jay123035
Posts: 32
Joined: Wed Nov 21, 2012 5:24 am

lm2576-5 circuit failing with 13v output

Sat May 03, 2014 1:55 am

I have a project that uses the basic "sample" power circuit for the LM2576-5 (found in issue 6 of MagPi)

Image

but all of a sudden the output jumped to 13v instead of 5v. (fried the Raspberry pi it was powering)

I am wondering what let the smoke out.
the caps?
the choke?
the IC?
the Zener diode?

My first guess was the IC but I cannot think of a way to test it without rebuilding the complete circuit. and that seams like a lot of work for a test.


Any ideas or suggestions would be really appreciated.

User avatar
Burngate
Posts: 6289
Joined: Thu Sep 29, 2011 4:34 pm
Location: Berkshire UK Tralfamadore
Contact: Website

Re: lm2576-5 circuit failing with 13v output

Sat May 03, 2014 11:37 am

jay123035 wrote:I am wondering what let the smoke out.
the caps?
the choke?
the IC?
the Zener diode?
First off, do any of those look like they suffered?

How could a cap go faulty? Either short or open circuit.
If either cap went short, the output would go to zero, so it wasn't that.
If Cin went open, it's possible for the chip to oscillate, going from 0 to 13v and back rapidly. Probably not sit at 13v permenantly.
If Cout went open, the result would depend on what's being fed - the Pi's C6 may be enough to smooth the o/p, or maybe not.
If not, the feedback would immediately tell the chip the o/p was too high, and switch off, so (probably) you'd get spikes of 13v separated by 0v, which would be smoothed out by the Pi's caps to less than 5v

How could the choke go faulty? Again, either short or open circuit.
If it went open circuit, you'd get no output at all.
If it went short, the result would be somewhat similar to Cout going open, but the chip wouldn't like the large current into Cout, and would either shut down or fail.

The zener?
It isn't a zener, for a start, it's a schottky diode - different thing.
If it went short, you'd get 0v (and a blown chip)
If it went open, there'd be no path for current during the part of the cycle where the chip output was off. That would mean a large negative voltage being driven back into the chip.
It would also mean (probably) low output.

So we're left with the chip.
Or a bad connection on your circuit - such as the feedback pin being open circuit.

hampi
Posts: 223
Joined: Fri May 31, 2013 11:29 am
Contact: Website

Re: lm2576-5 circuit failing with 13v output

Sat May 03, 2014 3:46 pm

These DIY circuits need to be tested before use. You need to have a fuse between 13 V power source and the LM2576. When you connect the power first thing to do is to measure the output voltage. Then you should connect the supply to a load resistor 2 - 5 ohms/25 W and check the output voltage again.

My LM2576-ADJ works and get the login message

Code: Select all

Linux rpia1 3.10.25+ #622 PREEMPT Fri Jan 3 18:41:00 GMT 2014 armv6l
* * * Solar powered computer experiment * * *
Last login: Sat May  3 16:25:07 2014 from pc
Battery at 12.4 V 78.0% 12.5 h left 38.6 h to empty at 14.2 °C CPU temp=22.9'C

jay123035
Posts: 32
Joined: Wed Nov 21, 2012 5:24 am

Re: lm2576-5 circuit failing with 13v output

Wed May 14, 2014 5:35 pm

thanks for your help. I swapped out the LM2576-5 and the circuit came back to life.


what would cause a LM2576-5 to go bad like that?

User avatar
Burngate
Posts: 6289
Joined: Thu Sep 29, 2011 4:34 pm
Location: Berkshire UK Tralfamadore
Contact: Website

Re: lm2576-5 circuit failing with 13v output

Thu May 15, 2014 8:14 am

jay123035 wrote:what would cause a LM2576-5 to go bad like that?
You may have to ask the manufacturer - though whether you'd get an answer I don't know.

Return to “Automation, sensing and robotics”