I have a question about sending high-speed signals using point to point wiring. By the way, I found your seminar to be most informative and inspiring.
Here is my situation:
The rise time of my signal at the output of my RS- 422 receiver (CMOS 5V) is 3 ns. The signal then travels through a 75 ohm series termination, then about 1-inch to a connector pin (on wire wrap wire), and then about 10 inches on 24 gauge wire to the input of an industrial counter. The input of the counter is a transducer with 243K-ohm input. My ground reference for the counter comes from the receiver board through the connector on the board and to the counter input - basically the same path as the excitation signal. My signal at the input to the counter rises to about 1.6 V and then shoots down back to 0V and then makes a gradual ascend to 5 volts with a 600ns rise time.
Since the rise time is so high I suspect that the counter is heavily loaded capacitively - as it is meant for standard industrial slow speeds. I would like to know how to get rid of this initial spike.
Any suggestions would be helpful. Is the best way to convert a strong driver to a weak driver (other than change the chip to a weaker logic family) simply to put a series resistor at its output?
Thank you very much.
Thanks for your interest in High-Speed Digital Design.
If you can tolerate the slow-speed risetime at the counter, what I would do is place a shunt capacitor after the 75-ohm resistor but before you enter the transmission line. This will slow down the signal enough to eliminate the irritating double-bounce at the end.
Your other alternative is to use a real 75-ohm coaxial cable (you can get miniature coax with a diameter as small as 0.05"). If you remember from class, a series-terminated transmission line can drive a big capacitive load fairly well. Whatever bounces back off the load comes to the driver and dies at the source termination. Your problem right now is likely that the characteristic impedance of the 24" wire connection is much higher than 75 ohms, so your resistor is not serving as a good series termination. If you make the resistor large enough to function (a couple hundred ohms) you will just further slow down the ultimate risetime at the load.
Your best choices are to either slow down the driver risetime a little bit so the whole thing acts as one big lumped-element circuit, or use a real 75-ohm transmission line.
Dr. Howard Johnson