Hello,
Was wondering if somebody would be nice enough to glance over this schematic? It's a servo controller based on the 328P, but the second board I soldered according to it is releasing smoke (seems to be Q3 specifically, but I don't know if anything else is going with it), but I carefully checked the wiring and can't see a problem.
If any additional info is needed please let me know.
anvoice:
If any additional info is needed please let me know.
Like the, um, schematic?
That embarrassing moment. Thank you.
Edit: Link to malware site removed by moderator.
anvoice:
That embarrassing moment. Thank you.
It's a good thing emails don't cost a postage stamp, or we'd all be poor from the emails we've sent with no attachment.
The diagram is hard to read 
What's the driver chip part number?
Do you mean "servo" as RC servo (PPM), or as PWM?
Hi,
Do you have options when you export an image from your CAD, JPG with maximum pixels resolution will help us read you diagram better.
You said the second board failed, what happened with the first, is it working.
Are any components SMD?
You may have shorted track or pad.
Thanks.. Tom... 
anvoice:
That embarrassing moment. Thank you.
Are you also embarrassed from posting a link to a site containing malware?
Next time attach the image to your post.
Hi,
Edit: Link to malware site removed by moderator.
Hmm. what malware? I got saved it to drive and looked at it in graphics editor, no malware...
Tom....
Pop-up when trying to view the image.
And I was supposed to know that how? It's photobucket that I always used. I will try, had no idea you could do attach to post.
The half-bridge (I assume you meant that by driver chip) part number is IRF9389PBF.
This is a PWM-based servo. All components are SMD. I can't tell what happens to the second chip for sure because I unplug before everything burns up. But based on the first board, it looked better than the second chip. I hope it's not a shorted track, it's the second board to do this and the tracks should all have good spacing.
I've attached the schematic as pdf, hopefully that should be easier to read.
Servo_controller.pdf (67.7 KB)
Hi,
We will need an image of you PCB too, so we can check layout vs. schematic.
Thanks.. Tom..
Hi,
The first board that worked is the same PCB pattern as the second that failed?
Do you have a current limited power supply and an oscilloscope.
If you power up with 328 chip not installed, what happens?
Tom..
TomGeorge:
Hi,
We will need an image of you PCB too, so we can check layout vs. schematic.Thanks.. Tom..
Hi Tom, do you know how I could import that from Kicad? I only know how to get a gerber file out of there. I attached a zipped version just in case.
Both boards (the same) failed in the same way, so seems like something is connected wrong. I do have a power supply and a scope.
Not sure what you mean by 328P not installed: it's the first thing I usually solder on so I can't test that, at least with these 2 boards. I have 1 last one I could try with, would take a while to solder it together.
In the meantime I'm trying to put together a breadboard model of this schematic, which will also take a while.
Gerber.zip (33.5 KB)
Hi,
In the meantime I'm trying to put together a breadboard model of this schematic, which will also take a while.
Did you make a protoboard/breadboard version to start with???????
OR did you make a PCB and expect it to work?
Tom....
TomGeorge:
Hi,
Did you make a protoboard/breadboard version to start with???????
OR did you make a PCB and expect it to work?Tom....
I made a version of that one: it's hard to make an identical circuit with the same PCB components. E.g. my switching regulator is SMD, so are the signal FETs and the half-bridges. In fact I got breakout boards now just to test them.
Realize it's pretty old at this point, but my breadboard circuit officially works. So the problem is not with the wiring. In addition, the circuit board passed the Design Rules Check, which means it's connected exactly as the schematic, plus all tracks have good clearance.
Any other ideas as to what could be wrong and how to troubleshoot? Possibly a bad batch of boards since 2 behaved exactly the same. I'd go in there with a scope and voltmeter, but it burns if I give it power so that's not much of an option.