On Tue, 2007-03-06 at 17:34 -0500, Bill Acito W1PA wrote:
> Jerry,
>
> Could I just swap in a Darlington pair like the 2N5686 (80V, 50A) and loaf
> along at my 18A, and be that much further inside the max current limit (and
> less likely to fail?)
That depends on circuit details. The Darlington will have a great deal
more current gain but at the cost of at least twice the base to emitter
voltage when forward biased. And it will have quite a bit greater
collector to emitter saturation voltage, so it may go out of regulation
sooner. And if the power supply design lets the regulator chip provide
significant output to be a current sample for current and temperature
limiting, the great current gain of the Darlington will keep that from
working at useful currents. And that's all supposing it doesn't
oscillate from the 100 times greater loop gain with the Darlington
transistor in th circuit.
>
> http://www.onsemi.com/pub/Collateral/2N5684-D.PDF
>
>
> Or as you say, I'm at the mercy of heat dissipation (buy the 2N5302 and a
> 12v fan for the pass heat sink)
If the heat sink won't keep the transistor case temperature in check
when the crowbar fires a fan won't help. Needs more metal in the heat
sink, preferably copper for greater heat capacity. The transistor can
only dissipate those 150 watts (or whatever its rated) if the case can
be held to 25 degrees C, or an infinite heat sink.
>
> Bill
> W1PA
>
Long ago, I designed a negative side regulator with a pair of 2N5302 and
a LM120K regulator. Then I didn't have to insulate the transistor
collectors since the were at the grounded output terminal of the supply.
MY dad built a couple, I built one and ran it 24/7 for years until
lightning got it. I have the two my dad built. He started a PE103
dynamotor on his in the shack. The other is in the Airstream I now have
for battery charging. When you go to two transistors you have to use up
a lot of the voltage drop in emitter resistors to get decent current
sharing, which is why some commercial supplies eat 2N3055. There's too
little voltage drop in their emitter resistors for good current sharing.
My present shack supply uses power MOSFETs as a negative side regulator
with Schottky diodes for the main rectifiers and a 12 volt RMS
transformer. The transformer is good for many amps, but the transformer,
diode and regulator drop run the MOSFETS out of regulating room (20
millivolts) at about 30 amps load. Tested at that load it runs some 92%
efficiency and is all linear. MOSFETS share current very well when
paralleled not needing the equalizing resistors of the paralleled
bipolar transistors. Its probably been running 8 or 10 years now 24/7
with no problems. And with the MOSFETS in the negative side, their
drains/cases are also grounded for better heat sinking. Though with only
16 volts unregulated at no load they don't dissipate much power and
don't need a crowbar to protect the radio should they short.
--
73, Jerry, K0CQ,
All content copyright Dr. Gerald N. Johnson, electrical engineer
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|