A thought...

Chuck Tomlinson tomlinsc at ix.netcom.com
Wed Oct 2 04:10:05 GMT 1996


> From: Arnaldo Echevarria <aec at ao.net>
> 
> I'm thinking, instead of going the 68hc11 route, using a pc.  Instead of
> building a a/d and d/a board, use two parallel ports.  That would give me
> 8 ins and 8 outs.  Timing would be resolved by the computer, hell, 386/486
> motherboards are cheap.
> 
> Question 1: Am I crazy? Anyone see an inherent flaw in all of this?

I've done a fair amount of PC-based real-time control.  The biggest 
obstacle you'll encounter is boot time.  Even with a flash disk, normal 
PC BIOSs waste precious seconds checking everything umpteen times.

If you don't mind waiting a long time for your engine to run, it can be 
done.  My experience is in chassis control.  The control loops were 
written in C, and ran from DOS or Windows.  We use fairly expensive 
Analog Devices (RTI-8xx) boards that interface to custom drivers.

Our control loops never needed to be faster than 1,000 Hz, so I don't 
have much experience with faster loops.  In pure DOS, it's a simple 
matter to reprogram the system clock interrupt to occur at 1.0 kHz 
(it's normally set to run as slowly as possible, at 18.2 Hz).  

Windows 3.x provides interrupt services with a 1.0 ms resolution.  
We do a lot of work on 486s using the minimum 1.0 ms interval and 
never had any latency problems.

In practice, waiting for a PC to boot is a pain in the ass.  For 
damper control, the car is uncomfortable until the code loads: no big 
deal.  But for brake control, you want that sucker loaded and settled 
before the engine starts.  We don't use PCs for brake control :-)

> Question 2: I need a good power supply to give me +12,+5,-5,-12. Should I 
> pull the car's power, use a 12/24 volt transformer, then get a couple of
> voltage regulators to give the voltages I need? Car voltage lines are 
> extremely noisy, any ideas on how to clean the signal?

We run our PCs from DC-AC inverters, mainly because we use the PCs for 
other tasks.  Electrical noise has not been an issue with newer inverters.

> Question 3: Any suggestions, ideas, or people interested?

If you can find a solid PC BIOS that boots in less than a second, I'd be 
interested in it.

--
Chuck Tomlinson <tomlinsc at ix.netcom.com>




More information about the Diy_efi mailing list