Debugging assistance please


 
D

Dale Swanston

New member
Hi - I have a HeaterMeter v4.3.4 (and RPi 3b) that I soldered up a few months ago. I've done 3 cooks to great success. But today the heatermeter will not cooperate.

I booted it this morning to come up with either blocks in the LCD display, nothing in the display or random characters with each power cycle and otherwise unreponsive buttons.

Not trusting my soldering skills I opened it up and reflowed a number of connections (I removed the ATMEGA when doing the socket connections) and voila it started to work. Put back into case, still worked. I moved it out to the Weber and it didn't work anymore. Repeat 2 more times. On the last go around it appeared to work fine, LCD, fan and wifi were all working like normal. But after 30-45 minutes or so it become unresponsive (last pit temperature was displayed on the LCD but the web interface showed just dashes for pit temp).

I was reading the thread "Why can't this thing just work... (frustrated)" and in there are some debugging techniques that I will try but I thought I might ask for some areas to focus on in the meantime.

Other data points
- it also doesn't work without the RPi attached so I don't think the Pi is the problem (guessing though)
- I ensured the LCD pins were NOT touching the USB connectors of the Pi when fully assembled
- in the last instance that it was working (for a while) the pit probe config had been changed (corrupted) away from the previous setting (ie it wasn't set to a thermoworks probe but something else ... I reconfigured it back to and all was fine). Does this suggest something with the ATMEGA connections possibly? Or just the result of numerous power cycles?

Anyhow, any ideas are most appreciated.
(Brisket is patiently waiting in the fridge.)
Many thanks in advance.
 
That's definitely unusual behavior. I'd say the loss of configuration is tied to the other issue you're experiencing.

The LCD should never be garbled, there's no configuration needed for the LCD to work so it should always initialize properly and display something meaningful (although the display mode might be wrong if the config is corrupted). If something is shorting somewhere causing the communication to be messed up, that could be browning out the atmega somehow as well causing the lockup and loss of config. Unfortunately I'm not sure what where to tell you to look apart from what you've already verified, which is that the LCD wasn't touching the USB ports. The fact that it does it with no Pi and doesn't even start properly is a big red flag, but that's not a condition that there's a definite cause for (like a single probe reading the wrong value, or the blower not blowing).
 
Is it possible that you're using a power supply that isn't up to the job? That might explain why it turns on and then is intermittent
 
@Bryan Thanks for the detailed responses and I'm glad I just haven't missed anything totally obvious. I'll take another look at the LCD connectivity to see if a joint is bridged there.

@Brad I became suspicious of the power supply as well (though it did work previously on 3 roasts) and checked the 12V supply and it was measuring 20V (I tried a few other 12V power supplies lying around and they all measured higher than 12V - maybe I'm doing it wrong but just checking by sticking voltmeter probes into the power supply). In desperation I purchased a new PS (amazon) but it too measured more than 12V so I am not sure what to think on that front.

My next plan of attack is to order another HM and try again, I think. Then if it works I should be able to debug the first one. There must be a solder bridge that I just can't see. Will do so when back from holiday.

Thanks again.
 
Dale
From your description you might be using a non-regulated power supply. That`s not a good thing. I have inserted a link for a supply from Amazon which would work well with HM. Also if you live in an area where the outside temps are running in the 90`s and higher, find some way to cool the case of the HM. They can get a little hot when its hot outside and that could and I have found with thermocouple HM cause temp reading to drift off because of the heat with the thermocouple amps. Electronics do not like heat.

https://www.amazon.com/dp/B07HNV6SBJ/?tag=tvwb-20
 
@Gary Thanks very much for the info and the link. I will definitely pick up that power supply.

Is this common knowledge or did I miss some specifications in the HM documentation (entirely possible)?
 
You must log in or register to reply here.

 

Back
Top