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

base ohm reading #127

Closed
RenegadeTitan opened this Issue Nov 3, 2016 · 15 comments

Comments

Projects
None yet
3 participants
@RenegadeTitan

RenegadeTitan commented Nov 3, 2016

any chance the base ohm reading could be extended an extra digit to thousandths of ohm in coil settings instead of hundredths of ohm? i believe this would help a lot in calibrating displayed temperature readings and to a lesser extent slightly increase accuracy of temp control regulation and algo pid adjustments. my current coil is .096 ohms and when locked in at .10 it requires 350f setting to fire correctly or locked in at .09 requires 520f setting to fire correctly. quite a wide range for closest options to actual ohms. i see option in nfe configuration for setting thousandths of ohms. im having an issue uploading any changes made in nfe to my device. this is a separate issue and i will experiment before opening new issue for that.

@ClockSelect

This comment has been minimized.

Owner

ClockSelect commented Nov 4, 2016

Assuming TCR=120 and Rreal=0.096Ω,
In the first case (0.100Ω, 350°F) your target rez is 0.115Ω (185°C/365°F)
In second case (0.090Ω, 520°F) your target rez is 0.117Ω (214°C/418°F)
It's a bit different but let's assume you really want 200°C (392°F), we're around a target rez of 0.116Ω.
You want the rez to be 0.116Ω when T° shows 200°C
Say you lock your rez at 0.090Ω (better lower, avoid sub-zero temps).
You should be good by setting your TCR around 160.

Ok, it's a workaround.
But I must see what your change implies - rez are used everywhere in calculations and it's easy to forget something somewhere.

Problem in your case is that you use very low sub-ohm coils with low a TCR metal; combination of the two gives temp control a very high milliohm sensitivity (around 10~15°F/mΩ).

@ClockSelect

This comment has been minimized.

Owner

ClockSelect commented Nov 4, 2016

SS316L, 0.75Ω, RX200S, 75W

Before (10mΩ accuracy):

After (1mΩ accuracy):

Seems there's a little gain in temp stability, but that's not obvious with my setup (0.75Ω, I'm far from your needs).
Anyways, it works.

ClockSelect added a commit that referenced this issue Nov 4, 2016

@ClockSelect

This comment has been minimized.

Owner

ClockSelect commented Nov 4, 2016

Give it a try.

bluenazgul added a commit to bluenazgul/myevic that referenced this issue Nov 4, 2016

@RenegadeTitan

This comment has been minimized.

RenegadeTitan commented Nov 4, 2016

@ClockSelect i went back and forth between firmwares about 15 times each relocking base resistance at .09 @ 520 degrees and .096 @ 420-430. seems to be nearly identical fluctuations as it regulates (on further thought seems it should not make a difference in functionality as long as you avoid higher ohm locked in than actual coil). possibly slight improvement but dont see how it would(again referring to functionality of regulation). one noticed behavior that i do not understand is that when on .01 accuracy firmware i lock resistance to .10 then immediately to .09 to remove any previous set base resistance. performed test. then flash to .001 accuracy firmware (again) and .096 is automatically showing as locked (even if coil is warm and still above this) so it appears it is somehow pulling this from memory. maybe you used a different memory location/variable to function as displayed base ohm reading? not an issue just found it interesting.

as far as accuracy of displayed temperature this is greatly improved (through the entire range from 70-vape temp). allows much more room for calibrating lower ohm coils. mostly an aesthetic improvement but one that i enjoy a lot. i think this is a significant improvement. thank you once again :)

@ClockSelect

This comment has been minimized.

Owner

ClockSelect commented Nov 4, 2016

In fact, the "0.09" is stored as usual and the "6" elsewhere. Problem with dataflash is that if you wish to modify its data structure, you can only do so incrementaly, conserving content and meaning of previous locations, to allow compatibility between firmwares and versions of firmwares. Except if you accept a dataflash reset at some point.

@RenegadeTitan

This comment has been minimized.

RenegadeTitan commented Nov 4, 2016

small idea, maybe since most coils are under an ohm display on main screen .xxx accuracy unless over 1.00 ohm then display x.xx accuracy? just an idea for if your looking for something to do :)

@ClockSelect

This comment has been minimized.

Owner

ClockSelect commented Nov 4, 2016

That's indeed an idea.

@ClockSelect

This comment has been minimized.

Owner

ClockSelect commented Nov 4, 2016

Here's what you should get if the resistance reading was less noisy (VTC Dual); you can compare with graphes above:

@RenegadeTitan

This comment has been minimized.

RenegadeTitan commented Nov 4, 2016

comparing the 2 devices, what causes the noisy ohm reading, circuitry or 5/10?

that vtc dual is quite stable lol

@ClockSelect

This comment has been minimized.

Owner

ClockSelect commented Nov 4, 2016

Hm... Seems you're right. I'll cancel my previous comment.

Can't be 100% sure, but say 99% for circuitry. I have the same result with an RX that had his 510 replaced.

@t2dYuri

This comment has been minimized.

t2dYuri commented Nov 4, 2016

@ClockSelect thanks for the update. For now, i have locked 0.73Ω in the second row of main screen, and 0.739Ω in the coils management menu. I understand, that it's just visual differences, but may be it would better to observe the same value in both places?

@RenegadeTitan

This comment has been minimized.

RenegadeTitan commented Nov 4, 2016

:)

@ClockSelect

This comment has been minimized.

Owner

ClockSelect commented Nov 4, 2016

That was the point of 7 comments above^^

ClockSelect added a commit that referenced this issue Nov 4, 2016

@ClockSelect

This comment has been minimized.

Owner

ClockSelect commented Nov 4, 2016

Display adapted.

@RenegadeTitan

This comment has been minimized.

RenegadeTitan commented Nov 4, 2016

ok thanks. i already removed that question and reposted as new issue. i posted it on nfe site also.

live display of milliohms looks good

@ClockSelect ClockSelect closed this Nov 4, 2016

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