The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
Subject: Re: Philips MKT cap failures - LV pulse application
X-Newsreader: Microsoft Outlook Express 6.00.2600.0000
Date: Wed, 25 Sep 2002 05:43:11 GMT
NNTP-Posting-Date: Tue, 24 Sep 2002 22:43:11 PDT
Organization: EarthLink Inc. -- http://www.EarthLink.net
"Andrew Kohlsmith" wrote in message
> I've been using Philips MKT .012uF caps in a 5V sawtooth generator circuit
> for many years. Just recently I've been noticing parts failures both in
> test and in the field - the capacitance dropped to roughly 10nF or so,
> really screwing up the slope of the waveform :-)
> From looking at old boards I've noticed that the original parts were
> MKT-HQ (or HO, hard to tell)
> About a year ago these parts read
> 100V PH
> All the parts that are failing are marked
> 100V BC
> The datasheet for the MKT 370 series doesn't tell me anything about these
> two-letter markings but my gut instinct is telling me that it must have
> something to do with either dv/dt or clearing ("healing") energy
> Has anyone run across this? Does anyone know what this HQ/PH/BC suffixes
> are about? These are all 100V rated parts and their specified dv/dt
> characteristics are well within what I'm doing.
> My parts vendor claims there is no difference in these parts but my gut
> instinct is telling me that these two-letter codes are indicating
> *something*; I just can't tell what.
> Basic circuit is an LM339 arranged as an integrator -- the + pin at +4V
> the - at 5V through 10k. The output is connected to - through this cap,
> and there is an everyday generic LM3904 transistor across the cap which I
> activate to short it out and start the cycle over again.
> With this transistor and the driver (output from a 74HC XOR gate) I can't
> believe I'm exceeding the 110V/us dv/dt rating. Just to be sure, I've
> sub'ed in a specific pulse-rated self-healing cap (Wima FKP series) --
> Panasonic's ECQ series (ECQP1H123GZ to be exact) seems to be another
> substitute. Both have far greater dv/dt ratings but I'm wondering if
> anyone out there has any experience with either of these (or any other
> 370-footprint part) in low-voltage, pulse applications.
First off, if your circuit is sensitive enough to fail when that cap changes
from 0.012uF (12nF) to 10nF then you've got a big problem to start out with.
A 20% variance in value in caps should be expected -- maybe not for a given
unit but certainly from unit-to-unit.
Is this cap a through-hole film device? If it's a surface-mount ceramic then
your problem is probably simple to understand, as smt ceramics are notorious
for developing cracks due to differences in thermal expansion;/contraction
coefficients between the cap and the pcb. Most cracks happen during the
If the pulse current is too high (high dv/dt), and that's what's causing the
cap to change its value, then perhaps you could insert a small resistor in
series with the shorting transistor to limit the cap current when the
transistor is turned on.
We're currently experiencing problems with tantalums exploding due to too
high inrush current when the power supply is first applied to them.
You see, capacitors are like women -- you can't live without them and you
can't live with them.
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup