Forum Rule: Always post complete source code & details to reproduce any issue!
Results 1 to 14 of 14

Thread: Electronics question. Why this blow up. No teensys involved

  1. #1
    Senior Member
    Join Date
    Feb 2016
    Location
    Australia
    Posts
    209

    Electronics question. Why this blow up. No teensys involved

    Hello.
    I have been playing around with stroboscopic effect and though I would make myself a big strobe light.
    I ordered a few bits and thought this would work:
    Click image for larger version. 

Name:	mosCirc.PNG 
Views:	37 
Size:	105.3 KB 
ID:	15308

    This turns on the led panel fine if I pull the signal wirre high to 12Vdc, and turns off if I disconnect the wire.

    I then hooked up a cheap arduino nano and an adafruit clone 16ch pwm board running at 1.7khz which slowly sweeps duty cycle from 0-100%.
    problem is my lamp lit up for a second, then the pwm board let out some smoke.

    Nano is no longer detected by windows, and pwm board does not work.
    I have spares hooked up and ready for round 2 but I'm not really sure what went wrong.

    I should say I'm not great at electronics so probably something simple that I don't understand.

    Any Ideas let me know.
    Thanks.

    Click image for larger version. 

Name:	7C946E54-7114-4658-9B21-AA661EB59C49.jpg 
Views:	22 
Size:	134.8 KB 
ID:	15311

    12.6vdc supply powers led driver circuit above and also goes to arduino nano vin and gnd pins.
    Single wire from PWM signal was connected to my mosfet driver trigger. Wire length was about 20cm I suppose if that is relevant.

    Click image for larger version. 

Name:	m1.PNG 
Views:	13 
Size:	266.7 KB 
ID:	15312Click image for larger version. 

Name:	m2.PNG 
Views:	5 
Size:	255.8 KB 
ID:	15313
    Last edited by Gibbedy; 12-07-2018 at 03:50 PM.

  2. #2
    Senior Member
    Join Date
    Feb 2016
    Location
    Australia
    Posts
    209
    Killed another nano and pwm pca9685 board.
    My nanos have regulators marked 1117-5 they are going up in smoke.

    I used a pololu 5v 500ma buck converter board and it is working so far.

    Thats what I'll go with at this point.
    Note last nano that failed I only had nano and pwm board connected.

    Vcc and V+ are powered from arduino regulator. I believe it should consume very little power as I'm not driving anything but a mosfet driver but perhaps there is some other consideration. Here's my attempt to show what I did here. Click image for larger version. 

Name:	I2C NANO BOARD_bb.jpg 
Views:	17 
Size:	87.4 KB 
ID:	15314 incoming black gnd should have gone to gnd rail not as shown in pic.

  3. #3
    Senior Member
    Join Date
    Apr 2017
    Posts
    136
    What PWM frequency are you driving the mosfet (Arduino through U1) with 1.7kHz, how are you getting that? I see no timer divisor capable of getting that frequency. Maybe add a current limiter between arduino and input on U1--maybe a 100 ohm will do.

  4. #4
    Senior Member
    Join Date
    Feb 2016
    Location
    Australia
    Posts
    209
    Iím using an Adafruit (clone) pca9685 breakout board just because I have it to add a buffer between an expensive teensy and the mosfet driver which I donít fully understand. Iím using a nano here to test things out and glad I did.

    I was thinking of adding a resistor as you say but while I was setting up scope on nano with only pen board connected I smoked it so Iím doing something wrong with Pam board or my nanos have fake 1117-5 regulators.

  5. #5
    If you only connect the nano and the pca9685 together does it cause failure?

    According to your diagrams the nano is receiving 12V input to the Vin pin? - this is fine
    5v output from nano is being used to power VCC and V+ of the PCA9685? - this is fine
    The PCA9685 pin is to control your mosfet circuit which is switching a 12v power line? - i would disconnect this and check that your nano-pca connections are working without any damage. If this works as expected then I'd assume there may be something wrong with your mosfet setup that is either drawing too much current or puting the 12v back into the pca9685.

