Feather turning off when NeoMatrix plugged in to 5V power

EL Wire/Tape/Panels, LEDs, pixels and strips, LCDs and TFTs, etc products from Adafruit

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
User avatar
dlinsalata
 
Posts: 33
Joined: Sat Jul 22, 2017 5:55 pm

Feather turning off when NeoMatrix plugged in to 5V power

Post by dlinsalata »

Hey all -

I am so close to finishing my first big Neopixel project, but I'm running into a problem I can't figure out: I have an Adafruit Feather nRF52840 Express connected to an 8x8 NeoPixel NeoMatrix. Per advice from this forum, I'm driving them both off a 5V power adapter.

Everything worked on the breadboard, and everything worked when I soldered the pieces into a PCB. The only thing left was to remove the alligator clipped connection between the DIN on the NeoMatrix and the Feather and solder the wire from the NeoMatrix to the resistor.

However, as soon as I soldered that (and just that), the Feather turns off each time I plug in the NeoMatrix. I figured something was shorting, but I can't see where that would be. (I know my soldering is absolute beginner, but the only change was literally one wire to another.)

Does anyone have suggestions on what I should check? Photos attached, including a close up of the signal wire I soldered this morning.

Thank you so much!
Attachments
2022-10-28 12.52.15.jpg
2022-10-28 12.52.15.jpg (373.77 KiB) Viewed 119 times
2022-10-28 12.51.55.jpg
2022-10-28 12.51.55.jpg (250.9 KiB) Viewed 119 times
2022-10-28 12.52.36.jpg
2022-10-28 12.52.36.jpg (240.44 KiB) Viewed 119 times

User avatar
adafruit_support_mike
 
Posts: 67391
Joined: Thu Feb 11, 2010 2:51 pm

Re: Feather turning off when NeoMatrix plugged in to 5V power

Post by adafruit_support_mike »

I see several pads that haven't been covered with solder, and those could easily cause trouble.

Most of the joints look clean and shiny, which is good.. it means you haven't kept the solder hot for too long. The surface of molten solder oxidized in contact with air, and the oxide breaks down into a powder that makes the solder pasty. You've avoided that, so bravo.

Give each solder joint a dab of flux and reheat it until the solder covers the whole pad and sweeps smoothly up to the pin. The rosin in the flux is made of acids that do very little at room temperature, but become aggressive evough to dissolve thin layers of copper and tin oxide on the PCB pads at soldering temperature. That gives the solder a clean and 'active' surface for bonding (lots of metal atoms with free electrons that can form bonds).

The flux also makes a barrier between the molten solder and the air, preventing oxidation. Flux burns off over time though, but a dab of paste flux gives you a few extra seconds to make a clean joint.

User avatar
dlinsalata
 
Posts: 33
Joined: Sat Jul 22, 2017 5:55 pm

Re: Feather turning off when NeoMatrix plugged in to 5V power

Post by dlinsalata »

Thank you for looking and the advice! I'll go do that and report back.

Out of curiosity, why does not having the pad potentially cause this problem?

(On a side note, of course as soon as I posted this, everything started working...but don't worry: as soon as I moved the project, it stopped again :P )

User avatar
adafruit_support_mike
 
Posts: 67391
Joined: Thu Feb 11, 2010 2:51 pm

Re: Feather turning off when NeoMatrix plugged in to 5V power

Post by adafruit_support_mike »

The last line answers the second one. ;-)

If the circuit starts and stops working as you move it, it suggests marginal connections opening and closing. Joints that don't cover the pads are suspect because it's hard to be sure they've made good contact with any of the other metal the joint is supposed to connect. Well-formed joints have all the externally visible clues that suggest reliable mechanical and electrical connections.

User avatar
adafruit_support_bill
 
Posts: 88040
Joined: Sat Feb 07, 2009 10:11 am

Re: Feather turning off when NeoMatrix plugged in to 5V power

Post by adafruit_support_bill »

Mike wrote:Well-formed joints have all the externally visible clues that suggest reliable mechanical and electrical connections.
Which is why circuits for critical applications (medical instruments, avionics, military etc.) usually undergo visual inspection in addition to electrical tests. A circuit may function perfectly off the assembly line in a test jig. But a close look by a technician may reveal signs of joints prone to failure.

User avatar
dlinsalata
 
Posts: 33
Joined: Sat Jul 22, 2017 5:55 pm

Re: Feather turning off when NeoMatrix plugged in to 5V power

Post by dlinsalata »

Thank you, both! That makes sense.

I followed your advice, and I'm a lot happier with the job now. Every pad is covered, and it helped me get better at the teardrop shape of the solder as I went.

Then I tried testing it, and the RTC wasn't working. So I double-checked everything with a head-mounted magnifying glass, used some rubbing alcohol to get the excess solder off, and then used a toothpick to carefully get even more of the excess solder off.

Well...then I plugged it in, and it...erm...caught on fire. One of the solder points, which was either the ground or the space between the ground and power. (Circled in photo.) Does this mean I've probably soldered them together, even though I don't see anything? Or could the flux itself be burning?

Thanks for the help!
1-2.png
1-2.png (489.05 KiB) Viewed 76 times
1-3.png
1-3.png (525.82 KiB) Viewed 76 times

User avatar
adafruit_support_mike
 
Posts: 67391
Joined: Thu Feb 11, 2010 2:51 pm

Re: Feather turning off when NeoMatrix plugged in to 5V power

Post by adafruit_support_mike »

dlinsalata wrote: Mon Nov 07, 2022 9:06 pm Well...then I plugged it in, and it...erm...caught on fire.
That can happen. It's common enough to be called 'letting the macgic smoke out' (components don't work after the smoke is gone), and we've all done it more times than we'd like to admit.

It takes a fair amount of current to generate the heat necessary to make smoke, and that usually means a short circuit. That can either be an unexpected connection or a wiring error.

Wiring errors are easy to make when you work on both sides of a board. It's easy to forget to flip something. Since it is so easy to make mistakes, it's always a good idea to draw front and back wiring diagrams so you can check as you go.

It usually takes a couple of tries to get a wiring diagram with all the connections and everything in the right place, so I do a few quick sketches in pencil until I find something I like. Then I make a copy of it using markers on typing paper, so it's easy to see the lines when I flip the page over.

The usual color-coding rules apply:

- Use red for the most common positive voltage, orange for the next most common, and yellow after that (very few systems have more than three positive voltage rails, and if they do, you need a more formal line marking scheme).

- Use black for 0V/GND, and green for anything with a connection to Earth Ground.

- If you have negative voltages, use blue for the most common, then purple, then grey.

- Use other colors for non-supply signals. Use the same color for the whole node (any two points that should have zero resistance between them), and try to use different colors for nodes that run close to each other so you can tell them apart. Don't be afraid to use text labels, and use them anywhere it helps reduce confusion.

By the time you've done all that, you'll be surprised how much information there is even in the diagrams for simple circuits.

When your diagram is done, flip it over and make a copy from the back side. I generally draw then on 3x5 cards (or larger) so they lay flat on the bench while I'm working.

Locked
Please be positive and constructive with your questions and comments.

Return to “Glowy things (LCD, LED, TFT, EL) purchased at Adafruit”