OF
Well-Known Member
@OF
Also I don't follow this thread much so I don't know where you guys stand, but over in the Project thread we all switched to the "myevic" open source firmware for our joyetech devices. It offers a new PID algo and much improved TC stability.
So I finally got around to trying this out (thanks again @KeroZen). Lots of fun stuff for geeks to play with (half of which I haven't messed with.....), and it fixes a personal pet peeve: The stupid "Art Deco" clock. It's hard enough reading the time on that rascal without making a round dial square on a whim. The software author is right when he said his (round) clock was "close to the one viewed on the VTwo box; just better, imo."
And you can enjoy the clock when using TCR mode, factory only allows this in fixed power or voltage due to screen use?? The myevic fixes this elegantly.
I'm not sure about the stability part, I see the occasional 500 plus temperature flash past when it's supposed to be under 390...... I suspect that's some artifact of hardware, but both SW sets would have to deal with it. Such sensor errors could dominate that stability thing? And moreover there's damping in the load going on, the temperature indication isn't the load temperature (no sensor there). Both sets are, I think, sufficiently stable for everyday use.
However, the timing is very lucky. Because I'm also fiddling with the Gen 2 herb cart it was natural to use that rather than V3. This meant switching my Gen 2 cart around between mods. This uncovered another TC issue that needs consideration: It seems Gen 2 is much more sensitive to exact temperature than V3 or the other DT offerings for concentrates.
To me this makes some sense from an output POV. You can 'dab' concentrate on scary hot surfaces without serious issue. In fact, I hear some guys even like doing that. The heater being say 20 or even more degrees hotter would not even be noticed? The concentrate is just going to boil away, holding the surface temperature down most likely in the process. Not so a herb cart. 20 degrees is the difference between smooth sailing and combustion. With little evaporation taking energy away, the surface will get hotter and hotter. Not what we want.
So, I think a critical factor is the 'cold' (saved) temperature. The one 'locked in'. The mod calculates the number of degrees rise to the target (say 390F minus seventy for 320 degree rise needed?). That is then multiplied by the 'percent per degree' factor, the "m" value. In my case it's 245. That's in a strange numbering system (divide by 100,000 not the 'more metric' even million....). This means for each degree, the resistance will go up (down if the TC were negative) by '245/100,000'.....almost exactly 1/4 of a percent. Most metals are in that range, .3 something for Copper IIRC. Any way, somewhere in here we have to correct for those upstarts and convert from degrees F to C (since the 'percent per degree' number is already in C). This means multiplying by 5 and dividing by nine since there are 180 degree F (32 to 212) and 100 C (0 to 100) from melting to boiling for water (this is the scale factor, no need for offset correction here). So we have 177 degrees C change times .245 % per degree C multiplied together (the 'degrees C parts cancel) and we're left with 43.8% increase.
If it 'remembers' say .80 Ohms, it's looking for that same .80 plus 43.8% more for 1.15 Ohms (.35 Ohms more). That's what it sees the goal to be, heat to there. A resistance calculated against memory and settings.
But, if there's an error of say .05 Ohms, nearly trivial and common, this can be an issue I think. 6.25%. On our 320 degree rise that's a 20F error. Over two full steps on Solo? That might not be too big a deal with concentrate or juice, but with herb it's a significant difference. My experience is 'resetting' the atty will often make it produce better with the numbers I'm using. That is it can 'run cold' only to improve by unlocking and resetting to a lower cold reference resistance (lower 'locked' value). Often by a few tenths of an Ohm. FWIW the exact same cart can read different values on another mod, often does. But since control is relative there, I suspect that part doesn't matter much?
Happy Friday to the wage slaves, best regards to all.
OF