debounce (wait msecs) option for pulse type filament monitors

[rob-miller]

Add 'Wnnn' option for pulse type filament monitor to specify number of milliseconds to ignore subsequent pulses after an initial pulse is received.

Documentation change something like:

'''Additional parameters for a pulse generating filament monitor'''

• '''Lnn''' Filament movement per pulse in mm
• '''Wnnn''' Wait nnn milliseconds between counting pulses (debounce)
• ''S, R, E''' As for Duet3D laser filament monitor<

[dc42]

Please explain why you think this change is necessary. Are you using a filament motion monitor that uses a mechanical rotary encoder? Or are you using the optical rotation sensor that has a known hardware design fault?

The PIO in the SAM4 has a hardware debounce facility, and if the debounce time doesn't need to be configurable then it would be more efficient to enable it.

[rob-miller]

Hi David, I'm sorry I did not see this message when it came in, odd as I was looking for a response - thank you for your quick reply. This is indeed for the optical rotation sensor with the design fault that you identified in my thread on the forum. Regardless of the source of the problem, I think a feature like this is necessary to support similar and more random/crappy pulse-type filament monitors. I currently suspect there may be a fundamental problem printing sufficiently accurate optical encoder wheels but I will continue to investigate that. I'm currently travelling but on this trip have obtained an oscilloscope to help me better understand the issues around this particular device and how I can fix them. I expect that most users won't have this type of equipment, so a dumb debounce facility which they can activate will probably be useful - but I did not know about the hardware support in the SAM4 so will investigate that and rework the patch to be more appropriate. Also on my mind is development of an 'S2' mode for the M591 Gcode to support measuring filament travel without printing for extruder calibration, but have not started any code for this yet.

…

On Thu, May 24, 2018 at 10:05 AM, dc42 ***@***.***> wrote: Please explain why you think this change is necessary. Are you using a filament motion monitor that uses a mechanical rotary encoder? Or are you using the optical rotation sensor that has a known hardware design fault? The PUI in the SAM4 has a hardware debounce facility, and if the debounce time doesn't need to be configurable then it would be more efficient to enable it. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#176 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ADNtwO6F0U0tSEeEIRIpAJEzmxnckIG3ks5t1uh2gaJpZM4UHBir> .