From: Winfield Hill
Subject: Re: circuit wanted-SQUARE PULSE GENERATOR
Date: 26 Nov 2002 03:38:30 -0800
Organization: Rowland Institute
References: <1038144646.265208@athprx02> <email@example.com> <firstname.lastname@example.org>
X-Newsreader: Direct Read News 4.11
Tony Williams wrote...
> Winfield Hill email@example.com wrote:
>> ... internal resistance below 1V, so R4 and C2 provide a
>> fixed 25us time for thoroughly discharging C each cycle.
> I'm still trying to see why that 25uS discharge time
> doesn't result in a 2.5% non-linearity in the control.
> Knowing you, you have it covered somewhere, but I just
> can't see where. Must be going thick...........
No, Tony, you have found me out! It's unavoidably there, a
[1/(1 + 2.5% Vin/Vmax)] term in the frequency formula. :>)
The discharge could be shortened to 5 or 10us, but it'd still
be there. Fixing the issue would require a switched negative
tracking current sink. And the added complexity (two opamp-
controlled current sources, two steering diodes) would likely
force one to abandon the cmos 555 approach entirely, choosing
instead a opamp triangle integrator / comparator scheme, etc.
There's another fault to my circuit. The O.P. specified a
"square wave." If we assume he meant 50% duty cycle, rather
than simply rectangular logic pulses, the 25us negative-going
pulses my circuit puts out are not acceptable. On the other
hand, modifying the circuit with the above-mentioned tracking
current sink would also solve this issue.