3
\$\begingroup\$

I am assembling a board for a wireless charger based on this evaluation module (see here)

I own a preheat bed, a hot air gun and a quality soldering iron.

I am afraid that after soldering the chip and all its accessories, I have burned it.

I took the chance at verifying every input of the chip (using a microscope and oscilloscope) to determine if I have the correct values.

Here's a diagram: enter image description here

The red question marks are outputs that should show some voltage, but show 0.

The green checkmarks or x are valid inputs or the part is not needed.

I made sure all the inputs were correct, but I get nothing for output. I should see some PWM signal through PWM_A and PWM_B and also LED_A, LED_B, and LED_C

I also checked the stability of the 3.3V line, and it sits within 10mV enter image description here

I used the following technique to solder the chip:

  1. I applied solder paste to the chip
  2. Placed the board on top of the preheating bed and turned the device to heat up to 280 celsius
  3. I applied some flux and then I started using the hot air tool at 360 degrees for about 20 seconds
  4. When the chip looked stable on the board, I turned off the heat bed and waited for it to cool.
  5. Finally, I used a soldering iron at 370 (with a wide tip) to remove excess solder

Do you have any thoughts on this? Did I kill it?.

Later Edit:

Photos of the chip itself from three different angles so you have an idea:

enter image description here enter image description here enter image description here enter image description here enter image description here enter image description here enter image description here

To me it looks fairly well soldered. I took the pictures using my 1500x microscope (the same one I used to test the pins themselves with the oscilloscope)

Another argument here would be that I tried to "feel the resistance" of the lead with the tip of the oscilloscope probe and the visible lead tears formed between the pads and the chip pins were there.

Still, any idea?

Later Edit 2:

enter image description here

Photo of the chip in PCB design

\$\endgroup\$
8
  • 2
    \$\begingroup\$ upvote for well presented question \$\endgroup\$
    – jsotola
    Oct 9, 2020 at 16:32
  • 1
    \$\begingroup\$ Get a 10x eye loupe and look at the edges at an angle so you can really see. It's unlikely you "burned up" the chip though you could easily damage the board by lifting pads, etc. Your technique was a bit non-standard, typically you'd stencil past onto the board or do it all with wire solder and an iron and drag technique. Pictures would help a lot... \$\endgroup\$
    – Chris Stratton
    Oct 9, 2020 at 16:35
  • \$\begingroup\$ @ChrisStratton I will add pictures about this soon \$\endgroup\$
    – bem22
    Oct 9, 2020 at 16:52
  • 2
    \$\begingroup\$ Very unlikely that you damaged the chip with heat. Inspect those joints well, and use an iron and lots of flux on any ones that look iffy. Look for a meniscus shape for good joints, vs. bulging or non-wetting for bad joints. \$\endgroup\$
    – evildemonic
    Oct 9, 2020 at 22:10
  • 1
    \$\begingroup\$ @bem22 Because the PCB will burn up long before any typical IC would. \$\endgroup\$
    – evildemonic
    Oct 10, 2020 at 23:59

1 Answer 1

Reset to default
2
\$\begingroup\$
  • You applied solder paste by hand and soldered with hot air.
  • The chip has a large thermal pad underneath that is connected to ground.
  • You have pins reading 0 V when there should be some higher voltage present.

I think you've got short circuits between the thermal pad and some of the pins. The shorts will be under the chip where you can't see them.

Measure the resistance between your anomalous zero volt pins and ground. I think you will find things are shorted.

  • Remove the chip with your hot air station.
  • Install a new chip. I think you'll want to replace the old one because there's no telling what damage may have been caused by the short circuits.
  • Use less solder paste on the thermal pad than you did the first time.

Next time, measure the resistance between all pins and ground before you apply power to the circuit.

If you have time and resources enough (moola to have a new PCB made,) then you can make it easier to install the chip.

Place a large plated through hole under the chip. You want it as big as possible, but still smaller than the thermal pad.

Install the chip as usual from the top side any way you like (I'd solder it by hand with a small tipped iron, but I'm weird that way.) Do not put solder paste on the PCB pad for the thermal pad.

Once all the pins are soldered and there are no short circuits, flip the board over.

Put a drop of liquid flux in the plated through hole for the thermal pad.

Use a large tip on your soldering iron, and solder the thermal pad to the plated through hole. Keep feeding in solder as long as it'll go in.

The solder should wick its way in under the chip and make a good connection between the thermal pad and the PCB.

\$\endgroup\$
6
  • \$\begingroup\$ Nice one. I will do exactly this and come back with the results. You made me add another photo of what the pad underneath looks like. Thanks for your great suggestion. \$\endgroup\$
    – bem22
    Oct 10, 2020 at 15:29
  • 1
    \$\begingroup\$ Disagree. Especially given that it looks well aligned, bridging of a QFN thermal pad to a periphery pads seems unlikely. Sure, it's possible but fairly unlikely with a QFN - these are more in the habit of being subtly unsoldered in a way that is very hard to see. It is a little more possible with QFN styles that have metalization on the bottom and not just the side. Excessive solder on the central pad is more likely to result in chip floating making the peripheral pads not connect at all. \$\endgroup\$
    – Chris Stratton
    Oct 10, 2020 at 15:49
  • 1
    \$\begingroup\$ If one pushed the chip down in the melt, and held it while it cooled, then bridging might be a concern. Otherwise, unlikely. In terms of the large via, while that is a hand assembly technique where hot air is not available at all, it's not really needed - and it makes a board that will probably be deemed incompatible with later automated assembly. \$\endgroup\$
    – Chris Stratton
    Oct 10, 2020 at 15:50
  • \$\begingroup\$ @ChrisStratton , this sounds like something I might have actually done, especially when the air was too high, moving the chip around - I tried to move the chip back and probably pressed on it \$\endgroup\$
    – bem22
    Oct 10, 2020 at 16:04
  • 1
    \$\begingroup\$ Pressing on it would likely require that you actually held it down as you moved the heat and let it cool, otherwise it tends to just float right back up. That said next time around you could try shrinking the central pad (really those fillets to connections) just a tiny bit. Give the large number of vias (I'd use few more central ones) unless they are plugged it's relatively unlikely you have a puddle of excess solder on that pad, as the vias will wick it away. \$\endgroup\$
    – Chris Stratton
    Oct 10, 2020 at 16:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.