Subject: Re: Constant current source with low voltage drop
Date: Sat, 12 Oct 2002 11:56:13 +0100
References: <3D9A5E9A.email@example.com> <3D9AF62C.firstname.lastname@example.org>
NNTP-Posting-Host: modem-30-18-60-62.vip.uk.com (22.214.171.124)
X-Newsreader: Microsoft Outlook Express 5.50.4522.1200
X-MIMEOLE: Produced By Microsoft MimeOLE V5.50.4522.1200
"Winfield Hill" wrote in message
> "markp" wrote...
> > Winfield Hill wrote
> >> Indeed, OK, here's a more simpler current-source circuit that's good
> >> to 10 or 20% and meets the goal of a low 250mV sense-resistor drop.
> >> 5.0V R1 2.0 ohms
> >> o--+-----/\/\----,
> >> | |
> >> E R2 100 |
> >> Q1 B --+-/\/\--+ |
> >> C | | |
> >> | \ R3 | V
> >> | / 127 | 156 mA
> >> | \ |
> >> | | E
> >> +-----+--- B Q2
> >> | C
> >> \ R4 | --> 150 mA
> >> / 180 '------O with compliance
> >> \ (2) 2n4403 up to 4.45 volts
> >> |
> >> gnd
> >> The 850mV Vbe of Q2 is used with two resistors to offset the 700mV
> >> Q1 Vbe reference voltage down to 250mV or so across R1... Like the
> >> previous more complex design this simple circuit works over a range
> >> of supply voltage.
> > Apologies for the delay in responding. Thanks, this looks just what I'm
> > after! With 0.7V Vbe for Q1 and 0.85V Vbe for Q2 I calculate around
> > 200mA rather than 150mA (neglecting base currents), is this correct?
> Yes, thanks. I'll claim to have made a typo, 187 instead of 127 ohms
> (both are standard 1% parts). :-) See corrected values above, now
> includes beta calculations. I also changed the bias resistor R4 to 180
> ohms for better regulation. I used a minimum beta of 20 for Q2, which
> may be operating in saturation. At least 20mA through R4 insures that
> Q1, whose Vbe acts as a reference voltage, has 5 to 15mA of current, a
> reasonable 3x range, limiting its Vbe change to under 5% (AoE pg 91).
Thanks, I'll modify my circuit accordingly.