So far so good. Nice and stable.
thatās great!!! I like your rig and who doesnāt like those little crisp, clear OLED displays?
You are probably at a point where you want to program in a bit of hysteresis such that an erroneous event (that still seems to be happening, although very infrequently) can just be āignored.ā
It isnāt uncommon to dump such values by comparing to trailing averages (e.g. 1, 2, or 3 second period sampled every 100ms) and tossing unreasonable numbers out. for temperature controllers, in particular where you want to avoid turning on/off high current devices too frequently. You can do the same thing on your POT valueā¦
Yes, currently it does nothing if the difference from the last reading was <64 which worked well until things went really crazy. Now, after 4 hours or so the max difference has been 25 with most reading being 2 or less different.
For my purposes checking a moving average one that threshold is exceeded should be easy to kill outliers.
I think this finishes my stabilization of sensors now I need to look at doing something intelligent with them. I have a background in electronic inventory management with sensors so I am going to add some features to help me manage my Oil tank. You get the cheapest prices if you manage your own tank and can then also watch oil prices to see when the best time to buy would be. Iām doing this for myself but having an eye toward expanding it for other managed commodities like propane etc.
Not to mention saving energy in the process. I have an ancient coal furnace that was converted to oil.
I also have some interesting math ahead of me because the system still heats after the flame is off. Because of this it has a simple old thermostat with an interesting feature in that it has a built in heater that heats the thermostat to āfoolā it into stopping before the room has reached the set-point. I think I can simulate this in software and smooth out jitters in the temperature at the same time.
I know one thing, Iām giving the antenna end of the chip an even larger avoid zone when I finally make the board.
You could actually move further the LCD from the photon
I like what you have there !!
Congrats!!
not too much without buying longer leads. But with more study Iām sure I can place leads better based on where they plug in.
Iām interested to hear what happens with your unruly Photon. Any chance itās the same cause? The Core was just fine in the same place so itās definitely something to look out for.
I think Iām calling this officially solved. Things have been stable and I even moved the display further away. I still get some blips but not large enough to cause trouble. Iāll write some software to assure that outliers are suppressed, Thanks for the help @BulldogLowell and @frlobo I think I might play around with isolation as well.