HM4.1 fail... short somewhere?


 
John - Thanks for validating this. I thought I'd done something wrong on OSH Park. It's a good thing I had a couple of spare boards lying around to check this on those because that extra trace to the ground plane was not visible with the OKI-78SR-5 mounted above it. I knew there was something wrong when I found such low resistance between 12V+ and ground at the power connector and then just continued to track through the board from there. The schematics were quite helpful.

BTW, this thing just looks so cool! Amazingly, that was really the only issue that arose. Everything else seems to be working fine so far now.
 
Last edited:
CraigMW said:
John - Thanks for validating this. I thought I'd done something wrong on OSH Park. It's a good thing I had a couple of spare boards lying around to check this on those because that extra trace to the ground plane was not visible with the OKI-78SR-5 mounted above it. I knew there was something wrong when I found such low resistance between 12V+ and ground at the power connector and then just continued to track through the board from there. The schematics were quite helpful.

BTW, this thing just looks so cool! Amazingly, that was really the only issue that arose. Everything else seems to be working fine so far now.
Click to expand...

yeah, I think I even know now, what happen.

Bryan Mayland said:
Board file is updated with the LED1 trace fixed. Pushed to github, shared on OSHPark and updated capnbry.net's HeaterMeter PCB.

I also nudged the 5V regulator up 0.01" to allow the ground plane trace to reach it directly. This may alleviate the need for the "white wire" ground fix.
Click to expand...

I bet when he did that something fouled up because it looks part of the pad was copied and moved about the .01 he moved the regulator
 
That's exactly what happened. I moved the other two traces but didn't see the ground one because it was part of the ground pour. Theoretically it shouldn't have shorted, it should have just come really really really close, but it is closer than the 6mil minimum clearance of the PCB fab.

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:
Bryan Mayland said:
That's exactly what happened. I moved the other two traces but didn't see the ground one because it was part of the ground pour. Theoretically it shouldn't have shorted, it should have just come really really really close, but it is closer than the 6mil minimum clearance of the PCB fab.

New board file is on capnbry.net, or OSH Park, or github.
Click to expand...

Awesome, Bryan. Thanks for the quick update!
 
I just put together a Cat5e cable and connected wires 4&5 to black and red on the blower. Everything works fine. It's really cool watching the PID function as I set the pit temp to 94, and then touch the end of the pit probe to warm it up. The blower slows down when the pit probe nears body temp, and then if I let it go, the blower speeds up. I still need to test the servo connection, but anticipate no issues.
 
OK I ordered my boards April 21st and just received them today. I have not had a chance to check them out. Would they be ok? Or does someone mind posting some better images/directions for us non technical folks to see if they have the same issues. Thanks
 
In the picture below, at the bottom center of the pic see the three oval pads? At the top of the pad on the right see how it is not isolated on the top? You need to cut that connection there so that pad is isolated at the top, only leaving the trace that shoots off to the right connected.
otdenk.jpg


Below is an actual picture of the board, note the same location, the connection may be hidden by the white silk screen printing...
317gvw3.jpg
 
I got my 4.1.3 boards yesterday - all three are showing continuity between the 1st and 2nd pins. Before I begin surgery, is the culprit a connection under the white silkscreen? (circled on the right on Ralph's picture)
 
I circled the culprit on both the real pic and the graphic above, and yes, the connection could be covered by the silk screen cause it's right in that area. It's just a little bit of a connection on the top of the solder pad on the right...
 
Yeah it probably just barely touches the pad. I'm a little disappointed because the reason I moved the component up 0.01" (and created this issue) was that the ground pour was supposed to now fill all the way across under IC1. I guess it was too close to the dimension? (sad face)
 
G.Godoy said:
I got my 4.1.3 boards yesterday - all three are showing continuity between the 1st and 2nd pins. Before I begin surgery, is the culprit a connection under the white silkscreen? (circled on the right on Ralph's picture)
Click to expand...

OK. just to be absolutely clear here I made another pic. In the picture below I have scribbled across the ground plain, the solder joint on the right of the OKI should NOT touch the ground plain. I drew an arc in the area between the right solder pad and the ground plain, this is where you need to cut the connection with a razor. I also put an arrow to the bottom pad of D2, notice how this connects to the ground plain in four places? When you are done cutting the trace on the OKI it should no longer have continuity with the bottom solder pad on D2

HMGND.jpg
 
You must log in or register to reply here.

 

Back
Top