Opto-coupler Question

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
rsfairman
Posts: 5
Joined: Sat Apr 03, 2010 10:09 am
Contact:

Opto-coupler Question

Post by rsfairman »

Is anyone able to recommend some fast opto-couplers appropriate for relatively high speed digital signals (say 100,000 pulses/sec)?

I'm nearing completion of a project, but I've hit a hurdle; I think that I'm using the wrong opto-coupler. I have a PC communicating with a CNC lathe over the parallel port. For those who know a bit about CNC, I'm using Mach 3 on my PC, and I'm trying to build a "break-out board" (yes, one can buy breakout boards, but this is a 25 year old lathe with some oddball innards and building the board myself is a learning experience). The parallel port is 3.3V, and the output of the opto-coupler will be 5V.

I'd like to use opto-couplers to isolate the PC from any errant high voltage coming from the lathe. My circuit works provided that I take the opto-couplers out of the system, but when I put the opto-couplers into the system, the output is either full on or full off, with no pulse detection. Here's the datasheet for the ones that I'm using now:

http://www.vishay.com/docs/83638/83638.pdf

These are Vishay part number ILQ66-4 (the quad-channel IC).

My guess is that the source of the problem is that the pulses coming over the parallel port are less than 5 usec long, but this opto-coupler isn't fast enough to respond. The datasheet is vague on the response time. The only numbers that I can find (3rd page) say that the rise and fall time could be as much as 200 usec, max, and that's obviously *way* too long.

Would this work instead:

http://www.vishay.com/docs/84732/6n137vo2.pdf

or maybe

http://www.vishay.com/docs/83604/83604.pdf

Thanks,
Randall
User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Opto-coupler Question

Post by MrAl »

Hi there,


If you can get away with the logic outputs which are different than transistor outputs then the
6N137 and associated parts on that data sheet look very good, as they should easily handle
100k bits/sec signal.
If you need a transistor output, then you would have to go with the 6N135 or 6N136, but they
will be slightly more difficult to use and their bandwidth is 10 times lower, although they might
still work ok with the right load resistor and possibly a base resistor.

A couple things to keep in mind using the opto couplers:

1. A specific load resistor is usually required for highest bandwidth.
2. A base resistor is often required for highest bandwidth too on devices that have such.
3. Important is the isolation between ALL the inputs and ALL the output pins: any pc traces or leads
or other components should be mounted far enough apart so that they can never touch the other
leads or components. The inputs are on one side of the package, the outputs on the other side and
that is done for a reason, so that the inputs and outputs have enough physical distance between each
other. Any parts connected to input or output leads must maintain this distance or the isolation is lost
which can cause some nasty problems like human shock hazard as well as blowing out expensive
equipment. That makes attention to pc board layout quite important as well as how the leads connect
to everything. Probably a good rule of thumb is to keep ANY leads or components that connect to the
input pins at least a quarter inch away from ANY that connect to the output pins, but a half inch even
better. For example note that on the 6N136 there are two input pins and four output pins to consider.
LEDs vs Bulbs, LEDs are winning.
rsfairman
Posts: 5
Joined: Sat Apr 03, 2010 10:09 am
Contact:

Re: Opto-coupler Question

Post by rsfairman »

Good info. I'll be glad when I've assembled this and it works.
Thanks.
Post Reply

Who is online

Users browsing this forum: Bing [Bot], Google [Bot] and 34 guests