Heatermeter Crapped Out


 
S

SteveKran

New member
I figured I should start a new thread for this as I had brought this up in a thread of a different topic.

I'm going to try and "import" the conversation the best I can through quotes.

SteveKran said:
Now here's something strange... twice today so far... the Heatermeter will "freeze up" and not show any of the probes plugged into it. I cycle the power and it sees everything again and starts outputting with the fan.

The first time it did this the Servo pulse duration went back to the default and I didn't catch it. I ended up going to sleep for a couple hours and the egg got to about 400°F because even though the fan was off, the damper was about 25% open. Not sure for how long.

The second time this just happened and the Servo values stayed the same when I cycled the power.

Weird...
Click to expand...

Bryan Mayland said:
Eerrrr that sounds like something drastically bad happening. The HeaterMeter runs independently of the Pi so it should be rock solid despite any software weirdness happening on the Pi or something crashing there or losing network connectivity. If the HeaterMeter locks up to the point it stops at whatever output it was at and not show any probes attached on the display, something is horribly horribly wrong in hardware causing something to completely affect the microcontroller's physical operation. Or are you saying that just the webui wasn't working?

The servo losing its configuration is pretty odd too, because the only way it could reset to default is if the EEPROM became corrupted on the microcontroller and it had to reset itself to the defaults, which is everything on the configuration page. Although if the microcontroller is locking up for some reason, which would likely be power-spike related, it definitely could also reset the EEPROM as one of the symptoms.
Click to expand...

SteveKran said:
Bryan thanks for the input. I believe the actual heatermeter has taken a crap on me. I can't remember what the output was doing, but none of the probe inputs were reading. No thermocouple or probe 1 or 2.

I can log into the WebUI and see the Uptime it's been turned on.


Now the LCD screen isn't lighting up when I power it on. I AM getting all 3 LED's on the front to light up (green/yellow/red). I logged into the WebUI and noticed a couple things you might find helpful.

Any ideas on how to troubleshoot this further? It's looking like I'm going to cook ribs old school this weekend.

24ky3c4.png


Normally the alarm Thresholds show up from the previous cook. They are all blank.


2z5pzpc.png
Click to expand...

RalphTrimble said:
That does look like your HM board crapped out. I would pull it from the rPi and start by checking voltages on the HM board, 12v, 5v, 3.3v to make sure power is right to start off. If you find the power is good I would look at the ATMega socket for bad solder and make sure the chip is seated properly in the socket. There are some voltage charts available that may be helpful to narrow down where the issue may be if you have to go further than that.
Click to expand...







Ralph, I found the image below and I have some of the expected voltages....others I do not. What's the best approach from here? Is there a recommended way to go through this circuit?

Seriously....thank you for all the help! This has been a little frustrating, but I'm sure I can get it resolved. You guys are great!





Link to image: http://heatermeter.com/devel/pcb/hm-4.3/HeaterMeter434BaseV.png

HeaterMeter434BaseV.png
 
I'm going to list all the voltage readings on the ATMega chip (still plugged in). Maybe this will give you an idea of what's going on. I'm using a Fluke 189.

LE1: 3.26
No Label: 3.26
MO:3.26
MI: Changing from 1V to 1.4V
SCK:0
AVC:3.26
ARE:3.26
AGN:0
BTN:0
BFB:0
F03:3.22
F02:3.22
F01:3.23
PIT:3.18

NEXT ROW
SV0:.12
No Label: Changing from 0.08 to 0.28
ALM:0
LBK:0
X2:0.38
X1:0.49
GND:0
VCC:3.26
LCK:0
BLW:1.32
No Label:2.04
TX:3.26
RX:2.91
RST:3.25


On the P-MOSFET. I'm getting 12.2V on each of the outside legs. The middle leg is at 0.0V
 
Last edited:
Good idea starting a new thread, easier to keep track of. Something you might want to try is going to the AVR Firmware tab and trying From Online repository and then selecting snapshots/trunk/reset-eeprom.hex to flash. This can check two things, one if the AVR is functioning and can flash, and secondly it will clear its EEPROM contents completely. Another user a few weeks ago contacted me and discovered an issue where the EEPROM became corrupted somehow and the bad data would prevent the HeaterMeter from booting. The code has since been modified to work around this issue but that's pretty new so your device probably does not have the fix. Flash the reset-eeprom, and if it actually successfully programs the chip, wait about a minute then flash the hm.hex from the online repository and see if that gets you back up and running.
 
Bryan Mayland said:
Good idea starting a new thread, easier to keep track of. Something you might want to try is going to the AVR Firmware tab and trying From Online repository and then selecting snapshots/trunk/reset-eeprom.hex to flash. This can check two things, one if the AVR is functioning and can flash, and secondly it will clear its EEPROM contents completely. Another user a few weeks ago contacted me and discovered an issue where the EEPROM became corrupted somehow and the bad data would prevent the HeaterMeter from booting. The code has since been modified to work around this issue but that's pretty new so your device probably does not have the fix. Flash the reset-eeprom, and if it actually successfully programs the chip, wait about a minute then flash the hm.hex from the online repository and see if that gets you back up and running.
Click to expand...

Thanks Bryan!

So I just hooked everything back up to the PI and plugged it in. IT'S WORKING!?!? And it crapped out again...LOL

Not sure what happened...or why....

I was about to flash the EEPROM and it crapped out again.

So I just flashed the EEPROM and got a message on my LCD screen "Clearing EEPROM Done!" Below is from the WebUI

Downloading 'https://heatermeter.com/devel/snapshots/trunk/reset-eeprom.hex'
Connecting to 104.223.92.144:443
Writing to '/tmp/hm.hex'

/tmp/hm.hex 100% |*******************************| 7766 0:00:00 ETA
Download completed (7766 bytes)
Stopping LinkMeter nil poll

LinkMeter platform is BCM2708
Loading SPI modules...
AVR fuses ffd705 OK

1865990ffe76b193ff37f3cc9ab18965 /tmp/hm.hex
hmdude: compiled on Oct 12 2017 at 13:08:37
Using port: /dev/spidev0.0
Loading ihex file: "/tmp/hm.hex" (2752 bytes)

0% | | 0 (0.0s)
0% | | 0 (0.0s)
5% |## | 138 (0.0s)
10% |##### | 276 (0.0s)
15% |####### | 414 (0.0s)
20% |########## | 552 (0.0s)
25% |############ | 688 (0.0s)
30% |############### | 826 (0.0s)
35% |################# | 964 (0.1s)
40% |#################### | 1102 (0.1s)
45% |###################### | 1240 (0.1s)
50% |######################### | 1376 (0.1s)
55% |########################### | 1514 (0.1s)
60% |############################## | 1652 (0.1s)
65% |################################ | 1790 (0.1s)
70% |################################### | 1928 (0.1s)
75% |##################################### | 2064 (0.1s)
80% |######################################## | 2202 (0.1s)
85% |########################################## | 2340 (0.1s)
90% |############################################# | 2478 (0.2s)
95% |############################################### | 2616 (0.2s)
100% |##################################################| 2752 (0.2s)
Update successful
Starting LinkMeter OK
Click to expand...




When I flashed the hm.hex from the online.... the screen went blank, and all 3 LED's (red/yellow/green) are now staying lit.

Downloading 'https://heatermeter.com/devel/snapshots/trunk/heatermeter.hex'
Connecting to 104.223.92.144:443
Writing to '/tmp/hm.hex'

/tmp/hm.hex 100% |*******************************| 63967 0:00:00 ETA
Download completed (63967 bytes)
Stopping LinkMeter OK

LinkMeter platform is BCM2708
Loading SPI modules...
AVR fuses ffd705 OK

0db1b2834b87d78a65143ab19e56d5b2 /tmp/hm.hex
hmdude: compiled on Oct 12 2017 at 13:08:37
Using port: /dev/spidev0.0
Loading ihex file: "/tmp/hm.hex" (22734 bytes)

0% | | 0 (0.0s)
0% | | 0 (0.0s)
5% |## | 1138 (0.1s)
10% |##### | 2274 (0.1s)
15% |####### | 3412 (0.2s)
20% |########## | 4548 (0.3s)
25% |############ | 5684 (0.4s)
30% |############### | 6822 (0.5s)
35% |################# | 7958 (0.5s)
40% |#################### | 9094 (0.6s)
45% |###################### | 10232 (0.7s)
50% |######################### | 11368 (0.7s)
55% |########################### | 12504 (0.8s)
60% |############################## | 13642 (0.9s)
65% |################################ | 14778 (0.9s)
70% |################################### | 15914 (1.0s)
75% |##################################### | 17052 (1.1s)
80% |######################################## | 18188 (1.1s)
85% |########################################## | 19324 (1.2s)
90% |############################################# | 20462 (1.3s)
95% |############################################### | 21598 (1.4s)
100% |##################################################| 22734 (1.4s)
Update successful
Starting LinkMeter OK
Click to expand...



Not sure where to go from here? :confused:
 
Last edited:
If your HM is working at random like that I would suspect intermittent bad solder joint(s). I would go over the board with a soldering iron and reflow the solder on each joint quickly and then power back up and see if it is stable.
 
RalphTrimble said:
If your HM is working at random like that I would suspect intermittent bad solder joint(s). I would go over the board with a soldering iron and reflow the solder on each joint quickly and then power back up and see if it is stable.
Click to expand...

Thanks Ralph! I'll give it a go with I get off work in the morning. I'm a nightshift guy... 5:30pm-4am
 
Got it Heatermeter working again. Thanks for the help guys... I really appreciate it!

I'm guessing a couple solder joints were the cause of the problem. When heated I heard a little flux burn off and got a tiny bit of smoke.

Now the latest issue is the that the thermocouple and probes are reading off temp. I'm pretty sure there's more flux on the board playing with the resistance the chip is seeing. I'm thinking of pulling the ATMega chip and putting the entire board in an ultrasonic cleaner with 91% isopropal alcohol then cleaning it with demineralized water.

Thoughts/comments/concerns...?
 
No need to go with the ultrasonic cleaner, do like Steve said... Use a toothbrush and iso alcohol and scrub the area around the probe jacks and the resistors that stand on end (solder side of board of coarse), let the board dry, and then power up and see if temps read right....
 
I scrubbed the board really well. Helped some...
I swapped out the ATMega chip out. All my temps were suddenly within 3 degrees of each other (all at room temp).

Sounds like that took care of it. Thanks again guys! If you're ever in Columbia I'll buy you all a beer!
 
Update... I think there was a power surge. I found tripped surge protector. May have explained why the chip swapped helped?
 
You must log in or register to reply here.

 

Back
Top