Author Topic: Hardware debugging  (Read 3233 times)

0 Members and 1 Guest are viewing this topic.

Offline JaneTopic starter

  • Frequent Contributor
  • **
  • Posts: 381
Hardware debugging
« on: November 18, 2015, 08:18:20 pm »
I need to repair a device that uses a dual synchronous
step-down switching regulator controller LTC3850.
Datasheet is here
http://cds.linear.com/docs/en/datasheet/38501fc.pdf
Let's suppose it uses the typical application design according to the attached picture.
How would you start to find out the faulty part if there is  no
Vout1, nor Vout2? The input voltage Vin is correct.
Thank you for you ideas.
 

Offline fivefish

  • Frequent Contributor
  • **
  • Posts: 440
  • Country: us
Re: Hardware debugging
« Reply #1 on: November 18, 2015, 08:23:24 pm »
Spend the $5 and try replacing the LTC3850. :)
 

Offline JaneTopic starter

  • Frequent Contributor
  • **
  • Posts: 381
Re: Hardware debugging
« Reply #2 on: November 18, 2015, 09:13:11 pm »
I did replace  LTC3850 and yet no output voltages
 

Offline Aodhan145

  • Frequent Contributor
  • **
  • Posts: 403
  • Country: 00
Re: Hardware debugging
« Reply #3 on: November 18, 2015, 09:16:45 pm »
What voltage are you measuring with your DMM?
 

Offline Brutte

  • Frequent Contributor
  • **
  • Posts: 614
Re: Hardware debugging
« Reply #4 on: November 18, 2015, 09:26:22 pm »
I'd do some optical inspection, then test for shorts/opens, values of passives and the transistors, in-circuit. In case nothing suspicious was found, I'd try ramping up supply voltage with current limiter and check voltages. These two regulators are independent so the fault must be close to power supply or the reference (which are common).
 

Offline KL27x

  • Super Contributor
  • ***
  • Posts: 4099
  • Country: us
Re: Hardware debugging
« Reply #5 on: November 18, 2015, 11:27:50 pm »
I'd turn it on and probe the FETs with a scope. Something is obviously wrong, there. Starting at the obvious lack of Vout, work back from there.

Vin > top FET > inductor >Vout. There's the first path to check out, in my opinion. Perhaps the bottom FETs are burned/shorted out and the top FETs are ok. That would be one way to get 0V on the outputs.
« Last Edit: November 18, 2015, 11:36:39 pm by KL27x »
 

Offline fivefish

  • Frequent Contributor
  • **
  • Posts: 440
  • Country: us
Re: Hardware debugging
« Reply #6 on: November 18, 2015, 11:36:59 pm »
I did replace  LTC3850 and yet no output voltages
You didn't mention that in your 1st post.
It's probably not switching, maybe shorted or open transistors, so try replacing all the FETs.
Also check inductors if they're open/burnt?
Use the fingers... any component getting hot?
 

Offline JaneTopic starter

  • Frequent Contributor
  • **
  • Posts: 381
Re: Hardware debugging
« Reply #7 on: November 19, 2015, 10:10:53 am »
Thank you all for the ideas. Regarding FETs, do you think the checking by DMM( digital multimeter) is OK?
Is it necessary to desolder them first or I can check them in the circuit?
 

Offline KL27x

  • Super Contributor
  • ***
  • Posts: 4099
  • Country: us
Re: Hardware debugging
« Reply #8 on: November 19, 2015, 11:28:17 am »
In circuit, you will probably best to check them with a scope with the circuit powered.

You might have to take them out of circuit to make sense of them with DMM. Unless they're burned and you get continuity between the gate and D/S (with the circuit unpowered, of course).

IMO, in any sort of switching supply like this, it is usually a good idea to check the FETs, first, before you replace the driving IC. If the FET is smoked, it'll usually kill the IC. An output pin doesn't normally last too long when it's pinned to one of the rails.
« Last Edit: November 19, 2015, 11:40:24 am by KL27x »
 

Offline deephaven

  • Frequent Contributor
  • **
  • Posts: 796
  • Country: gb
  • Civilization is just one big bootstrap
    • Deephaven Ltd
Re: Hardware debugging
« Reply #9 on: November 19, 2015, 12:31:39 pm »
The first thing I would do is (with the power off) check that there isn't a short or low resistance between the output and 0V. There might not be anything wrong with the regulator.
 

Offline JaneTopic starter

  • Frequent Contributor
  • **
  • Posts: 381
Re: Hardware debugging
« Reply #10 on: November 19, 2015, 02:52:49 pm »
@KL27x:Can you please let me know how I can check FETs ,in circuit,with scope?
Thank you
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf