New HM 4.3 Build HM Not Communicating


 

DaleH

New member
Hi All,

I recently completed a HM 4.3 kit build that my wife got me for Christmas 2017 (an indication of my available free time) :p Unfortunately I'm unable to get the HM to do pretty much anything at all. I've combed through the forums at length and have tried a few things, but without any success. Here's what I have, the symptoms and what I've tried:


HeaterMeter kit for a 4.3 build with a thermocouple option
Raspberry Pi Model 3B 1GB revision 2
v14 of the HM Firmware

I've assembled and connected all the components and loaded the firmware.
I can boot the device and connect via ethernet or wireless connections
The HM shows up in the devices page
I can log in via root to change configuration in the web interface

However I have these issues
- The linkmeter tab provides the error reported by others in the forum
HeaterMeter serial communication can not be established. Configuration requires bidirectional serial operation. Possible causes of failure:
No HeaterMeter board attached
No HeaterMeter (AVR) firmware installed. See AVR Firmware
Incorrect baud rate in /etc/config/lucid
Hardware malfunction

- I've tried updating the firmware through the interface as well as the command line and AVRUpated. In both cases I receive the AVR Fuses error reported by others and cannot proceed further.

- I've confirmed the ATMEGA chip is installed correctly

- I've reheated all the solder joints and cleaned up some that weren't fully filled.

- I've cleaned the boards with denatured alcohol the best I could.

- I've confirmed the LCD adjustment setting


After each of these steps the results were the same:

- The HM registers on the devices page

- I can log in to the web interface and update settings via wired or wireless connection

- Other than the the LCD showing black boxes while adjusting to confirm the correct setting there are no other signs of life from the HM; no LED lights at any time.


At this point I don't know where to go next. I know I need to test continuity and voltages, but I'm not sure where to start. I have a multimeter, but the use of it on circuit boards is far from my weakest strong point. I did manage to confirm the ground connections, 5v (4.9x) supply to the button board and 12v out of the power supply connector, but that's about it. I'm not sure about the 3.3v points or what the items highlighted in green are meant to represent in the diagrams on the basic troubleshooting link.

I think that's about it. Any suggestions from anyone would be more than appreciated. Thanks!
 
Make sure you have not mixed up the 3.3v regulator with a BS170 transistor on your HM board. They look just alike except for the printing on them. Without the regulator in place your ATMega will be DOA......
 
Thanks for the suggestion Ralph! I saw that note in the build instructions and took great pains to make sure I didn’t mix them up. I did just double-check them to be sure and fortunately/unfortunately I have them in the right locations. HOWEVER, as I was inspecting the board I noticed that while I have the ATMega inserted the socket in the right orientation, I have the socket soldered to the board...backward. Unbelievable. I’ll switch it around and retest when I have a little more time and hope it works. If not I had already ordered another ATMega in the event I accidentally damaged it during the build or resoldering processes.

I’ll post back my results as soon as I can. Thanks again!
 
Success!

Thanks again Ralph for the suggestion! Had you not suggested I take a look at the regulator and transistors, I never would have noticed I had the ATMega socket soldered in the wrong orientation. Fortunately just flipping the ATMega in the socket was all that was needed.
 
Aww yeah! Great work on that. Yeah the orientation of the socket itself doesn't matter as long a the ATmega is in the orientation specified on the PCB silkscreen (with the notch closest to the board edge). Nice job on the rest of the soldering and getting up and running after just a brief 1 year delay. :-D
 
As Bryan noted, it really doesnt matter what orientation the chip socket is installed, as long as you install the chip in the orientation that is shown by the printing on the board. (the chip can go in the socket backward, no problem). Perhaps it was just some bad soldering on the socket that got cleaned up when you switched it around. I echo the congrats on removing the socket and putting it back in without damaging any traces on the board, nice job!
Hope you enjoy your HM as much as the rest of us!
 
Thanks Bryan and Ralph for the comments! I AM excited to have this working and have been having fun monitoring the temp out of our upstairs furnace vent with the pit probe since I can’t get it hooked up to the smoker just yet.

