heatermeter suddenly died?


 
It has crossed my mind.....maybe the card is damaged. Will try it with a new one as soon as i get home. It would be kinda great if that was the problem.... i will ofcourse keep you all updated
 
Cards do die, particularly the micro sd. I had one suddenly get stuck in read only for no apparent reason. I tried everything and couldn't get it set right.
When you said you swapped rPi with a neighbor I assumed you meant he had a HM and you moved his rPi and sd card, now I realize he may have had an rPi for other purposes. By all means, try another SD card, particularly because your rPi is using the micro SD.
 
I tried another (week old) card, but unfortunately didn't gave me the pleasing result. Contacted my neighbour, and he may have a colleage that can help me out with searching and maybe solving the error (for a bottle of belgium beer he said he would give it his best try). In the mean time i will get going with finishing my dry-aging fridge.... keeps me active i guess. As soon as i have more info, i will let you guys know (with the error that caused it). If it really is an endless problem, i will order a new one, but for now i will get my hopes up on my neighbour (fingers crossed).
 
Just a little update.... mr Peter F respectfully is out of time to help right now. The neighbours colleage found some errors, but didn't find the solutions yet. It isn't in the sd card, he checked the soldering parts, no problems there. Personally i think it's a fault in the chip, but that is just a noob guess, and no idea how to check that. He will continue to check (in his free time), so i will post here as soon as i know more, with the errors that caused it all.
 
I've lost track of the finer details of your issues and what you tried etc... but if you suspect the ATMega is bad I would remove it from it's socket and reinstall it, then re-flash it and see what happens.
I would also point out that the the HM is a dual layer board, meaning there are traces/circuits on both sides of the board (not just the solder side) So "looking" at the solder joints on the solder side only gives you half the picture, there may be a solder joint on the component side of the board that is not complete because solder didn't flow to the other side of the board. That is why when I hear about odd situations like this the first thing I recommend is "REFLOW SOLDER" on the respective circuits. In this case that would be the ATMega and rPi sockets. IDK what your neighbor did, but if he is handy with a soldering iron I would ask him to reflow solder on the ATMega and rPi sockets, melt the solder that is there, add a little more, make sure you heat the trace on the board and wiggle the component leg around a bit with the soldering iron to encourage solder to flow to the other side of the board.
 
RalphTrimble said:
I've lost track of the finer details of your issues and what you tried etc... but if you suspect the ATMega is bad I would remove it from it's socket and reinstall it, then re-flash it and see what happens.
I would also point out that the the HM is a dual layer board, meaning there are traces/circuits on both sides of the board (not just the solder side) So "looking" at the solder joints on the solder side only gives you half the picture, there may be a solder joint on the component side of the board that is not complete because solder didn't flow to the other side of the board. That is why when I hear about odd situations like this the first thing I recommend is "REFLOW SOLDER" on the respective circuits. In this case that would be the ATMega and rPi sockets. IDK what your neighbor did, but if he is handy with a soldering iron I would ask him to reflow solder on the ATMega and rPi sockets, melt the solder that is there, add a little more, make sure you heat the trace on the board and wiggle the component leg around a bit with the soldering iron to encourage solder to flow to the other side of the board.
Click to expand...

I thought the through holes were tinned (plated) through, so if solder was on one side, it should have continuity to the other side? I agree that it's a good idea to get solder to flow all the way through, but if it's not, it should still have an electrical connection to any traces associated on the other side of the board. Am I wrong?
 
Last edited:
You would think, but I have seen personally on multiple occasions circuits not being made on the other side of the board. On one HM board I had a problem with a resistor in the blower circuit, tested continuity on the solder side and found it to be good and then the component side and found it to be lacking. I resoldered several times and ultimately had to resort to actually soldering the component side of the board and that fixed the problem. I think the process used by OSH Park is a rapid prototype process that is not the same as traditionally produced boards, perhaps that has something to do with it. Whatever the reason I just deal with the reality.
 
Cool, this is good to know. Going forward, I think I'll check continuity on both sides of the board before soldering.
 
It is possible that continuity is there before soldering, but perhaps the tin between layers is thin and a small amount of pressure on a component (like when holding the board) could break that tin if solder hasn't flowed through to the other side of the board. Then over time oxidation in the crack could make the connection deteriorate.
 
I sent a link of all this posts to my neighbour, he will give it to his colleage. Allthough this man is soldering for his profession, he might have some use to all of the answers. What Ralph says about checking both sides, looks to me that is something that could be "overlooked" at. So hopefully this gives him new ideas.
 
Ooooh noooo..............
I searched for days and days, as i had enough time because the neighbours colleage still had it for checking. Now i found a thread that got me a little crazy :
https://tvwbb.com/showthread.php?69166-New-User-needs-someone-to-hold-his-hand

It says that you should not install the 10K resistor if you have the TC version, which i have.

