temp probe bleeding over to another


 
Darin J

Darin J

New member
temp probe not reading correctly

I built up my rPi Heatermeter v4.0 last week and the only issue I had was the buzzer not working but that doesn't concern me much really. However, I had only used the pit and #1 food probe until I got 2 new Maverick probes in yesterday.

I enabled #2 food and switched #3 to a Maverick preset instead of the thermistor and am getting bleedover to #2 if I have a probe plugged into #1 or #3. I've followed all tracings and don't see any solder bridges or other issues (yet, will check again after work today).

Based on the schematic there are a few likely places for a problem, any other suggestions to look for? I haven't gotten into voltage checks yet.
 
Last edited:
So when you have a probe connected to 1 or 3 you get the same reading on probe 2 connection, even when there is no probe connected. Do you have any issues when the probe 2 is connected and the other two probes are not? If you have multimeter with a continuity checker, check to see if you get any connection between the different probes. Someone else actually received a board that had trace missing, so maybe you have a board that has a trace connected when they shouldn't be.
 
I didn't get to test last night so brought it in with me but had the rPi off it and forgot it at home, so hopefully this evening. I need to make sure I'm giving the correct symptoms.
 
I'm getting basically an open at 9M-ohm between the contacts on probe #2 where the others (except #3 due to thermistor) show a capacitor charging up. I have continuity from each probe tip to the associated resistor and the other side of all 4 resistors are bonded as well as all 4 probe sleeves. More digging...
 
I have continuity from each probe tip to the ATmega328P pins on the chip so the traces/socket/etc are all good. I'm not seeing any cross-connections either, each circuit is isolated as it should be. I'm baffled right now. I'll do some live checks when I get home.
 
Yeah that is confusing as all getout. Something that might help is the ADC or resistance mode. If you log in and go to the configuration page (when you get your Pi connected) and set the setpoint to 0A (zero A) or 0R (zero R) the food probes will spit out their ADC read or the calculated resistance. That might help you see what's happening with a probe disconnected (which would normally read "off").
 
I took the thermistor out for this testing just so I can be consistent across all inputs.

No probes connected:
pit 0
#1 0
#2 ~7900 +/-100 fluctuating
#3 0

Just one probe (same one, brand new) tested in each jack at a time:
pit 7878 steady
#1 7879 steady
#2 0 solid
#3 7879 steady

I checked all 4 of my probes on a single input and all 3 of my food probes were all exactly the same resistance and the pit probe was just a hair lower.

With all 4 plugged in:
pit 7871
#1 7864
#2 0
#3 7881

With probes in pit/#1/#3, no probe in #2
pit 7871
#1 7872
#2 7560 +/-5 fluctuating
#3 7880

Probe #2 jack only has ~.8vdc across it, so I'll have to look in that direction now.

Thanks for the advice on checking resistance. There is only so much troubleshooting you can do dead, live checks help a ton. I'm still leaning toward a solder bridge somewhere since I didn't have a fine tip or fine solder, well didn't have the patience to wait until Monday.
 
So the thread title is incorrect now, it's not bleeding over, it's just not reading that one correctly.
 
I'm just going to microblog my troubleshooting here.

3.29vdc across C6, correct
0ohms from negative side of C6 to sleeve on each probe jack, so good grounding on all 4
9.92kohms from opposite sides of R5/R16/R17/R18 to respective socket pin on atmega328p with chip out of socket

And I just found it while doing that last check through the resistor to socket, it's a bad solder joint on the socket side of R16. Easy to fix, hard to find. When I was checking resistance, I was pushing down on the solder with my probe making contact but when I checked by touching the resistor lead from the other side it showed an open and when I looked closer could see that I could move it slightly.

It's understandable based on me not having a fine tip for my soldering iron or thin solder the night I started. I had time to get started, just not the patience to wait a day.

Next up is to see if I have another bad joint keeping the buzzer from sounding.
 
Good work! It is really difficult for me to troubleshoot hardware issues, as it can be something that looks fine but turns out to not be.

The buzzer in v4.0 is super simple, AVR -> Resistor -> Buzzer -> Gnd, should be easy to troubleshoot that one by comparison!
 
Get some resin flux at a local radioshack and dab a bit with a q-tip on the bad joint. Then resolder and adding more solder, that should get you a good joint.
 
I agree on hardware troubleshooting, it just took me sitting down with some time to dig in and do the work. I figured it had to be something simple and it was. I just resoldered that joint and all is well. Thanks for the resistance tip though, that was helpful.

The buzzer circuit checks fine, maybe I just don't have it configured? It's driven off the resonator, right? Or just straight 3vdc?
 
And the buzzer was just me not configuring an alarm properly, all good. I think I'm 100% functional now.

Thanks for the help!
 
You must log in or register to reply here.

 

Back
Top