Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations KootK on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Voltage Drift

Status
Not open for further replies.

point7

Electrical
Sep 21, 2005
2
Hello,

I have a switching power supply that distributes voltage to numerous loads. I run a 10 hour test to verify the stability of it.

It fails 75% of the time. Besides a possible short or noise in one of the loads, what other factors can I consider the cause of this voltage drift?
 
Replies continue below

Recommended for you

point7; We can make few (if any suggestions) if you don't provide better information...

What specifically do you mean "it fails"??
 
What's the spec and how does it fail?
Did it meet the spec before some event in the past?

TTFN



 
A possible explanation is that the design is marginal as compared to the limits. 25% of the time the test sample just happens to scrape by the test limits.

Try extending the test to 30 or 40 hours. Once the failure rate is closer to 100%, then the way forward might become more clear.

The 25% is perhaps just a distraction. (Or maybe not...)

 
Here is a more detailed description of my test failure:

I have a switching power supply that outputs loads at:

24VDC @ 12A failed
8VDC @ 3.22A passed
15VDC @ 4.03A failed
-15VDC @ 3.22A failed
8VDC @ 830mA passed
8VDC @ 840mA failed

Maximum peak deviation is approximately .4%.Any output that falls out of this range is considered out of spec and faulty.
 
Did it meet the spec before some event in the past?

TTFN



 
It may be good to look at the statistics of the measurements on all of the units, not just the failures. You could then look at the distribution of one of the measurements and see where your failures are and if your distribution look Gaussian. You would want to compare this to the prediciton distributions your design engineers generated and see if they matched up. You can also see how the distribution changes with time or with any production changes.
 
Are you saying that it fails test because the tolerances on the outputs are exceeded, or that it fails 'Ka-Boom' and requires repair or replacement? Don't drip-feed information - your initial post should list as much as you can detail about your problem. We can't help fix a problem that you can't describe!


----------------------------------

One day my ship will come in.
But with my luck, I'll be at the airport!
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor