figured this (not-so-earth-shattering-insight) out by ruining a $4 weight scale bought off amazon:

I think this button -- the cover of which is visible on the table next to the circuit board, and which originally covered the part of the circuit labeled S3 -- I was able to pull it off the circuit board with a gentle pliers-tug -- is the same design as that on the walkie - talkie.

So, my game plan for the next week is:

  • Order a few optoisolators from Newark (darn, the price has risen to $0.30 ea)
  • Follow the OpenMoCo tutorial on controlling an LED via an Arduino using an optoisolator
  • Use a similar circuit with minimal modification to try to tweak the walkie talkie pager button. Should be a matter of simply soldering wires to the inner and outer part of the button, if the walkie talkie button looks like what's in the last picture above.

Then we'll have Arduino-controlled transmission of a signal! (In case that might be useful. In any case, it'll be fun.)

Meanwhile, I'll also try to get my hands on a CHDK-compatible Powershot (thanks for your tips on doing this, above) for the next step -- controlling a camera and pinging a walkie talkie ...

Wheee!

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 21, 2012

Member

That's a very nice tutorial on optoisolators. Is the final result on the breadboard basically a solid state relay? An SSR is generally built around an optoisolator, so maybe just starting with an SSR would work. Just apply a current to one side of the SSR and the other side closes a circuit (pushes a button). Stop the control current and the circuit opens (releasing the button).

Too bad about your fish scale.

Here is a Powershot SD 1100 on ebay that you might be able to win tomorrow for $40. A good camera for testing and flying. This one has battery, charger, SD card.

Member

Fastie commented Aug 21, 2012

That's a very nice tutorial on optoisolators. Is the final result on the breadboard basically a solid state relay? An SSR is generally built around an optoisolator, so maybe just starting with an SSR would work. Just apply a current to one side of the SSR and the other side closes a circuit (pushes a button). Stop the control current and the circuit opens (releasing the button).

Too bad about your fish scale.

Here is a Powershot SD 1100 on ebay that you might be able to win tomorrow for $40. A good camera for testing and flying. This one has battery, charger, SD card.

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 22, 2012

Member

+1 on the SSR: Rui (my local electronics / Arduino guru) just gave me a handful of MOSFETs (I believe they're approx $.01 ea.), and echoed your suggestion. I guess an optoisolator would be more appropriate when one is worried about blowing up sensitive electronics via a voltage mismatch (as in the case of connecting an Arduino up to a digital SLR as per the optoisolator tutorial above); but the walkie talkie circuit's voltage is close enough to the Arduino's that I guess this isn't much of a worry. From the brief tutorial Rui gave me this afternoon -- in line with your suggestion re: SSRs above -- we should be off to the races very soon.

Guide to using MOSFETs ...

Cheers re: the SD 1100 -- I'll see if I can jump on that presently!

Member

dwblair commented Aug 22, 2012

+1 on the SSR: Rui (my local electronics / Arduino guru) just gave me a handful of MOSFETs (I believe they're approx $.01 ea.), and echoed your suggestion. I guess an optoisolator would be more appropriate when one is worried about blowing up sensitive electronics via a voltage mismatch (as in the case of connecting an Arduino up to a digital SLR as per the optoisolator tutorial above); but the walkie talkie circuit's voltage is close enough to the Arduino's that I guess this isn't much of a worry. From the brief tutorial Rui gave me this afternoon -- in line with your suggestion re: SSRs above -- we should be off to the races very soon.

Guide to using MOSFETs ...

Cheers re: the SD 1100 -- I'll see if I can jump on that presently!

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 23, 2012

Member

Hi Chris,

MOSFET: check! I've got an Arduino telling one FV300 to page the other every 6 seconds (and to light up an LED, for good measure). Video here. More details to follow -- for now, I need to put in an order of electronics by 9PM EST ...

SUCCESS (for now) ...

Member

dwblair commented Aug 23, 2012

Hi Chris,

MOSFET: check! I've got an Arduino telling one FV300 to page the other every 6 seconds (and to light up an LED, for good measure). Video here. More details to follow -- for now, I need to put in an order of electronics by 9PM EST ...

SUCCESS (for now) ...

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 23, 2012

Member

I have never been more excited to see such a crappy video (with an excellent name)! That is terrific progress. Nice surgery on the handset. Hooray for MOSFETs.

I'm afraid that SD1100 on ebay is already up to $48 + $11 shipping. We need to get you a camera right away. Should I mail you one?

Member

Fastie commented Aug 23, 2012

I have never been more excited to see such a crappy video (with an excellent name)! That is terrific progress. Nice surgery on the handset. Hooray for MOSFETs.

