Maker Pro
Maker Pro

Low quiescent xtr coil driver

J

James Karmen

Jan 1, 1970
0
I am looking for a design for a low quiescent CCS using only discrete
transistors (no op amps) suitable for driving a magnetic relay coil at
a frequency of several tens of Hertz. Input is a 4V sinewave direct
from a PIC micro. Yes, it is not a squarewave in this case.

This seems to suggest a darlington pair, with perhaps an FET being the
first stage.

Can anyone advise what are the design considerations that will result
in an absolute minimum quiescent current as would be suitable to a 9V
battery powered device? And, if possible, provide an example circuit.

Thank you for your suggestions.

James Karmen
 
E

Eeyore

Jan 1, 1970
0
James said:
I am looking for a design for a low quiescent CCS using only discrete
transistors (no op amps) suitable for driving a magnetic relay coil at
a frequency of several tens of Hertz. Input is a 4V sinewave direct
from a PIC micro. Yes, it is not a squarewave in this case.

This seems to suggest a darlington pair, with perhaps an FET being the
first stage.

Can anyone advise what are the design considerations that will result
in an absolute minimum quiescent current as would be suitable to a 9V
battery powered device? And, if possible, provide an example circuit.

Thank you for your suggestions.

Since when does a CCS have a 'quiescent' condition ?

And how do you propose to 'drive' one ?

Your post suggests to me that you really don't know what you're talking about.

Graham
 
S

skenn_ie

Jan 1, 1970
0
One of my favourite devices for driving directly from a uC is the BST72
from Philips.
 
T

Tom Bruhns

Jan 1, 1970
0
Eeyore said:
Since when does a CCS have a 'quiescent' condition ?

And how do you propose to 'drive' one ?
I just love when people use TLA's (Three Letter Acronyms) without
defining them.

CCS: Coast Clutch Solenoid
Combat Control System
Common Channel Signaling
Common Command Set (SCSI)
Common Communications Support
Continuous Composite Servo
Hundred (C) Call Seconds
Hundred Call Seconds [Telephony]
Simon Bolivar airport (code) [Venezuela]
Or, just maybe, "constant current source."

OK, back to the OP (original post). Why? Why would you want a
constant current source (switchable, on and off, presumably)? Why
would you not want to drive the relay coil, especially if you want that
sort of speed, with something a bit more aggressive that would SAVE on
power? HOW? How is it that a sinewave can come _directly_ from a PIC?
Does your PIC have a built-in DAC? WHERE? Where on the sinewave
(assuming you really do have to deal with one) do you want the relay to
pull in, and where do you want it to release? How well do you know the
characteristics of the relay you'll be driving? Can it really follow
"several tens of Hz"? How long will it last at that rate?

If you have control over what goes on inside the PIC and you have even
just one pin available, you can probably do a lot better letting the
PIC drive the pin with a PWM (pulse width modulated) signal that will
pull the relay in fast and then go to low average holding current.
Then a simple mosfet would work fine to drive the relay coil, with very
few additional parts required: just an appropriate snubber across the
relay coil. You may be able to do even better if the relay drive
frequency isn't arbitrary but is some constant value. And finally, if
you really want to minimize power, you may be able to do even better--a
lot better--with a particular type of solid state relay, depending on
what the relay output needs to drive.
Your post suggests to me that you really don't know what you're talking about.

Graham

Yes, you may be right, Graham. Or at least doesn't know enough yet to
be asking such specific questions. It reminds me of the guy who comes
into the shop with something behind his back and asks to use the drill
press. The "something" turns out to be a thin piece of sheet metal,
and he wants to put a round half inch (or 1.2cm) hole in it, and next
to the drill press is a nice Rotex punch with a good assortment of
punches in it, including the right one.

If we know just what you're trying to accomplish, we have a better
chance to suggest optimal ways to do it.

Cheers,
Tom
 
Top