Re: Bad cap topologies

On Mon, 17 Aug 2009 12:16:47 -0700, D Yuniskis
<not.going.to.be@xxxxxxxx> wrote:

Hi PlainBill,

PlainBill47@xxxxxxxxx wrote:
On Mon, 17 Aug 2009 06:22:49 -0700, D Yuniskis
<not.going.to.be@xxxxxxxx> wrote:

[much elided]

What I would like to know is which circuit topologies
tend to aggravate this problem. From my casual
observations (I've done most of my repairs without
the benefit of any design documentation), the failing
components either seem to be proximate to heat sources
*or* in configurations where they see high ripple
currents (suggesting this is a problem with the
devices' ESR -- internal heating).

Does anyone have any *definitive* answers about this?
And, long-term remedies? (i.e., does replacement with
a good, high temp, low ESR cap *solve* the problem or
just kick it down the road?)

Are there lessons to be learned when *designing* these
types of circuit topologies to avoid these failure
modes? (besides picking good vendors)

There are a number of factors involved.
1. Poor quality capacitors.

Yes, as I mentioned in my original post. But, my question
is intended to address the *expected* results if "good"
quality capacitors are used in the same circuit topologies.
I.e., will they also exhibit similar failure modes -- just
further down the road? (i.e., what is it about the topology
that causes the failures)

2. High surge currents found in SMPS and the backlight inverter

I.e., caps that handle the large ripple currents.

3. Proximity to heat producing components, chiefly heat sinks

Yes, but this doesn't seem to be as reliable a predictor of
failure. Often there are caps literally *touching* parts
that run VERY hot; yet they don't appear to fail as often
as other parts "free standing" (i.e., nothing within an inch!)
elsewhere in the circuit.

4. Poor ventilation of the electronics portion of the monitor.

Again, that would tend to affect every component in the
circuit (roughly) equally. No doubt it is a contributing
factor -- no doubt alol of the above are contributing
factors!

Because of the current manufacturing / distribution pattern, there is
limited feedback from consumer to designer. Still, the designs of LCD
monitors continue to evolve. A few years ago a 5V 4A power supply was
common. Today the 5V supply is less than half that. The monitor
logic card is being integrated into the LCD panel electronics, further
reducing component count and cost, and improving reliability. With
LED based backlight systems power demands will drop further.

As far as existing monitors, my recommendation is to replace all caps
(except the 150 µF 450 Volt one) with good brand low ESR parts. My
personal preference is Panasonic FM and FC series', but others have
equivalent success with Rubycon and Nichicon.

I've been using the Panny parts as (historically) they have been
"very good to me" :> But, I ownder if I am just buying a little
more time before similar failures remanifest.

And, as a *designer*, I am interested in determining the real
cause of the problem(s) to ensure that I don't repeat these
problems in my own designs...
It depends on your definition of 'a little time'. My primary source
of information is www.badcaps.net/forum It would appear you are at
least doubling the MTBF (mean time between failures) by using high
quality caps. That would mean if you replace the caps in a two year
old monitor, it will probably last an additional 4 years before it is
necessary to replace them again. At worst, that is a significant
improvement.

When I look at the cost, power consumption, clarity, and design of 4
year old LCD monitors vrs those one or two years old, I doubt that in
5 years you would be asked to repair many 7 year old monitors.

I would agree, having to redo the replacement every two years could
cause speculation about your skills.

PlainBill
.

Relevant Pages

  • Re: Series Termination
    ... reduced the trace width necessary for a particular transmission line. ... So far there is no useful output from the FPGA, ... much design there yet) and transmitted. ... loops with multiple caps on the supply lines. ...
    (sci.electronics.design)
  • Re: for all those who believe in ASICs....
    ... Experienced team A works diligently, ending in a heroic failure ... defect managment is functional issolation by designing for 100% test ... suspect one of the most important parts of the process will be design ... Now, I don't know how much of Xilinx's yield is scrap today, or would ...
    (comp.arch.fpga)
  • Re: High Voltage Caps for Plate Supply ??
    ... failure every three years, your failure rate is nearly insignificant. ... so you've got to have a pretty good supply charging those caps. ... been made to cooling the capacitor bank. ... voltage across the bank is 3600 volts, the transformer rated at only 0.8 ...
    (rec.radio.amateur.homebrew)
  • Re: SI and EMI training
    ... signals require 4 different power supplies (an entire EMC/EMI issue in ... few extra caps into the thing. ... I had a device (one of the x-point switches in the above design as it ... happens) that was flip-chip and had a hefty heat sink. ...
    (comp.arch.embedded)
  • Re: Learning embedded coding, which uC?
    ... >> There are engineering practices that help to make sure ... and that includes documentation during the design process so you ... > can hold proper and regular design reviews. ... > consequences of a certain failure are. ...
    (comp.arch.embedded)
Loading