I'm afraid that SD1100 on ebay is already up to $48 + $11 shipping. We need to get you a camera right away. Should I mail you one?

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 23, 2012

Member

Oh wow, thanks -- but I'm currently the highest bidder on the SD1100! Eeek -- how high shall I go? I'm new to this bidding procedure ...

Member

dwblair commented Aug 23, 2012

Oh wow, thanks -- but I'm currently the highest bidder on the SD1100! Eeek -- how high shall I go? I'm new to this bidding procedure ...

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 23, 2012

Member

Oh wow, thanks! But I'm currently the highest bidder on that camera, so I
might have already purchased it :) -- bated breath ...

BTW it should be possible to turn on the PTT button via the Arduino just as
easily and "hold it down" ...

On Wed, Aug 22, 2012 at 8:37 PM, Fastie notifications@github.com wrote:

I have never been more excited to see such a crappy video (with an
excellent name)! That is terrific progress. Nice surgery on the handset.
Hooray for MOSFETs.

I'm afraid that SD1100 on ebay is already up to $48 + $11 shipping. We
need to get you a camera right away. Should I mail you one?


Reply to this email directly or view it on GitHubhttps://github.com//issues/2#issuecomment-7956770.

voice / SMS: +1-651-252-4765
skype: dwingateb

Member

dwblair commented Aug 23, 2012

Oh wow, thanks! But I'm currently the highest bidder on that camera, so I
might have already purchased it :) -- bated breath ...

BTW it should be possible to turn on the PTT button via the Arduino just as
easily and "hold it down" ...

On Wed, Aug 22, 2012 at 8:37 PM, Fastie notifications@github.com wrote:

I have never been more excited to see such a crappy video (with an
excellent name)! That is terrific progress. Nice surgery on the handset.
Hooray for MOSFETs.

I'm afraid that SD1100 on ebay is already up to $48 + $11 shipping. We
need to get you a camera right away. Should I mail you one?


Reply to this email directly or view it on GitHubhttps://github.com//issues/2#issuecomment-7956770.

voice / SMS: +1-651-252-4765
skype: dwingateb

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 23, 2012

Member

$59 is a good deal but not a bargain. I got two of those 18 months ago for $50 and $60 (incl shipping). You can get a refurbished A495 for $70 plus shipping which would be a better deal. But it's nice to experiment and learn to fly with only a $60 dollar investment. Somebody will probably outbid you at the last moment anyway (I never place any bids until the last moment).

Member

Fastie commented Aug 23, 2012

$59 is a good deal but not a bargain. I got two of those 18 months ago for $50 and $60 (incl shipping). You can get a refurbished A495 for $70 plus shipping which would be a better deal. But it's nice to experiment and learn to fly with only a $60 dollar investment. Somebody will probably outbid you at the last moment anyway (I never place any bids until the last moment).

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 23, 2012

Member

A495 for $51 "Buy it now"

Member

Fastie commented Aug 23, 2012

A495 for $51 "Buy it now"

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 23, 2012

Member

Hah ... okay, in all the excitement (and because there was a Canon A2200 at Radio Shack for $80 this afternoon -- which seemed from my 30 second Google search to be an inferior camera), I had already put in a maximum bid of $70 -- so we can all watch what happens ...

Re: Arduino-controlled FV300 pager: I'll be posting pics of the assembly process tomorrow if I can, but here's a quick summary:

  • I'm not sure yet, but it seems as though the circuitry doesn't like receiving repeated "pages" spaced less than around 4 or 5 seconds apart. Background: I started to goof around with short clicks (as a baby step towards sending different "codes", where each code might e.g. correspond to a different camera status), but the circuit seems to want to recover for a few seconds before it's ready for the next page.
  • The assembly process is pretty straightforward: just remove a few screws, and solder two wires to the front of the circuit board.
  • What's next? The two main options from our prior discussion seem to be a) getting the Arduino to listen to the camera via CHDK / PTT, and then ping the pager circuit as a function of the camera status; b) having the Arduino monitor the voltage across the speaker; when you want to listen in on the rig, you press page; the Arduino hears this noise on the speaker, and this triggers the Arduino to turn the PTT button "on" for 30 seconds in order to listen in on the rig.

I'm interested in doing a) in any case, just to learn how to control / listen to a camera with an Arduino. On the other hand, I'm not yet quite sure how to do b), but the MOSFET trick ended up being quite straightforward (with your guidance), so I'm optimistic that listening in on the speaker won't be too hard to accomplish, either ...

