Warning!!! A fault on version 4.1 board Github files bottom ground plane


 
J

John Bostwick

TVWBB Wizard
if you look at this thread

http://tvwbb.com/showthread.php?51184-HM4-1-fail-short-somewhere

Craig used the version 4.1 GIthub files and had a short. He narrowed it down to the bottom ground plane and the DC/DC converter pad

2j1swvl.png


The red circle shows the pad in question and if you have ordered a new board within the last couple of weeks, since the last upload, you may have to cut that pad from the ground plane, so that you will have a functioning board.
 
It appears that the bottom layer silk screen might be the culprit.

These are from v4.1.3
l30JPsp.png

jlemUuF.png


This is from v4.1.2
Q4biM85l.png
 
Last edited:
I wouldn't cancel if I were you, it's rather simple to cut that connection with a razor or something before you build the board, then you are good to go...
 
if you look closely at it you can see the tops of the pads mis-aligned up, its like it was copied and then not pasted back to the correct spot
 
RalphTrimble said:
I guess the most important question is... Is this board functioning properly/fully after this trace is cut?
Click to expand...

Yup. Mine is working just fine now that I cut the trace. I had to remove the OKI power supply daughter-card to do so, but the fault was apparent on the two spare boards. It's kind of a good thing that I had these spare cards to isolate this problem. I have yet to assemble my blower/servo controller, so I cannot comment on those functions yet. However, the board boots up properly using either the RPi's power supply (microUSB) or from the 12V supply plugged into the HM4.1. The LCD comes on and functions properly, as does the web server. I kept the unit plugged in for a bit of a "burn-in" period, and all seems to be functioning properly. Once I get my blower configured, I'll post on how it is working. Thanks for posting the warning!
 
Steve_M said:
It appears that the bottom layer silk screen might be the culprit.

These are from v4.1.3
l30JPsp.png

jlemUuF.png


This is from v4.1.2
Q4biM85l.png
Click to expand...

Wow, amazing you found this Steve! I know nothing about the whole process of PCB printing, so it is good to have folks on here with the ability to find this stuff. I'm totally impressed by how cool all of this is!
 
CraigMW said:
Yup. Mine is working just fine now that I cut the trace. I had to remove the OKI power supply daughter-card to do so, but the fault was apparent on the two spare boards. It's kind of a good thing that I had these spare cards to isolate this problem. I have yet to assemble my blower/servo controller, so I cannot comment on those functions yet. However, the board boots up properly using either the RPi's power supply (microUSB) or from the 12V supply plugged into the HM4.1. The LCD comes on and functions properly, as does the web server. I kept the unit plugged in for a bit of a "burn-in" period, and all seems to be functioning properly. Once I get my blower configured, I'll post on how it is working. Thanks for posting the warning!
Click to expand...

What trace exactly do I have to cut.
 
FrankH said:
What trace exactly do I have to cut.
Click to expand...

The bottom pad which is shown in Steve's pic of the 4.1.3 board on the right. The way I did this was with an X-acto razor blade, digging a channel between the solder pad and the ground plane. The purpose is to electrically isolate that pad from ground.
 
Whoa holy smokes, looking at the source it's definitely messed up. Putting up new source files now.

EDIT: New board file is on capnbry.net, or OSH Park, or github. Note these say "HeaterMeter v4.1.4" on the silkscreen rather than just v4.1.
 
Last edited:
Thanks, Bryan. After cutting that trace to isolate the bottom solder pad, all is working fine now. Great work on this! I can't wait to try it out on my next cook for Mother's day (brisket and pork shoulder).
 
There are so many things that can go wrong with a "simple" fix. I go through this at work with management when they want to do a "quick spin" and just "tweak" one thing. They figure it is so small there is no risk. No such thing.

dave
 
Bryan Mayland said:
Whoa holy smokes, looking at the source it's definitely messed up. Putting up new source files now.

EDIT: New board file is on capnbry.net, or OSH Park, or github. Note these say "HeaterMeter v4.1.4" on the silkscreen rather than just v4.1.
Click to expand...

Yah, this one was a lot bigger problem than the LED1 issue, glad you got it fixed up, wonder how many boards were printed with that error? At least it's an easy fix if someone has one....
 
Looks like I will have 3 with the error and 3 without, I thought I canceled my order, I just got word that both sets have been sent to fab. Not a big deal, it's a simple fix, just wanted the cleanest board to start off with.
 
Well, damn... I think once you see how simple it is to cut that trace, and how it doesn't effect anything else you will understand why I suggested that you just go with it instead of re-ordering with the fix...
I guess at some point you are gonna be the source to get your HMv4.1 board for a few forum members....
 
haha Ralph, they did finally get back with me after the boards had gone to fab and canceled it. So, i'm only getting the 3, I would have populated and sold the boards for others if they didnt want to deal with cutting the trace. I have extensive electronics background, but mechanical, meh... I got nothing.
 
You must log in or register to reply here.

 

Back
Top