Soft Power-Latch Circuit (#262) - Surface Mount Problems

[ERudisill]

I recently found vlog #262 and decided to try and implement it on a 3V battery powered radio tag.  However, it's not working.  As in, it neither turns on when the battery is inserted, nor when the button is pressed.  Since it is all SMD, I'm sure I've just selected some mismatched components.  But due to my lack of EE skills (software dev by trade,) I'm not seeing the problem.  Most of the components are 0402 because I'm size restricted.  That's also why I opted for a dual transistor IC rather than two independent components.  I've triple checked the footrprints and am fairly positive they are correct.

The schematic is attached below.  Is there something I'm missing?  Does this look rational?  Is there something I should be doing different for small SMD vs the pull-through components used in the video?

Thanks for ANY assistance, including internet berating as long as it points me in a helpful direction to learn

[katzohki]

Get out your DMM (go buy one if you haven't got) and start checking that voltages are where you expect them. Also, try powering your device directly to make sure it's not that.

0402 is just below the limit of what I could do without a microscope or some other magnification. They're so small, I would definitely take a close look for accidental bridges.

Did you make a board yourself or something? How have you got these parts put together?

[exapod]

i don't know your budget but you can use the MAX16054 http://www.maximintegrated.com/en/products/power/supervisors-voltage-monitors-sequencers/MAX16054.html .

[ERudisill]

Get out your DMM (go buy one if you haven't got) and start checking that voltages are where you expect them. Also, try powering your device directly to make sure it's not that.

0402 is just below the limit of what I could do without a microscope or some other magnification. They're so small, I would definitely take a close look for accidental bridges.

Did you make a board yourself or something? How have you got these parts put together?

Yes.  We had the board printed, and we've got a guy in the office that is quite good at soldering under a microscope.  There certainly does not seem to be any bridges, it all looks very clean.  The 0R resister was a CYA move that I put in, just in case the circuit doesn't work - and when we bridge the pads there, the device powers up.  So I'm fairly certain it is in my design.  Thanks!

i don't know your budget but you can use the MAX16054 http://www.maximintegrated.com/en/products/power/supervisors-voltage-monitors-sequencers/MAX16054.html .

Thanks! I forgot to mention in my OP that if anyone has an alternative to the circuit, I'd be willing to change designs.  At first glance, this looks really nice.  I'll dig into it a bit (price, size, etc.) and see if it will work for the application.