Frustrated "No Pit Probe"


 

JayDBalt

New member
OK I am a noob here. I got everything built. Up and working, and seeing temps with (some) probes. I got a thermocouple probe today, plugged it in and it worked. I was having some trouble getting a meet probe to work. I got Maverick ET 73 probe for temp and that was working fine, and showing temps on all 3 ports. I was having trouble to get a reading from an ET-732 food probe. I was changing the settings in the browser, and did an AVR reset, and it lost connectivity. Now if I power everything up, I just get 3 LED's and No Pit probe on the display. No buttons work. I feel like I have good solder joints everywhere, I am no expert, but do OK. Is there a way to reset the AVR?

Any help would be much appreciated.

Thanks,

Jay
 
Found how to push firmware. Does not work. Do I have an AVR chip problem?

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% |*******************************| 64041 0:00:00 ETA
Download completed (64041 bytes)
Stopping LinkMeter OK

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

5bcfbd117f3c318dc22f7a34cf075a73 /tmp/hm.hex
hmdude: compiled on Oct 12 2017 at 13:13:56
Can't set AVR programming mode (0x00)
Using port: /dev/spidev0.0
Loading ihex file: "/tmp/hm.hex" (22758 bytes)
Update failed
Starting LinkMeter OK
 
Err yeah that's a problem because if it can't flash firmware it probably isn't a software problem. The "Reset AVR" on the configuration page doesn't clear anything, it just sends a software reset command to the AVR to reboot it. Interesting that it turns on and starts at operating the LCD though, and lighting all 3 of the LEDs is done on startup to show they're all working and then they're supposed to turn off after 1 second. Sounds like it is locking up or something before that 1 second mark?

You can try it without a Pi attached and see if it operates normally that way maybe? You could also try to flash the "snapshots/trunk/reset-eeprom.hex" and see if that takes and resets the HeaterMeter config (after which you can flash it with the regular snapshot) but I don't think it will flash either if it can't enter programming mode. You can also maybe try to see if it operates connected to the Pi but without the LCD/Button board attached (boot it up without the secondary board). It's kind of an unusual situation you've got there, especially since it worked before.
 
It is weird that it worked fine. And just trying different probes seems to have killed it. I looked over all of the solder joints. I tried another Atmega328-PU, but am not sure how to get it to be initialized by the rPi. I don't know enough to test the board. I think I made a fancy paperweight :(
 
Brian, which maverick probe required was it that if pushed all the way in would not work, you had to back it out abit? Maybe thats his problem??
 
If you put a blank atmega328 in there (P or non-P, they both work the same) then HeaterMeter will attempt to program the fuses and firmware when it starts up, after about 10 seconds of not receiving any data from the chip. It just assumes it is a new install and resets it to stock. If you put in a new chip and still are getting nothing, even the "AVR Firmware" flash isn't seeing it, then there's a problem on the SPI bus or a 3.3V power problem at the chip (check VCC/GND and AVCC/GND for 3.3V). Also try to flash it from the webui without the LCD/Button board attached which will take a few things out of the equation.

The probes with the toolong connectors are those ET-732. They click in about 1mm short of being inserted all the way and if you push them in further they will go back to being offline. When pulling the plug back out slowly, you'll feel it go easily for the first 1-2mm then sort of snap into place, then take a bit extra force to continue pulling. An easy fix to this is to put some sort of small o-ring on the mono pin to keep it from getting pressed in all the way and disengaging. That could have been the initial problem but now there's definitely something bigger happening.
 
Guys,
I truly appreciate the help. I just ordered a pre-soldered board. I do not have the knowhow to diagnose the problem. I do not know how to check VCC/GND and AVCC/GND. I do have a voltmeter :). I just do not understand why it worked, and then failed. I did try a different chip because I thought maybe the AVR got corrupted. No workie. I had several probes from an old wireless monitor. And a couple of new sets. But I definitely got a non-meat probe to work in all ports. And the thermocouple worked fine. I was just trying to get one of the other meat probes to work. (I'll play with the plug when I get the new board.) It does bug the crap out of me that something is amiss. I am an IT guy, and face weird stuff like this all the time. You are one simple step away from a fix, you just don't know it. And sometimes you do not know exactly what you did to fix something. But learning the electronics as I go. I just want to cook :) Stay tuned!!!! And thanks for the help!
 
OK after further review (and GTS (Google that S**t)) I started testing voltage all over the board. And then (I did read before) I noticed some components did not have solder all of the way through. Several on the AVR chip header/plug. I went back through and redid the soldering. No luck. I'm 99% sure I was a) really lucky that it worked at all in the first place, b) waiting on the pre-built board.

One other thing I remembered. I was working with the meat probes (with the pins that are to long) at the time. I wonder if the braided shield got under the board on my pad. I'm assuming they are grounded and it would not be good to have them touch the board. But nothing looks smoked. I got some new ATMEGA328 P with a bootloader and no difference.

I do see the LCD and LED's blink after I apply power after roughly 90 seconds. And sometimes when I plug in I get - No Pit Probe - and sometimes I don't. After 30 seconds or so the - No Pit Probe - message goes away. And it acts the same if there is a Pit Probe (that worked before) is in or out.

I'm just posting all of this for the next person.

One last thing. On the troubleshooting page there is a board with voltages. I assume this is for a 4.2 board. Is there a drawing for the 4.3?
 
So, I have gone through everything twice just checking voltages. I thought for sure it was something on the header to the rPi. Question: How do you measure the high frequency signal at 3.3 volts on the headers? Oscilloscope? The new boards work perfectly BTW. Thanks.

As an aside. This all sucks because one of the most awesome electronics shops around was 5 minutes away. They closed last year, and now it is a Sonic drive in..... I could have walked in and said help And there would have been a huddle with 4 people working on it.
 
Question: How do you measure the high frequency signal at 3.3 volts on the headers? Oscilloscope?
You can use an oscilloscope, but you can also sort of see it on a multimeter. Like 0+ means you'll see 0V but it will bounce up as data is moving but probably not to the same value each time it moves and there's no telling what you'll get because it depends highly on the multimeter and how much data is crossing. So for those you're just looking for voltage either usually close to 0V or 3.3V but with some amount of variation.
 

 

Back
Top