I had this colleage solder my HM when i bought it, but i think he didn't read all of the instructions very well (no offence to him, my bad!)
Now that i am looking at my first pictures in my first post, i think i see the 10k resistor installed, but it actually should not have....
I do not have the HM here yet (hope to have it back tomorrow so i can check it), but could that be the big problem?

If it is, is my HM really dead, or is it just a matter of unsoldering it?
I just hope it will fix my problem, allthough i ordered another one mr Mayland (means that i soon will have 2 to install :D )
 
having the 10k pullup resistor installed will not hurt anything, it'll just make the TC pit probe not work properly. Just remove it or even clip one leg off. This has nothing to do with your problem...
 
Although it's not a very usefull update, i'm just letting you know what's happening... the new HM arrived last week (customs made it take a while to get here). I gave it to the tech-guy. He will, somewhere next week, assemble this one and hopefully see what the difference is with the new one and the old, broken one. I was with him the other day while he looked over the components, but could not find anything. That is why i ordered the new HM. Comparing both to maybe spot the problem. Hope to have news by the end of next week.
 
Finally the solution is here...:
It took a while because the person who worked on my hm had to do this at his job, so it was laying on a shelf for a while. The problem was the Atmega chip itself! For some reason the file on the chip was damaged. When i was there with him (some weeks ago) he took another Atmega, but copied the damaged files from the old one to the new one. 2 problems, but not visible if you don't exactly know what should happen. Assuming that the file was in order, he kept on searching. In the mean while i ordered another hm, so he could build a new one and start comparing the 2. When that happened, he saw the fault of the Atmega and the files. I know that explaining this in my own words is maybe a little different if the "repairman" would, but i hope you all understand what happened. I'm just happy the hm works again, only i can never find out why the chip died.
 
I'd glad your friend got it working for you, but it seems he took the long road to accomplish it. I know it may be hard to follow what is actually going on between the HM and the rPi, as a result sometimes people make things more difficult than need be.

The HM software goes on the SD Card and loads on the rPI, the rPi communicates with the HM hardware and looks for an ATMega with the HM code on it. If it finds an ATMega that does not have the HM code on it the rPi will flash the HM code to the ATMega automatically. So, as long as the HM software on your rPi is not corrupt all you should have to do is install a blank ATMega in the HM and the rPi will program it, there is no need to copy the code from one ATMega to another. I have read a few posts from people having trouble with their HM saying "I have experience flashing ATMega's" and I did XYZ... There is no need for any of that (unless you are building a stand alone HM and don't have access to a rPi), just let the HeaterMeter Software on the rPi do it's job and deal with the ATMega and you will get up and running.
 
Last edited:
Thanks for this clear and understandable info! As you may have noticed, i am not the tech-guy here, but the way you explain it, makes sence to me! For my understanding : IF this problem occurs again, i can simply replace the Atmega with a new, blank one and the rPi will do the rest? Still thinking what the reason was for the Atmega to stopworking, but that probably will never be sorted out. I have no idea why he took the long road, but i am glad he finally found the error.
 
I've never had an ATMega fail, when I first starting messing with HM's I certainly did everything I could to kill one! LOL So I have no real explanation for your failure. I would think the easiest way to mess up the ATMega would be during flashing an upgrade, did you do that right before the failure? Otherwise I am at a loss as to what happened here.
With the above in mind, I wouldn't expect your new ATMega to fail. If you do have another failure you are correct, if you put a blank ATMega into the HM the the software on the rPi should see it and flash it automatically and you should be back in business....
 
Well i did upgrade it, but i can't exactly recall when. Might be the last thing i did before playing with it again.... for now i'm just a happy man with, in the end, 2 working heatermeters. I would say win-win!!
 
Alright, so here is one scenario... You were upgrading and the power was interrupted while the new code was flashing to the ATMega, perhaps it was still in progress but you didn't realize so you pulled the plug and put it away with the programing incomplete.
In that scenario you should be able to fix things up by manually initializing the flashing of the ATMega either via the HM web interface, if the HM web interface isn't accessible you can connect a USB keyboard and HDMI monitor the the rPi and type a command to manually initialize the flashing of the ATMega directly on the rPi.
 
Last edited:
It could be that this is what happened. Don't know, but for everyone who ever will get to this problem, it's a nice manual for repairing and understanding this issue. Thank you (all) for helping me with getting to the solution!!!
Hope my next post will be about brisket, pulled pork, beercan chicken, moink balls, ribs,.......as long as it isn't technical :cool:
If it is, i know where to ask :rolleyes:

Thanks al lot, all of you!!
 
You must log in or register to reply here.

 

Back
Top