Member

dwblair commented Aug 23, 2012

Hah ... okay, in all the excitement (and because there was a Canon A2200 at Radio Shack for $80 this afternoon -- which seemed from my 30 second Google search to be an inferior camera), I had already put in a maximum bid of $70 -- so we can all watch what happens ...

Re: Arduino-controlled FV300 pager: I'll be posting pics of the assembly process tomorrow if I can, but here's a quick summary:

  • I'm not sure yet, but it seems as though the circuitry doesn't like receiving repeated "pages" spaced less than around 4 or 5 seconds apart. Background: I started to goof around with short clicks (as a baby step towards sending different "codes", where each code might e.g. correspond to a different camera status), but the circuit seems to want to recover for a few seconds before it's ready for the next page.
  • The assembly process is pretty straightforward: just remove a few screws, and solder two wires to the front of the circuit board.
  • What's next? The two main options from our prior discussion seem to be a) getting the Arduino to listen to the camera via CHDK / PTT, and then ping the pager circuit as a function of the camera status; b) having the Arduino monitor the voltage across the speaker; when you want to listen in on the rig, you press page; the Arduino hears this noise on the speaker, and this triggers the Arduino to turn the PTT button "on" for 30 seconds in order to listen in on the rig.

I'm interested in doing a) in any case, just to learn how to control / listen to a camera with an Arduino. On the other hand, I'm not yet quite sure how to do b), but the MOSFET trick ended up being quite straightforward (with your guidance), so I'm optimistic that listening in on the speaker won't be too hard to accomplish, either ...

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 23, 2012

Member

Ah, just saw your "buy it now" link -- thanks! Well, if I lose this bidding war, then I'm I'll definitely simply buy that A495!

Member

dwblair commented Aug 23, 2012

Ah, just saw your "buy it now" link -- thanks! Well, if I lose this bidding war, then I'm I'll definitely simply buy that A495!

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 23, 2012

Member

It looks like you have mastered all the skills you need for b). Essentially replace the speaker with a MOSFET and have current on the speaker circuit close a circuit which... Oh yeah, how do you alert the Arduino that the MOSFET just detected current on the speaker circuit? Hmmm

But that should be easier to figure out than a). I agree that learning how to communicate with a Powershot would be a great accomplishment. Especially if two-way communication is possible. Then the Arduino can tell the camera when to shoot, and also learn how many shots it has taken, etc., and generally keep track of camera operation. When the Arduino decides that something unusual has happened, you already know how to have it phone home! (Which is very cool.)

Sorry I didn't notice that "buy it now" A495 yesterday.

Member

Fastie commented Aug 23, 2012

It looks like you have mastered all the skills you need for b). Essentially replace the speaker with a MOSFET and have current on the speaker circuit close a circuit which... Oh yeah, how do you alert the Arduino that the MOSFET just detected current on the speaker circuit? Hmmm

But that should be easier to figure out than a). I agree that learning how to communicate with a Powershot would be a great accomplishment. Especially if two-way communication is possible. Then the Arduino can tell the camera when to shoot, and also learn how many shots it has taken, etc., and generally keep track of camera operation. When the Arduino decides that something unusual has happened, you already know how to have it phone home! (Which is very cool.)

Sorry I didn't notice that "buy it now" A495 yesterday.

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 23, 2012

Member

Congrats on the SD 1100!

Member

Fastie commented Aug 23, 2012

Congrats on the SD 1100!

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 23, 2012

Member

Wheee -- an SD 1100 coming right up -- thanks for the pointer to it! Will be taking photos of my house from above via hydrogen balloon in no time ...

Re: option b), yeah -- I guess I just implement the "reverse" of what I've done already? The speaker circuit turns a MOSFET "on", which completes a circuit on the Arduino, and the Arduino senses that ... I'll definitely look into it tomorrow.

I'm going to take a walk outside now and see how far I can walk before my FV300 can no longer hear the paging signal. Not exactly line-of-sight, but a nice coda to the flurry of activity. Will post pictures of the process tomorrow, hopefully. Thanks AGAIN for all the encouragement and guidance -- I wouldn't have attempted such intimidating-for-me electronics tomfoolery without it ...

Member

dwblair commented Aug 23, 2012

Wheee -- an SD 1100 coming right up -- thanks for the pointer to it! Will be taking photos of my house from above via hydrogen balloon in no time ...

Re: option b), yeah -- I guess I just implement the "reverse" of what I've done already? The speaker circuit turns a MOSFET "on", which completes a circuit on the Arduino, and the Arduino senses that ... I'll definitely look into it tomorrow.