I did run into an issue with my last step that I didn’t expect considering how well everything worked once I put the ATMega chip in the correct orientation. Now I’ve noticed that my blower will not work. Blower works fine when connected directly to power, but there doesn’t seem to be power getting to the blower pin at the RJ45 outlet.

Any suggestions on the best way to trace that back to where there’s a break in the circuit?

Thanks for everything!
 
There is an image with the voltages on it here to help debug such issues (which assumes blower is running 100%):


In particular you'll want to follow from BLOW on the RJ45 -> 47uF/25V capacitor -> inductor -> the three pins on the power MOSFET -> Q1 BS170 driving the MOSFET -> BLW pin on the ATmega. Your problem is somewhere along that line.
 
Thank You Brian! It may be a week or more before I get to it, but I'll pull out the multi-meter and start prodding around as soon as I get a chance.
 
Hi Brian,

I finally got around to checking the follow circuit as you described above. I have continuity in all the sections I would expect to have continuity, but when I power the HM (without the button board), the expected 10.5V out of Q1 BS170 is actually showing 12V. That transfers up to the MOSFET so I'm showing 12V on the outermost pins of the MOSFET and ~176mV reading on the center pin of the MOSFET which should be reading 12V. While I'm pretty sure this is wrong, I'm not quite sure where to go from here. Any suggestions you might have would be appreciated!

Thanks!
 
That's exactly what you'd get if the MOSFET is fully off, like 0% output. Was the output set to 100%? If so then it means Q1 isn't turning on, because that is what turns on the blower MOSFET. In that case check the 0+ leg closest to the front of the board and verify it has the same voltage as BLW on the microcontroller. If it is 3.3V, then that means the HeaterMeter is trying to turn it on but it isn't, which is usually an indicator of a broken leg on Q1.

If Q1's close pin is just 0V with no variation then it sounds like the blower output is 0% for some reason, perhaps in the webui the blower has some configuration that is preventing it from running? Not sure at that point.
 
Hi Brian,

I’ve confirmed the blower output is 100% in the webui and have experimented by changing the set point to also force it to 0%. When the fan is at 100% the 0+ pin goes to 3.3v as expected and when the fan drops to to 0% the 0+ pin of Q1 drops to 2.9mV. The closet pin of Q1 does vary, but only within a few mV (.1mV at 0% and .7mV at 100% fan)

During all of these changes, the center pin of the MOSFET never changes much and hovers around 88 to 91mV with the Blow pin at the RJ jack steady at 206mV.

With the MOSFET not really changing the center pin output is it possible I’ve damaged the MOSFET somehow and should consider replacing it?

Thanks for everything!
Dale
 
Maybe these notes will help you. There are more notes on my v4.2 image so I am including it, the circuit is the same as v4.3 but the layout is different.

If the center pin of Q1 is going from 0/3.3V when blower is off/full but the trace that connects Q1 to Q3 isn't going from 12/0V you probably have a bad Q1. Also verify Q1 is a BS170...

HM_Blower_Notes_V4.3.jpg


HM_Blower_Notes.jpg
 
Thanks for the information Ralph! That pretty much sews it up I think. The leg from Q1 to Q3 never varies from 12v whether the fan is full on, full off or anywhere in between. Apparently I’ve damaged Q1 somehow as it appears the solder connections look good and I don’t see any broken leads. I’ll order another BS170 or two and replace it to see if that solves the issue. I’ll post back as soon as I get the part and can get it swapped out. (I also confirmed I didn’t have it switched with the MCP1700.

Thanks again!
Dale
 
Brian and Ralph - apologies for the delayed update regarding your suggestions. Been a little busy on this end, but finally had the opportunity to swap out Q1 and, as expected, that solved the problem! Fan turns on and output varies as it should. Thanks so much for all the great advice and the time you guys spend on this project and the forum. Now to find some time to actually use the HM in practice!

Thanks again for everything!
Dale
 
Glad you got it working. Those little transistors can be overheated during soldering, you're not the first one to take one out....
 

 

Back
Top