> There are a lot of parts there...

Actually just a single (and *very* cheap) ULN2003A and some passives. If
I swap the 15V zeners for resistors the passives is just 2 x 4-resistor
networks. The 1N5819's would be needed anyway, unless the body-diodes
can be trusted, which I understand is not a good idea. And they are
dirt-cheap too.

> 1. SINK level: you might look at "logic-level" components. 
> That way, you drive them
> just like you drive a logic chip (in SINK mode) and it reduces the 
> design costs and
> effort. Drive straight from the PIC, maybe through a 1K resistor to 
> isolate the PIC
> from motor spikes.

I selected these FETs because I can get them easily and cheap. The
logic-level FETs I have seen are either more expensive, or have lower
current capacity, or a higher RDSon (or a combination).

> 2. For the DRIVE level components- again, use LOGIC-LEVEL 
> devices but drive
> them with your ULN2003 or a transistor. The pulldown voltage resistor 
> can be very
> high, 10K or so.

I don't see the advantage of a logic level FET? I have the full 24V
swing anyway, I need to reduce it to comply to the max gate voltage! I
guess that would be even lower for a logic-level FET.

> 3. I use a bi-directional transorber across the load to protect from 
> transients. What
> you have won't stop motor spikes.

Probably a good idea, but as yet I am testing with a 2k2 + 2
antiparralel LEDs as load. I don't think they will generate much spikes
:)

> ULN2003 runs warm always.

Well, that chips does not run warm. It is the bloody P-FET that is
supposed to be switched of :(

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: www.voti.nl/hvu
 

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