I'm going to take a walk outside now and see how far I can walk before my FV300 can no longer hear the paging signal. Not exactly line-of-sight, but a nice coda to the flurry of activity. Will post pictures of the process tomorrow, hopefully. Thanks AGAIN for all the encouragement and guidance -- I wouldn't have attempted such intimidating-for-me electronics tomfoolery without it ...

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 23, 2012

Member

Update -- was able to hear the pinging 5,300 feet away, non line-of-sight. BOOM.

Member

dwblair commented Aug 23, 2012

Update -- was able to hear the pinging 5,300 feet away, non line-of-sight. BOOM.

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 23, 2012

Member

Wow, welcome to the Mile Away Club. GMRS or FRS? I hope no one in your neighborhood has an FRS home intercom system on the same channel;^)

Member

Fastie commented Aug 23, 2012

Wow, welcome to the Mile Away Club. GMRS or FRS? I hope no one in your neighborhood has an FRS home intercom system on the same channel;^)

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 23, 2012

Member

Thanks very much! I do have to say that it was quite cool to be walking around Amherst late at night, listening to a ping being sent by an Arduino at home. I ended up in a pretty dark part of town quite late at night, and the pinging sound was somehow comforting. I imagine this is how Voyager II might be feeling (I do hope we're still pinging Voyager II?) ...

Yes, I was a bit worried that someone nearby was getting quite annoyed by the pinging. I did manage to get the paging sound down to a brief "click / abbreviated beep" noise by reducing the amount of time the paging button was energized. Less time, and all one would hear would be a brief "click" noise, akin to when someone presses the PTT button momentarily without saying anything. If this ends up being used for a message sending, might be worth figuring out the least annoying sound.

Also: occurred to me that we could use a simple piezo to replace the speaker, if we wanted to generate a little vibrating plate instead of / in addition to an audible noise. That way one could e.g. wear the piezo against one's wrist, so that one wouldn't have to worry about holding the walkie talkie close enough to hear. Although I guess one could simply plug in an earpiece to the FV300, too. Anyway, piezo (or a blinking LED) would be other options ...

Re; GMRS vs FRS: this was channel #1 on the FV300, which I believe in the manual was listed as "GMRS / FRS" ... will check.

Member

dwblair commented Aug 23, 2012

Thanks very much! I do have to say that it was quite cool to be walking around Amherst late at night, listening to a ping being sent by an Arduino at home. I ended up in a pretty dark part of town quite late at night, and the pinging sound was somehow comforting. I imagine this is how Voyager II might be feeling (I do hope we're still pinging Voyager II?) ...

Yes, I was a bit worried that someone nearby was getting quite annoyed by the pinging. I did manage to get the paging sound down to a brief "click / abbreviated beep" noise by reducing the amount of time the paging button was energized. Less time, and all one would hear would be a brief "click" noise, akin to when someone presses the PTT button momentarily without saying anything. If this ends up being used for a message sending, might be worth figuring out the least annoying sound.

Also: occurred to me that we could use a simple piezo to replace the speaker, if we wanted to generate a little vibrating plate instead of / in addition to an audible noise. That way one could e.g. wear the piezo against one's wrist, so that one wouldn't have to worry about holding the walkie talkie close enough to hear. Although I guess one could simply plug in an earpiece to the FV300, too. Anyway, piezo (or a blinking LED) would be other options ...

Re; GMRS vs FRS: this was channel #1 on the FV300, which I believe in the manual was listed as "GMRS / FRS" ... will check.

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 26, 2012

Member

Don, Here is the PLOTS guide for installing CHDK on your new SD1100. And here are some tips for using a Powershot for aerial photography. These guides emphasize the Canon A495, but most things apply well to the SD1100.

Member

Fastie commented Aug 26, 2012

Don, Here is the PLOTS guide for installing CHDK on your new SD1100. And here are some tips for using a Powershot for aerial photography. These guides emphasize the Canon A495, but most things apply well to the SD1100.

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 31, 2012

Member

Hi Chris -- thanks for the link re: CHDK. I'll be looking to install this very soon!

And just now, I realized how brilliant Galen's timer suggestion is, and what a simple and straightforward solution it could be for listening in on the camera as per the Old FART approach you've been outlining above. You may have already suggested this particular solution (if so, then it just seeped in to consciousness, apologies) but it suddenly clicked for me (so to speak) that -- now that we've got the Arduino "pushing buttons" on the walkie talkie -- we could simply have the on-board Arduino trigger the PTT of the on-board walkie talkie. I.e., the on-board Arduino will:

  1. Turn the walkie talkie on
  2. Press and hold the PTT button
  3. Release the PTT button before the 30 sec max interval is up
  4. Turn the walkie talkie off
  5. Wait X minutes
  6. Repeat

Depending on the length of the flight, we could reduce the complexity further by only doing step (1) once, and omitting step 4.