  6. #6
    Senior Member
    Join Date
    Feb 2016
    Location
    Australia
    Posts
    209
    Hello JWScotSat.
    Yes. On last attempt only PCA9685 and arduino nano connected when the last pair blew up. It ran for a little bit before smoking.
    Nano 5v reg is failing passing full 12v which kills everything.
    So It looks like the regulators on the nano's i'm using are not up to spec although I use them all over the place. One sitting on my doorbell is at 15VDC all day. I just don't understand most of this.
    My solution was to use a pololu 5v 500mA buck converter which is running everything fine.

    What exactly do part numbers refer to.. I mean does someone invent a 5v regulator and name it 1117-5 just because they feel like it.. then does everyone decide to make the same thing and just give it the same name if it meets the same specs?

    I believe the tc427 mosfet driver should protect me from the 12V.

  7. #7
    I'm concerned that on its own the pca9685 is blowing your nanos regulator without anything else connected as I'm looking to use the PCA9685 in my own projects.

    According to the datasheet for the PCA9685
    IDD supply current operating mode; no load; fSCL = 1 MHz; VDD = 2.3 V to 5.5 V - 6 to 10 mA

    In a stand by scenario the PCA9685 should only pull 10mA of load. Can you again confirm that you have ONLY the nano and the PCA9685 connected together. Nothing else connected to those devices? Can you provide pictures and a full schematic of this setup and detail how it fails?

  8. #8
    Also, I assume this was a placement error and you weren't placing 5v and gnd on the same line? Click image for larger version. 

Name:	Untitled.jpg 
Views:	13 
Size:	45.4 KB 
ID:	15335

  9. #9
    Senior Member+ Frank B's Avatar
    Join Date
    Apr 2014
    Location
    Germany NRW
    Posts
    4,765
    That's the reason why, in cases like this, photos are way better..

  10. #10
    Senior Member
    Join Date
    Feb 2016
    Location
    Australia
    Posts
    209
    Yeah I edited the note but didnít re-do the image

  11. #11
    Senior Member
    Join Date
    Feb 2016
    Location
    Australia
    Posts
    209
    I should have re-connected to take a photo. It is all on pcb and glued in position with just a socket for nano and owm board.


    Interestingly last night I added some rotary encoders and used pwm pins as an easy point to steal 5v and gnd.


    After this the pwm board stopped working on startup. Power light would flash and light Dim. (Maybe I bumped something)

    I got a wire and bridged gnd terminal on screw connector to gnd of USB socket on nano and it lit up bright.

    Put a wire In place to fix that.
    So perhaps I have a loose gnd track or wire underneath my board . I really need to pull the pwm board out Check it out.

  12. #12
    Senior Member
    Join Date
    Feb 2016
    Location
    Australia
    Posts
    209
    JW I have no pics of the way it was wired for test sorry.
    I only have my own recollection which I wouldn’t expect you to rely on or me really..

    Last night I was a bit more observant but the setup now has another layer of complication that won’t be helpful.

    I will say what I found with nan/pwm board/mosfet driver/pled display for whatever reason...
    Measuring from gnd screw terminal (for pwm output power) to gnd pwm output pins gave high resistance.

    Current on supply showed pulling 500mA.

    Putting meter to current measurement and using as a test wire I put one probe on same gnd screw terminal where I had done continuity test and other terminal on USB socket of nano. Pwm board led instantly lit up. And complete system functioned. Adding a wire fixed this latest issue.

    Now did I blow a gnd track.. dunno.
    I’m pulling .1A now.

    This whole post probably isn’t much use for anything.

    With my original problem I haven’t ruled out dodgy nano’a because yesterday one blew up connected to nothing on 12.6vdc. But as soon as it happened I wasn’t sure if board slipped and shorted on bench or not.

    My supply looks good on the scope.
    Last edited by Gibbedy; 12-12-2018 at 06:15 AM.

  13. #13
    So it seems you've had grounding issues with the boards. I couldn't understand why the PCA would blow your arduino. It seems your setup is far more complicated than your initial post hinted to.

  14. #14
    Senior Member
    Join Date
    Feb 2016
    Location
    Australia
    Posts
    209
    The pca and nano that blew up was connected by pins/sockets/circuit board with solder tracks I make.. (what u call that) so I’m guessing gnd connection on solder pad or socket or something was crook. I suppose the theory is that a dodgy ground causes something on pca board to fail in a high current pulling way?

    It’s all silicones into a box now and will be put on the shelf as another rediculous project. Can’t post photos from iOS for some reason.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •