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.
From: "Christopher R. Carlen"
Subject: Re: Bad LM338K?
Date: Wed, 11 Dec 2002 14:31:30 -0800
Organization: Sandia National Laboratories, Albuquerque, NM USA
NNTP-Posting-Date: Wed, 11 Dec 2002 21:29:47 +0000 (UTC)
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
> Hi guys,
> I'm looking for a answer to a puzzling question ...
> I've build a power supply from a LM338K (TO3) with a 10.000 uF and a 100uF
> capacitor on the input line and a 1uF and a 0.1uF capacitor to output a
> fixed 24 volts DC supply.
> I 've got a 220 ohm resistor between LM output the adjust of the LM, and a
> 4k7 trimpot to the ground ... which I adjust to 24 VDC (with almost no load)
> ... the transformer delivers 35 volts DC to the input of the LM (no ripple).
> The transformer is rated min. 10 Amps.
> All correct so far? Fact is that it allows me to drive small loads ... eg.
> power resistor 51 Ohm ... 24 volts output is kept ... resistor warms up ...
> no problem ...
> Now I put a second 51 Ohm resistor in parallel to the first resistor ... and
> the output voltage of the circuits drops 4 volts (to 19-20 volts), to the
> best of my knowledge this is way under the power specs of the LM338, as it
> should be able to drive 5 Amps with input-output difference of 11 volts?
> How can this be explained? At this state even adjusting the trimpot has no
> effect on the voltages anymore ... it's like the regulator has found a "safe
> state" and does not want to provide 24 volts ... it almost does not produce
> any heat ... could it be damaged? Or is there a simpler explanation?
> Many thanks,
Double check the input voltage to the regulator. With Vin-Vout>10V, the
output current may be limited to less than 5A, but this shouldn't really
be the problem here since you're only at 1A. What is the current
capacity of your transformer/rectifier/filter cap system? Is it's
output drooping below 27V when you power both resistors? Also, what is
your heatsinking thermal resistance? You have to keep that LM338K cool.
If you want to get anywhere near the rated 5A output, you will need a
pretty big heatsink.
Finally, I have seen LM338K oscillate. If your output caps are close to
the regulator output, this shouldn't be a problem. But sometimes things
go awry because the output terminal isn't usually close to the ground,
in a chassis assembly. A PCB assembly is better for this. What I do
for a chassis wiring type of layout is run a fat wire from the LM338K
output lead to my binding post or whatever is the output terminal of the
system. And run a fat wire from the rectifier/filter minus terminal to
the minus output terminal of the system. These fat cables are less than
6 inches long. I put a 1uF Ta across the output terminals, and a big
Schottky rectifier in reverse biased configuration, in case the supply
will be in series with any other supply, this will prevent bad things
from happening (like, BANG!, speaking from experience).
The adjust pin I route to my potentiometer, and the ground side of the
pot goes to the ground output terminal of the system. That is the
wiring that the datasheet says to use. This should work well, and not
oscillate. Can't quite remember where I put the other resistor from
output to the adjust pin. Not so critical I think.
But you never can be sure unless you scope it.
I have built a dual 28V 5A supply for my home workbench using LM338K, so
I speak from my experience. You can see it here:
P.S. Try to post to only one newsgroup at a time.
Christopher R. Carlen
Principal Laser/Optical Technologist
Sandia National Laboratories CA USA
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup