From: Winfield Hill
Subject: Re: wiring harness using BJT (instead of relays) Check design please?
Date: 23 Oct 2002 07:08:25 -0700
Organization: Rowland Institute
X-Newsreader: Direct Read News 2.98
>> The 2N3772 is a Power BJT, why does it not have a low dissipation
>> like the power FET?
> A BJT, when switched "fully on", drops about 1.4 or so volts across
> its collector/emitter junction.
Careful Stuart, don't overstate the case for FETs! A "fully on" BJT
has a collector-to-emitter saturation-voltage, called Vce(sat), which
can easily be under 100mV at moderate currents. The 2n3772 data sheet
may mention a high 1.4V maximum C-E drop at 15 amps, but that's a good
reason not to use that transistor at such a high current! For example,
at 10A and 75C, Vce(sat) is about 0.60 volts with 2A drive, according
to an old Westinghouse data sheet I have.
> A FET has effectively inconsequential voltage drop across its drain/
> source junction. A smaller voltage drop (V) means less power (P)
> dissipated in the device at a given current, because P = V x I.
Inconsequential? The "fully on" FET is characterized by a parameter
Rds(on). For an IRF540 (as a 27A 100V FET this could be considered
a typical replacement for a 2n3772), this is about 0.066 ohms at 75C,
according to a 1998 Motorola data sheet. That's about 0.66V at 10A.
Of course there are more modern power MOSFETs, but then there are
more modern power BJTs too. :-)