Previous in Forum: Power Factor Improvement   Next in Forum: Power Generation - Neutral And Ground
Close
Close
Close
6 comments
Rate Comments: Nested
Guru

Join Date: Nov 2006
Posts: 729
Good Answers: 2

PCB Failure.

10/17/2007 2:23 PM

A customer has lot of pcb failures due to over voltage.The system in his premises consists of one 3150KVA transformer and about 3200KVAR (Total)capacitors connected across various loads. Incoming voltage has been monitored and has been steady.What causes over voltage?could it be a occillation when some loads are switched off between Transformer inductance and capacitors.He has been advised by some consultants to carry out harmonic analysis which I feel may not be helpful. what could be the cause and solution for preventing this.

__________________
To avoid crticism do nothing,say nothing,be nothing
Register to Reply
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.
Guru
Canada - Member - Our strength is our diversity

Join Date: Jan 2007
Location: Canada
Posts: 1024
Good Answers: 40
#1

Re: PCB Failure.

10/17/2007 3:30 PM

here are just a few possibilities:

  1. Large loads are switched off causing a feedback.
  2. Changing ground potential due to electrical storms
  3. Inconsistent generation
  4. Radio frequency interference, noise from electrical devices
  5. Solar radiation
  6. Meteorites
  7. Lack of surge suppressors, noise filters
  8. Electromagnetic pulses
  9. Poorly shielded conduits
  10. Static electricity
  11. Not using a static ground strap when handling the PCPs will cause damage before the PCB is even installed.

A good way to start, is to monitor the lines for a long period of time. That way when you have a failure, you may be able to identify what is causing the surges.

You could also monitor the lines and shut down each piece of equipment 1 at a time. When you find the culprit, either add a line filter, clamp, or find a way to shut it down in stages.

__________________
Perfection is a subjective and abstract concept.
Register to Reply
Guru

Join Date: Sep 2007
Posts: 1817
Good Answers: 7
#3
In reply to #1

Re: PCB Failure.

10/18/2007 1:57 AM

It is not just the switching off that is the problem, when you switch a machine on that contains an direct on line transformer, it will spike back at approximately 7 times the going load current for some 50 to 100 ms due to the magnetic field being build up inside the transformer. This building up is perceived as a moving magnetic field and thus generates a reverse current.

If this building or the other equipment has no protection against this the contractors should be found and asked to rectify this.

Register to Reply
Guru

Join Date: Aug 2007
Location: Indiana, USA
Posts: 579
Good Answers: 61
#2

Re: PCB Failure.

10/17/2007 5:35 PM

Harmonics is actually a good possibility. If the capacitors are 3-phase and all 3 poles are connected simultaneously, at least 2 of the 3 will connect at some point on the sine wave other than zero crossing. This causes either a spike or sag on that phase, depending on the charge state and polarity of the capacitor.

One of my manufacturing customers had a similar problem. I used a Dranetz disturbance analyzer to monitor the internal distribution feeder for 24 hours, and found that their own PF correction capacitors were imposing some severe harmonic spikes during switching. Since they needed the caps to avoid a PF penalty fee, I recommended they install a full-isolation UPS (the only connection between the incoming AC and the load AC is through the battery) and connect all of the machine control circuits to it. For eight years, they have had no further problems. They also sized the UPS to supply all of the computers in the offices, receiving & shipping, preventing any loss of data or productivity.

A Dranetz or similar full-spectrum analyzer should be available for rent from a test equipment house for a reasonable price, and it is very easy to connect and program. Even if the problem is not the capacitors, the analyzer will record the time and magnitude of spikes so you can find a pattern and track down the cause.

__________________
Experience: The knowledge you gain just AFTER you needed it.
Register to Reply
Power-User

Join Date: May 2007
Posts: 127
Good Answers: 6
#4
In reply to #2

Re: PCB Failure.

10/18/2007 2:00 AM

I agree on the harmonic analysis recommendation because it should help to identify the likely internal source of the voltage spikes.

If any of the loads are large three phase electric motors then dynamic braking voltage spikes during motor braking are a likely source for problems. These should be properly ground shunted but can severely back feed especially if the ground connection itself is insufficient or just old.

Register to Reply
Guru

Join Date: Sep 2007
Posts: 1817
Good Answers: 7
#5
In reply to #4

Re: PCB Failure.

10/18/2007 2:14 AM

Are these not shorted over large capacity resistors?

I work with dc motors and have yet to see a dynamic braking system that shorts to earth when the motor is slowed down. All the generated power is absorbed internally by the dc control.

Maybe it is different with 3 phase motors but I would not like to use my earth connection for that purpose.

Register to Reply
Guru

Join Date: Feb 2006
Posts: 1758
Good Answers: 6
#6

Re: PCB Failure.

10/20/2007 4:11 PM

It is suggested that:

Following points clarified:
1. Do the PCB(s) itself fails after assembly or
[PCB] = certain cards fail?
2. Are these Cards Professional designs & manufactured? And being used in an OEM Euipment?
3. Are these cards part of some Standard Equipment or p/o some Fabrication / Manufacture Setup
for Cards?
4. Are the failures during normal operation or due to Power Outages?
5. Can we get some summary of his findings on failed PCBs [Cards]?

I have some suggestions in general:

6. The Power Supply for Equipment or Indiginuous Setup be checked for:
a. Output Voltages' OV limits? Are these protected? Passively or Dynamic?
Some Discrete OV protectors be used.
b. The DC Filter capacitor:
A Hi-Capacity [Backup] capacitor in DC output may absorb Spikes in power line.
And Holds POWER to critical circuits for normal SAGs.

Register to Reply
Register to Reply 6 comments
Copy to Clipboard

Users who posted comments:

case491 (2); Haajee (1); Mr Gee (1); pwr2thepeople (1); techno (1)

Previous in Forum: Power Factor Improvement   Next in Forum: Power Generation - Neutral And Ground

Advertisement