It won't be as nice as triggering the on-board PTT button via a ground-based signal, but it might be a little more robust (maybe?) in that we only ever have to successfully broadcast one way.

Member

dwblair commented Aug 31, 2012

Hi Chris -- thanks for the link re: CHDK. I'll be looking to install this very soon!

And just now, I realized how brilliant Galen's timer suggestion is, and what a simple and straightforward solution it could be for listening in on the camera as per the Old FART approach you've been outlining above. You may have already suggested this particular solution (if so, then it just seeped in to consciousness, apologies) but it suddenly clicked for me (so to speak) that -- now that we've got the Arduino "pushing buttons" on the walkie talkie -- we could simply have the on-board Arduino trigger the PTT of the on-board walkie talkie. I.e., the on-board Arduino will:

  1. Turn the walkie talkie on
  2. Press and hold the PTT button
  3. Release the PTT button before the 30 sec max interval is up
  4. Turn the walkie talkie off
  5. Wait X minutes
  6. Repeat

Depending on the length of the flight, we could reduce the complexity further by only doing step (1) once, and omitting step 4.

It won't be as nice as triggering the on-board PTT button via a ground-based signal, but it might be a little more robust (maybe?) in that we only ever have to successfully broadcast one way.

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 31, 2012

Member

That's an excellent protocol. It's probably the most straightforward way to use your walkie-talkies to send camera sound to the ground. It could probably be done by 555 timers and MOSFETs. The advantage of that would be that a completed device could be produced in kit form for $30, and many people could benefit from it.

The advantage of employing an Arduino is the extensibility. The Arduino could eventually also trigger the camera(s), operate pan and tilt servos, change camera settings, and even acquire camera data and alert the user based on that instead of just sending audio of the shutter. And maybe a ZigBee could replace the walkie-talkie altogether making ground to air communication for control purposes easier.

I still want to hear more about using an Xbee without an Arduino. That could be an attainable entry point for more people.

Member

Fastie commented Aug 31, 2012

That's an excellent protocol. It's probably the most straightforward way to use your walkie-talkies to send camera sound to the ground. It could probably be done by 555 timers and MOSFETs. The advantage of that would be that a completed device could be produced in kit form for $30, and many people could benefit from it.

The advantage of employing an Arduino is the extensibility. The Arduino could eventually also trigger the camera(s), operate pan and tilt servos, change camera settings, and even acquire camera data and alert the user based on that instead of just sending audio of the shutter. And maybe a ZigBee could replace the walkie-talkie altogether making ground to air communication for control purposes easier.

I still want to hear more about using an Xbee without an Arduino. That could be an attainable entry point for more people.

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Aug 31, 2012

Member

I'm beginning to learn, I think, aspects of the flavor that all of these solutions should have: accessible, inexpensive, robust, etc. It'd be nice to make up a list of characteristics, guidelines. Seeing the PLOTS Kickstarter Spectrometer kit really helped to solidify some of these project traits for me. There are various kits, seemingly aimed at various use-cases, communities, and levels of technology involvement. It's very useful that you're separating some of these use-cases / communities for this project, as well.

I think you're right -- if the 555 timer board you're already using can trigger the cameras to fire, I'd think that it wouldn't be difficult to make a circuit that went through the above protocol with walkie talkies. Sorry if you're already sent it, but could you post a link to the info on the 555 timer board you use?

I agree that the Arduino is, at least in one sense, complete overkill for this purpose -- the Arduino truly becomes more of an advantage in terms of extensibility. But in some ways it's also a direct competitor even for the 555 timer based solution -- even with no thought to adding additional features -- in that the timing intervals and the triggering functions would be easily user-customizable, even out in the field if a laptop were present, without having to fiddle with electronics. I don't know enough about the 555 solution to know how hard-wired its timing is. In any case, I imagine the 555 solution is cheaper than an Arduino solution, as the Arduino itself is $25.

Ultimately -- as you're pointing to -- it's a matter of use-case / background / taste. Some people will prefer to have a kit that is plug and play, and wouldn't even want to fiddle with options -- "Just send me the kit, tell me what to buy at Radio Shack, what to plug into what, so I can start taking pictures." (I confess to being in this category for most things, especially when first trying something out).

Because I'm more comfortable with an Arduino than with a 555 timer circuit, I'll try the Arduino solution first; but I'm certainly convinced that we should also try the 555 timer approach.

Will be biking 3*2 hours with Ben Gamari this weekend, and as long as I can keep up, will grill him re: Xbee + Arduino.

Member

dwblair commented Aug 31, 2012

I'm beginning to learn, I think, aspects of the flavor that all of these solutions should have: accessible, inexpensive, robust, etc. It'd be nice to make up a list of characteristics, guidelines. Seeing the PLOTS Kickstarter Spectrometer kit really helped to solidify some of these project traits for me. There are various kits, seemingly aimed at various use-cases, communities, and levels of technology involvement. It's very useful that you're separating some of these use-cases / communities for this project, as well.

I think you're right -- if the 555 timer board you're already using can trigger the cameras to fire, I'd think that it wouldn't be difficult to make a circuit that went through the above protocol with walkie talkies. Sorry if you're already sent it, but could you post a link to the info on the 555 timer board you use?

I agree that the Arduino is, at least in one sense, complete overkill for this purpose -- the Arduino truly becomes more of an advantage in terms of extensibility. But in some ways it's also a direct competitor even for the 555 timer based solution -- even with no thought to adding additional features -- in that the timing intervals and the triggering functions would be easily user-customizable, even out in the field if a laptop were present, without having to fiddle with electronics. I don't know enough about the 555 solution to know how hard-wired its timing is. In any case, I imagine the 555 solution is cheaper than an Arduino solution, as the Arduino itself is $25.

Ultimately -- as you're pointing to -- it's a matter of use-case / background / taste. Some people will prefer to have a kit that is plug and play, and wouldn't even want to fiddle with options -- "Just send me the kit, tell me what to buy at Radio Shack, what to plug into what, so I can start taking pictures." (I confess to being in this category for most things, especially when first trying something out).

Because I'm more comfortable with an Arduino than with a 555 timer circuit, I'll try the Arduino solution first; but I'm certainly convinced that we should also try the 555 timer approach.

Will be biking 3*2 hours with Ben Gamari this weekend, and as long as I can keep up, will grill him re: Xbee + Arduino.

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Aug 31, 2012

Member

Personality determines a lot of this. Some would be challenged to solve the Old FART problem for $20 using 555s, others would spend $150 for Arduino components to do it right. An affordable 555 solution will have some immediate applicability to many others taking kite or balloon photos. Once you move to Arduinos, it becomes a personal hobby project until it can be refined enough to be worth $150 and the price of the learning curve. One fear is that PVOS will lose interest in the project before it reaches that stage, and few will benefit from your effort.

There is no question that starting down the Arduino path is enticing because it can lead to valuable real time control of KAP and BAP rigs with the potential to improve the reliability and quality of photographic results. It might take a great deal of effort to build a system that is worth the money and learning curve to many people, so the decision to go that route should be made carefully.

The basic 555 IC costs about $1.00, so the potential for an affordable solution is great. The kit Jeff shipped with the PLOTS IR kit is the Velleman MK111 timer kit for about $5.00. It is widely available. Here is a circuit diagram and assembly instruction insert. There are two problems with this kit: 1) the interval selectable with a potentiometer cannot be longer than 60 seconds, and 2) the old electro-mechanical relay runs on 12 volts, and is wasteful of current. KAP rigs often have 5v power supplies, but rarely 12v. We learned to modify the MK111 to work without the relay for triggering Powershots so it would need only a little current at 5v. Here is another kit with longer intervals, but it still needs 12v to run an old relay.

These kits only produce a single sequence of regular events. To actuate two different events or actuate an event on an irregular schedule, more than one 555 might be required. So building a tidy 555 solution will require some old fashioned electrical engineering skill. I sense the PVOS crowd is interested in applying a different type of skill.

Member

Fastie commented Aug 31, 2012

Personality determines a lot of this. Some would be challenged to solve the Old FART problem for $20 using 555s, others would spend $150 for Arduino components to do it right. An affordable 555 solution will have some immediate applicability to many others taking kite or balloon photos. Once you move to Arduinos, it becomes a personal hobby project until it can be refined enough to be worth $150 and the price of the learning curve. One fear is that PVOS will lose interest in the project before it reaches that stage, and few will benefit from your effort.

There is no question that starting down the Arduino path is enticing because it can lead to valuable real time control of KAP and BAP rigs with the potential to improve the reliability and quality of photographic results. It might take a great deal of effort to build a system that is worth the money and learning curve to many people, so the decision to go that route should be made carefully.

