I really like the idea of the large cap. Because I need the vibrator to
continue to work even when trying to draw more then 5A the current limiter
would be the best answer I think. The whole thing is being supplied by a
switching 12VDC 5A powersupply. If I were to allow the vibrator to draw more
than 5A it would cause the switching supply to shut down. That would cause
the F877a to also shutdown. I can work around it but it would just be more
convienient to make it so that everything continues chugging along even in
an overload state.

Thanks,
James

----- Original Message ----- 
From: "Russell McMahon" <apptech@paradise.net.nz>
To: "Microcontroller discussion list - Public." <piclist@mit.edu>
Sent: Tuesday, March 08, 2005 4:02 AM
Subject: Re: [EE] Best method of current-limiting a PWM circuit?


> > I'm driving a logic-level N-EMOS with the PWM output of an F877a at
> > 20kHz to controll a mechanical vibrator. I want to limit the current
> > the load can draw to 5A and I can't think of a good way of doing it
> > other than a big-fat resistor in series. One complication is that
> > the coil resistance can vary. Can anybody point me in the right
> > direction?
>
> If you feed the vibrator with a 5 amp current source it can draw LESS
> than 5 amps under PWM control but never more. If you want you can put
> a suitably large capacitor on the output of the current source so that
> you can source peaks greater than 5 amps but not more than 5A mean
> once the capacitor discharges and the current source starts supplying
> the full load.
>
> A current source can be as simple as a power transistor, a zener and a
> few resistors.
>
> The main problem is that it will dissipate significant power if the
> solenoid goes short circuit so some form of monitoring is desirable.
> If this is 12V then 5A = 60 watts. 2 x LM350 will also do this with
> thermal protection, depending on the voltage involved. Heat sinking
> may be necessary depending on how you want it to behave under
> overload.
>
> Simple but probably adequate transistor current source.
> Q1 PNP suitably rated.
> Emitter V+ via Rsense
> Base to V+ via zener
> Base to ground via resistor Rbias suitably dimensioned.
> Current limit =~~~ (Vzener-0.6)/Rsense.
> No thermal limiting.
>
> Keep Vzener low enough that power in Rsense is not excessive.
> Vzener CAN be two forward biased silicon diodes in series.
> Ilimit then become =~~ 0.6/Rsense.
> 5A Rsense = 0.6v/5A = 0.12R
> At 5A, power in Rsense is about 3 watts.
> Power in Q1 will depend on how much "excess" voltage there is
>
> You COULD arrange instead for a PWM cycle to be tripped off early in
> each PWM cycle whenever current exceeds 5 amps. This way there is
> minimal dissipation in the control circuitry.
> The trip off could be processor instigated OR use external hardware.
> This may not achieve what you want as current then drops to zero for
> remainder of that PWM on cycle whereas you MAY want it to limit at a
> steady zero. Depend son what you are trying to achieve.
>
> Tell us more and we'll tell you more :-)
> What voltage?
> Why do you want to current limit?
> Trip or maintain?
> ... ?
>
>
>
>         RM
>
>
> -- 
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist