LM350 Voltage Regulator Overheating
Chuck Ricketts
Hello, I recently built a power supply for a laser driver circuit that
I put together at work. I built a linear supply using parts rated for 3
amps or higher. The problem I'm having is keeping the damned adjustable
voltage regulator cool. This chip seems to overheat with relatively small
loads (given enough time) , and shuts itself down. I've tried mounting to
an aluminum heat sink (it's a TO-3 package) but the thing still overheats.
I've gotten it to work but only after installing a blower directly over the
chip. It would then stay cool with a dummy load pulling 2.3 amps. But it
still seems too hot to me. Are these regulators prone to overheat? It just
doesn't seem like you should have to go to such extreme measures to keep it
cool.
Thanks in advance,
Chuck Ricketts
David Vlack
Chuck Ricketts wrote in message <01bd39c5$769d4200$e09493d0@master>...
> Hello, I recently built a power supply for a laser driver circuit that
>I put together at work. I built a linear supply using parts rated for 3
>amps or higher. The problem I'm having is keeping the damned adjustable
>voltage regulator cool
What voltage are you dropping across the regulator?
Remember power= IxV. 3 Amps and a 10v drop is 30 watts!!
--
Dave
(Remove zzz from domain name to reply by mail)
Jerry G.
Hi there...
It seems that your powersupply is under-rated for your purpose. You must
get or build a supply with much more current load capability. The load
may be demanding very high current bursts for extreemly short time periods.
That is why when loading with a streight 2 or 3 amps the output does not
overheat.
One quick and dirty way around it is to have a very very large electrolytic
high current capacitor accross the output of the supply. Take care to make
a provision to charge it without blowing the voltage regulator. You may
need to make some kind of circuit to protect the regulator so that it can
charge the cap. What you are doing here is giving the power-supply "dynamic
headroom" for current load. This of coarse only applies if you need pure
DC. I don't know the required criteria here...
--
You must remove the NOSPAM from my internal
address to reply to me. Sorry about that...
Reply To: jerryg...@total.net
If It Works, Don't Fix It !
If It Don't Move, Slap It !
If It Don't Make Noise, Shake It !
If It's Broke, Then Fix It ... !
If You Don't Know How To Fix It, Give It Out !
W-Site http://www.total.net/~jerryg
-----------------------------------------------------------------------
Chuck Ricketts wrote in message <01bd39c5$769d4200$e09493d0@master>...
| Hello, I recently built a power supply for a laser driver circuit that
|I put together at work. I built a linear supply using parts rated for 3
|amps or higher. The problem I'm having is keeping the damned adjustable
Sam Goldwasser
What is the voltage drop? Usually, the dissipation is simply Vdrop*I. At
2.3 A, it doesn't take much to be a nice heater! You may need a BIG heat
sink. Also, I assume you have the proper caps on input and output.
--- sam : Sci.Electronics.Repair FAQ: http://www.repairfaq.org/
Usually latest (ASCII): http://www.pacwest.net/byron13/sammenu.htm
Lasers: http://www.misty.com/~don/lasersam.html
Henry van Cleef
In article <01bd39c5$769d4200$e09493d0@master> "Chuck Ricketts" <chkr...@HiWAAY.net> writes:
>amps or higher. The problem I'm having is keeping the damned adjustable
>voltage regulator cool. This chip seems to overheat with relatively small
>loads (given enough time) , and shuts itself down. I've tried mounting to
>an aluminum heat sink (it's a TO-3 package) but the thing still overheats.
>I've gotten it to work but only after installing a blower directly over the
>chip. It would then stay cool with a dummy load pulling 2.3 amps. But it
>still seems too hot to me. Are these regulators prone to overheat? It just
>doesn't seem like you should have to go to such extreme measures to keep it
>cool.
>
the equation. W=EI and 5 volts will give you 10 watts dissipation at
2 amps. Yes, they are "prone to overheat" as they are
power-dissipating devices, and power handling is part of the basic
design calculation.
There are some things you can do, depending on the characteristics of
the supply and the actual load. One is to put a shunt resistor across
the regulator to carry some of the current. Another is to put a
series resistor in front of the regulator to reduce some of the
voltage. Combination of the two sometimes is called for. The
resistors will, of course, get hot----you are moving the dissipation
from the LM regulator device to the resistors.
Be careful to assure that the supply side voltage is ALWAYS above the
load side---reverse current through regulators will blow them. Also,
watch out for parasitic oscillations. Be sure to scope the output when
it is operating.
A nice big heat sink, and mounting with grease, are called for if you
are going to dissipate more than 1-2 watts in the device. My "rule of
thumb" target is that the sink should not get too hot to touch (around
50C) operating in a room temperature (20C) environment.
--
===================================================================
Hank van Cleef
===================================================================
John R.Hepburn
What everyone has said is dead on . Now what to do. Typically when I design
power supplies I make sure the supply voltage d.c. minimum (ripple) is not
much greater than 20% over the regulated voltage maximum (with a 5 volt
minimum spread). This is assuming that you do not have radical swings in
your a.c. line voltage.If you do then the above is from a.c. minimum. If
your raw d.c. is too high, are there taps on the transformer to drop it
down?...or is changing the transformer practical?Could you install a fixed
regulator to get it down to reasonable levels?(wattage sharing). If not then
a good cooling system is your only real solution.Given your current *
minimum voltage drop, a heat sink is nessesary, but you should keep it
reasonable if possible.
Good Luck
John R. Hepburn
John's Electronic Services
jhep...@recorder.ca
Allan
Chuck Ricketts wrote in message <01bd39c5$769d4200$e09493d0@master>...
> Hello, I recently built a power supply for a laser driver circuit that
>I put together at work. I built a linear supply using parts rated for 3
>amps or higher. The problem I'm having is keeping the damned adjustable
>voltage regulator cool. This chip seems to overheat with relatively small
>loads (given enough time) , and shuts itself down. I've tried mounting to
>an aluminum heat sink (it's a TO-3 package) but the thing still overheats.
>I've gotten it to work but only after installing a blower directly over the
>chip. It would then stay cool with a dummy load pulling 2.3 amps. But it
>still seems too hot to me. Are these regulators prone to overheat? It just
>doesn't seem like you should have to go to such extreme measures to keep it
>cool.
>
>Thanks in advance,
>
>Chuck Ricketts
most of these Voltage regulators need two caps Very Close to the INPUT and
OUTPUT Pins to ground. the input cap should be around 10 uf and a greencap
of about .1 uf to .022 uf on the output should stop it getting Hot. It
sounds like you are getting RF floating around and its overheating the
Regulator.
Allan
all...@pronet.net.au
Dave Moore
If the current is relatively constant, you can
use a bypass resistor to offload some of the
current. You can put a resistor in parallel or
series with the regulator.
--DM--
Chuck Ricketts wrote in message <01bd39c5$769d4200$e09493d0@master>...
> Hello, I recently built a power supply for a laser driver circuit that
>I put together at work. I built a linear supply using parts rated for 3
>amps or higher. The problem I'm having is keeping the damned adjustable
>voltage regulator cool. This chip seems to overheat with relatively small
>Chuck Ricketts