The basic 555 IC costs about $1.00, so the potential for an affordable solution is great. The kit Jeff shipped with the PLOTS IR kit is the Velleman MK111 timer kit for about $5.00. It is widely available. Here is a circuit diagram and assembly instruction insert. There are two problems with this kit: 1) the interval selectable with a potentiometer cannot be longer than 60 seconds, and 2) the old electro-mechanical relay runs on 12 volts, and is wasteful of current. KAP rigs often have 5v power supplies, but rarely 12v. We learned to modify the MK111 to work without the relay for triggering Powershots so it would need only a little current at 5v. Here is another kit with longer intervals, but it still needs 12v to run an old relay.

These kits only produce a single sequence of regular events. To actuate two different events or actuate an event on an irregular schedule, more than one 555 might be required. So building a tidy 555 solution will require some old fashioned electrical engineering skill. I sense the PVOS crowd is interested in applying a different type of skill.

@dwblair dwblair referenced this issue Sep 12, 2012

Open

Sanitation #17

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Sep 18, 2012

Member

Burned 3 fingers while soldering the walkie talkies last night, and to no avail -- I think I done broke 'em. Will post photos of my failure asap. I'm going to grab another walkie talkie pair at Radio Shack Thursday evening, and do a more elegant, minimally-invasive-surgery hack -- with the idea that at the end, it'll look and work like the original walkie talkie, but will have two wires dangling out that allow for Arduino control. That's the goal, anyway. Hope I can get it put together in time for #leaffest2012 !

Member

dwblair commented Sep 18, 2012

Burned 3 fingers while soldering the walkie talkies last night, and to no avail -- I think I done broke 'em. Will post photos of my failure asap. I'm going to grab another walkie talkie pair at Radio Shack Thursday evening, and do a more elegant, minimally-invasive-surgery hack -- with the idea that at the end, it'll look and work like the original walkie talkie, but will have two wires dangling out that allow for Arduino control. That's the goal, anyway. Hope I can get it put together in time for #leaffest2012 !

@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Sep 18, 2012

Member

Can't wait for the photos of your fingers. But that's bad news about the walkie-talkies. I wonder if its worth considering the VOX route if you are going to make a fresh start. But that might cost more than $20.

Member

Fastie commented Sep 18, 2012

Can't wait for the photos of your fingers. But that's bad news about the walkie-talkies. I wonder if its worth considering the VOX route if you are going to make a fresh start. But that might cost more than $20.

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Oct 31, 2012

Member

Revisiting this project ...

Member

dwblair commented Oct 31, 2012

Revisiting this project ...

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Oct 31, 2012

Member

I decided to skip the MOSFET soldering project for tonight, and went for a (potentially) easier project: activating a servo via walkie talkie. Turns out that it's super simple:

  • Remove the speaker from one of the walkie talkies, and solder wires onto the + and - speaker wire spots on the walkie talkie circuit board
  • Stick those wires out the grill on the front of the walkie talkie, and replace the cover
  • Hook up the + wire to an analog i/o pin on an Arduino
  • Now you can simply "listen" on that pin for any voltage changes due to audio being received by the walkie talkie!

I did precisely this, and also added a servo into the mix -- when the receiver hears audio above a certain threshold, it triggers the servo to rotate a few degrees. The video is here:

http://youtu.be/JpUsNvUDvhg

-- it works!

Next steps:

  • Interpret more complex audio signals -- e.g., morse code -- in order to trigger more complex actions
  • Two-way communication (e.g., the servo-controlling walkie talkie send the other walkie talkie an "action complete" message
Member

dwblair commented Oct 31, 2012

I decided to skip the MOSFET soldering project for tonight, and went for a (potentially) easier project: activating a servo via walkie talkie. Turns out that it's super simple:

  • Remove the speaker from one of the walkie talkies, and solder wires onto the + and - speaker wire spots on the walkie talkie circuit board
  • Stick those wires out the grill on the front of the walkie talkie, and replace the cover
  • Hook up the + wire to an analog i/o pin on an Arduino
  • Now you can simply "listen" on that pin for any voltage changes due to audio being received by the walkie talkie!

I did precisely this, and also added a servo into the mix -- when the receiver hears audio above a certain threshold, it triggers the servo to rotate a few degrees. The video is here:

http://youtu.be/JpUsNvUDvhg

-- it works!

Next steps:

  • Interpret more complex audio signals -- e.g., morse code -- in order to trigger more complex actions
  • Two-way communication (e.g., the servo-controlling walkie talkie send the other walkie talkie an "action complete" message
@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Oct 31, 2012

Member

That looks like fun. You should tell everyone coming to Cocodrie to bring an FRS walkie-talkie and set them to the same channel. During the talks the servo can be on the podium with a big sign saying SPIN ME IF YOU LIKE THIS and the audience can give live feedback to the speaker.

