Robert Rolf writes:

> Kenneth Lumia wrote:
> >"Morgan Olsson" wrote:
> >>Serialports is possible and easy to optoisolate.
> >>For the current project i can not use USB in final test as the PIC
> >cirquitry is on high voltage.
> >
> >Actually, it is relatively simple and not much more expensive to isolate 
> >the
> >USB
> >port.  I did this on a project for a Circuit Cellar competition and it
> >proved
> >rather easy.  Take your USB and use a FT232BM chip to convert USB
> >to serial.  Add 2 optocouplers for TX and RX and connect the other side
> 
> Then you are NOT isolating USB, are you? You are isolating
> a MUCH slower serial signal. By that reasoning you could use
> Ethernet<=>USB modules and call it 'isolating USB'.
> 
> >to the projects serial port.  In my case, it was an AVR, however a
> >PIC would be similarly wired.  If you lay it out correctly, the worst
> >case isolation is the optocouplers rating.
> 
> Or you could use damn fast opto's like those that HP optical make
> (whatever they're called now) and USB transceivers to get to/from
> differential drive.

Would the GMR's described in Jeff Bachiochi's article in CC#142
"You're Not Alone Dealing with Isolation" be apropriate for this use?
They claim 2500V isolation and 100 MBaud data rates.

http://www.circuitcellar.com/library/print/0502/JEFF/index.asp

The raw devices are not more than a few bucks, and the il-485 (rs-485)
interface claims 35 Mbaud and goes for < $10, so it seems likely that
USB 1.x could be implemented rather easily and cheaply.

-p.
_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist