« Big Fish Eat Little Fish - Dick LaRosa | Main | Anti EMI and EMP Grounding - John Dunn, Consultant, Ambertec, P.E., P.C. »
Texas Instruments offers the TLV431 series of shunt voltage regulators. Someone had made use of this particular device in a circuit design and just to keep things nice and quiet, he had provided the chip with a 0.1 µF bypass capacitor.
Wrong move!
That value of 0.1 µF is right smack dab in the middle of the highlighted sketch below of capacitance values that can render that regulator unstable and oscillatory and it did.
As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.
Having trouble reading this image? View an alternate.
Comments are moderated, and will not appear until the author has approved them.
Your Information
(Name and email address are required. Email address will not be displayed with the comment.)
Learn about our
free engineering
consulting referral
service at:
We have over forty
electrical, electronic,
mechanical and
software engineers
with expertise in more
than sixty-five categories
of technology and business.
All are members of the
IEEE and adhere to the
IEEE professional codes
of ethics.
No fee is charged for the
referral service.
Each member is an
independent consultant
and negotiates his/her
own consulting agreement.
Editor: Jerry Brown
Contributors:
Carl Schwab
Unless otherwise noted,
reprinting or republication
of anarticle on this blog is
authorized by crediting the
author and prominently
displaying the following
sentence at the beginning
or end of the article,
including the hyperlink to
"This report is republished
with permission of IEEE
Consultant's Network of
Long Island"
John,
TI and Diodes Inc do not recommended to use capacitor 0.1uF by well known reason. The test circuitry that you copied from TI datasheet is for determining an effect of capacitance value on stability.
If somebody is not reading a datasheet, your article will not help him anyway.
Posted by: Vladimir Doubovis | November 08, 2011 at 10:56 PM
Indeed so, Vladmir. It's pretty much certin that the circuit's designer didn't read the data sheet, at least not carefully.
It's just that when the mischief was afoot and the TLV431 was in oscillation, there were several people there who were quite puzzled by the phenomenon and who didn't happen to think of the possibility that a shunt capacitance might be harmful. Hence, this caveat.
Posted by: John Dunn | November 09, 2011 at 12:15 AM
Vladimir: I don't entirely agree -- I know a lot of engineers who tend to only skim datasheets because they figure the device they're using is pretty self-explanatory and/or they've used similar devices before, but they read articles like John's blog here (or the articles in EDN or on enGenius or wherever) *specifically because they're looking for the "gotchas" that aren't obvious from a cursory pass over a data sheet*.
I would agree that with the TLV431 here, the manufacturers make it pretty darned clear that you need to be careful with the output capacitance. On the other hand, on many data sheets manufacturers seem to go to active effort to hide all the little warts of their designs! (It's also a fact these days that many engineers who only dabble a bit in the analog world don't even know what many of the data sheet specs mean...)
An interesting question is... is there a part comparable to the TLV431 that's unconditionally stable? -- That seems like it could be a significant competitive advantage to a manufacturer...
Posted by: Joel Koltner | November 09, 2011 at 12:35 PM
I would imagine all "431" devices are designed or processed differently so at the margin they may work differently.
Posted by: David Gillooly | November 09, 2011 at 01:10 PM