The two way communication option is intriguing. A problem with standard RC control of pan/tilt KAP rigs is that once the rig is out of sight range you don't know which way the camera is pointed. A ground-based Arduino could interpret the data coming from the air and keep track of how much each servo had rotated and display the direction the camera was pointed.

You could also have another servo on the podium with a SPIN ME IF YOU HATE THIS sign with a walkie-talkie set to another channel. (you could forget to put the batteries in that one)

Member

Fastie commented Oct 31, 2012

That looks like fun. You should tell everyone coming to Cocodrie to bring an FRS walkie-talkie and set them to the same channel. During the talks the servo can be on the podium with a big sign saying SPIN ME IF YOU LIKE THIS and the audience can give live feedback to the speaker.

The two way communication option is intriguing. A problem with standard RC control of pan/tilt KAP rigs is that once the rig is out of sight range you don't know which way the camera is pointed. A ground-based Arduino could interpret the data coming from the air and keep track of how much each servo had rotated and display the direction the camera was pointed.

You could also have another servo on the podium with a SPIN ME IF YOU HATE THIS sign with a walkie-talkie set to another channel. (you could forget to put the batteries in that one)

@dwblair

This comment has been minimized.

Show comment
Hide comment
@dwblair

dwblair Oct 31, 2012

Member

Brilliant!
Quick follow-ups:

  • Very simple initial version of the device you're describing: an accelerometer on the rig that transmits an audio signal whose frequency is a function of the rig's orientation, and broadcasts the sound via walkie talkie -- by listening in to the walkie talkie, one can tell a) the orientation b) how stable / unstable the rig is (will sound like a cartoon sound effect whistle)
  • We can even use voice activation to control the rig ... totally hands-free ... you whistle three short tones, and the rig moves one way ... you whistle one long tone, the rig moves another way ... and heck, a belt-housed accelerometer, so that a swing of the hips makes the camera shutter click ... a little hula dance rotates the servo and captures a panorama shot ...
Member

dwblair commented Oct 31, 2012

Brilliant!
Quick follow-ups:

  • Very simple initial version of the device you're describing: an accelerometer on the rig that transmits an audio signal whose frequency is a function of the rig's orientation, and broadcasts the sound via walkie talkie -- by listening in to the walkie talkie, one can tell a) the orientation b) how stable / unstable the rig is (will sound like a cartoon sound effect whistle)
  • We can even use voice activation to control the rig ... totally hands-free ... you whistle three short tones, and the rig moves one way ... you whistle one long tone, the rig moves another way ... and heck, a belt-housed accelerometer, so that a swing of the hips makes the camera shutter click ... a little hula dance rotates the servo and captures a panorama shot ...
@Fastie

This comment has been minimized.

Show comment
Hide comment
@Fastie

Fastie Oct 31, 2012

Member

I think you might need a GPS module as well as accelerometers to know which way the rig was pointed. Unless the Arduino can remember all the changes in position and keep track. Without GPS or at least a compass, you could calibrate at the beginning (point the camera north and horizontal and push GO). Then you have to be able to translate every accelerometer blip into a new camera orientation. You could do the same with servo moves if the time a servo was activated can be translated into angular rotation. I think stepper motors are better at this than servos because the rotation is a more precise function of signal (don't know though). With the servo approach, you ignore every swaying motion and assume that the camera rig is always hanging straight down from a kite line which is being blown in one direction by an unchanging wind. The Picavet or pendulum suspension maintains its orientation with the kite line. That should be good enough most of the time.

The belt-housed kite controller has already been done:, but the goal was somewhat different.

Member

Fastie commented Oct 31, 2012

I think you might need a GPS module as well as accelerometers to know which way the rig was pointed. Unless the Arduino can remember all the changes in position and keep track. Without GPS or at least a compass, you could calibrate at the beginning (point the camera north and horizontal and push GO). Then you have to be able to translate every accelerometer blip into a new camera orientation. You could do the same with servo moves if the time a servo was activated can be translated into angular rotation. I think stepper motors are better at this than servos because the rotation is a more precise function of signal (don't know though). With the servo approach, you ignore every swaying motion and assume that the camera rig is always hanging straight down from a kite line which is being blown in one direction by an unchanging wind. The Picavet or pendulum suspension maintains its orientation with the kite line. That should be good enough most of the time.

The belt-housed kite controller has already been done:, but the goal was somewhat different.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment