Rev 3: Hot MOSFET & low Voltage

More
11 months 1 week ago #10390 by Ty_Eeberfest
If the ideas I just posted don't solve the problem, give us some more pictures. Good close-ups of the low voltage + HV area - both sides of the board - would be helpful.

Look into it later when the dust is clearing off the crater.

Please Log in or Create an account to join the conversation.

More
11 months 1 week ago #10391 by rmcrae
Voltage coming out of the wall is good (12.4v).

Voltage coming out of the VIN and GND pins of header SV1 is 4.4v (I assume this is low?)

As far as the D3 (1N5819) and the D1 (UF4007) are concerned they are in the correct spots with the cathodes in the proper place. I went ahead and tested the voltage on each with the D3 putting out 2.6v and the D1 putting out 1.0v.

Please Log in or Create an account to join the conversation.

More
11 months 1 week ago #10392 by Ty_Eeberfest
Yes, the VIN pin I had you measure should read about 12V as it is connected directly to the center pin of the power input connector. You can confirm that by just looking at the traces on the board.

So one of two things is happening here: (1) Your power supply is broken, going way out of regulation under load, or (2) something on the clock board has gone wrong and is drawing so much current that your power supply can not cope and is going into an overcurrent protect mode to save itself.

Because the MOSFET gets hot I think (2) is most likely. Try pulling the ATMega328P controller out of its socket - this should eliminate any gate drive that may be getting applied to the MOSFET. Take VCC and VIN readings again and check MOSFET temperature. Does anything change??

Also please get us the pictures I requested in my post at the top of this page. Good close shots of the back of the board at as high a resolution as the site will allow you to upload could prove useful.

Look into it later when the dust is clearing off the crater.

Please Log in or Create an account to join the conversation.

More
11 months 1 week ago - 11 months 1 week ago #10393 by rmcrae
Did a couple of things:

Went ahead and tried a 9v power supply just to see--only got a read out of 0.4v on the GND to VIN.

Pulled out the ATMega328P Controller.

With the 12v power supply and the controller removed I'm still getting a draw of 4.3v on the GND to VIN and a draw of 2.6v on the GND to VCC.

MOSFET still very hot--no discernible change.
Attachments:
Last edit: 11 months 1 week ago by rmcrae.

Please Log in or Create an account to join the conversation.

More
11 months 1 week ago #10394 by Ty_Eeberfest
I think we can safely rule out a power supply problem then.

Thanks for the pictures. The back of the board looks good - I can't see anything that looks like a solder bridge - but a picture is never quite as good as being there. On the front side I'm wondering why your diodes appear to be laying on their sides rather than standing on end, tho this shouldn't matter unless a diode lead is touching something it shouldn't be.

Pulling the ATmega out should remove any gate drive from the MOSFET, so it shouldn't be conducting, but obviously it is since it's hot. The gate drive comes form Pin 15 on the ATMega. Try reading between Pin 15 of the empty socket and ground. Also read between the test point "HVD" (right next to the MOSFET) and ground. Both places should read essentially zero +/- a few stray millivolts.

If you read a significant voltage at those points then you need to find out how it's getting there, e.g. find where you got something shorted with a solder bridge or scrap of cut off lead or something.

But if you read 0 volts, all I can think of is bad MOSFET. If you are feeling lucky try to remove the MOSFET. Lucky because it's real easy to damage the board trying to de-solder it. If you do remove the MOSFET I'd expect to see VIN go up to 12V and VCC pop back up to 5V.

Look into it later when the dust is clearing off the crater.

Please Log in or Create an account to join the conversation.

More
11 months 1 week ago #10395 by Ty_Eeberfest
Also wondering what this stuff I highlighted on the picture is. Is it just a hair or a scratch in the soldermask, or is it something possibly conductive stuck to the board?


Look into it later when the dust is clearing off the crater.
Attachments:

Please Log in or Create an account to join the conversation.

Moderators: AccutronTy_EeberfestIan
Time to create page: 0.183 seconds

Search

Tube Suppliers

Go to top
JSN Boot template designed by JoomlaShine.com