From: "Spehro Pefhany"
Subject: Re: V regulator input cap size?
References: <email@example.com> <0001HW.B9D9CD54001A5D15163F7590@news.covad.net> <0001HW.B9DAE8B7003F6E14163F7590@news.covad.net> <3DC1A954.firstname.lastname@example.org> <1%hw9.136243$Q3S.email@example.com> <3DC1AF48.firstname.lastname@example.org> <3DC1C89E.email@example.com> <firstname.lastname@example.org> <3DC1DBEA.email@example.com> <firstname.lastname@example.org>
User-Agent: tin/pre-1.4-19990216 ("Styrofoam") (UNIX) (Linux/2.2.14 (i586))
Date: Sat, 02 Nov 2002 11:17:55 GMT
NNTP-Posting-Date: Sat, 02 Nov 2002 06:17:55 EST
In sci.electronics.design John Fields wrote:
> I've deleted the current limiting resistors for the LED's since
> they'll probably be driven by the 555's, but it shouldn't matter
> because they'll be on the 555 outputs and the current for them will
> have to be supplied by VLOGIC in any case.
I suggest open collector drivers allowing unregulated power to them, and
keeping the shunt regulator current down. Either the 200x driver I
suggested or "digital" transistors would work nicely. I doubt the
zener will have to be more than a single 500mW or, at most, a cheap 1W
unit. By keeping the series resistor high in value the shunt regulator can
withstand huge transients (MOF resistor, preferably) without requiring
expensive TVS barnacles, inductors or other such stuff. I'd also prefer to
keep the Vreg current as constant as possible, as 555s are sensitive to
voltage variations within their timing cycle.
If the LEDs actually alternate (one on/one off always) it's even possible
to use the LED current for the zener regulator, at some expense in the
(BTW, I'm NOT going to do a detailed design, the information
is too sketchy, my aim was to show a concept not to complete a design, and
I don't have time right now to make arbitrary assumptions and labor to
satisfy the resulting constraints. The process is obvious to anyone
"skilled in the art" as the patent guys say, and we've shipped a lot of
product with similar designs, so *I* know it works very well indeed).
> Also, I've added R2 to
> limit the current through the 12V relay coil. (A slight oversight
> on Spehro's part, I'm sure!-)
Not an oversight, John. I recommend changing the relay for a 24VDC one by
using a very small filter cap for the relay coil eg. 22uF/50V (NONE will
actually work with virtually all DC relay designs). A resistor was
mentioned in the text, but I don't think it will be required. A 24V coil
relay will cost about $0.005 more, in 10K quantity, than a 12V coil relay,
but I think it is worth it to keep all the power in the coil. ;-)
Fred: Sorry about changing the "rules" in order to get an elegant and
reliable design that can be produced cheaply.
(rest of John's detailed analysis deleted for brevity)
"it's the network..." "The Journey is the reward"
email@example.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com