Cross Talk

[Trinity79]

Hello All,

I finishing the last bits of my program, but ran into  problem with SCXI 1160 though.
I'm currently using this switch in a matrix format, where I basically jumpered alot of th terminal block.
I running into some interference problems.  Basically, I have a scope that is hooked up to the Switch, and the other end hooked up to a FG.
Here's my problem, I can switch the relay, and I can validate that the signals are correct.  Upon further investigation, I noticed by checking the other channels I stil get a sinusoidal wave(for instance). Although this signal is not as strong as the one coming in.  However, its amplitude is about 25% of the original signal. Just curious if anybody has a solution to this problem.

Thanx

[Nick_Storton]

Hi Trinity79,

         What frequency are you using to check this out at? The higher the frequency the more crosstalk you will get. From the spec sheet this switch offers better than -25dB channel to channel crosstalk @ 10MHz, This equates to about 6% of the original signal. The crosstalk spec also reduces by 20dB per decade so if you are using around 50 MHz this will give you about -15dB which is approx 25% of the original signal. It is possibe that this level of crosstalk is mostly due to the wiring of the terminal block, in which case you may need to re-wire using coaxial cable.

Hope this helps,

       Nick

[Guy Ferraro]

The SCXI-1160 is an SPDT general purpose relay that can switch medium power signals and was not designed for optimal crosstalk. Therefore we do not have any crosstalk specification pertaining to the SCXI-1160. The crosstalk specification given by Nick is probably the one fro the SCXI-1166.  

Since you are using the SCXI-1160 in a matrix configuration you are probably interconnecting a lot of SPDTs with wires crossing each other in the terminal block. All these wires crossings contribute to the crosstalk. What I would suggest is checking you wire layout and making sure wires are isolated from one another and grounded connected to the chassis ground. If you are switching differential signals make sure the pair is twisted, shielded and grounded as well.

Guy Ferraro
Switch Product Support Engineer
NI-USA 

[Trinity79]

Thanks for your response.


Nick,

I originally thought that the frequency was the issue, but I checked by operating at a low frequency at about 1K Hz.
There was still some interference, although not as large as when the frequency was > 1MHz.

Guy,

I feel that the problem is what you suggested.  I am switching a differential signal. Let me reiterate what you suggested:

1) Ground the incoming signal to the ground on the terminal block.
2) Use STP wiring in the terminal block.

BTW, I have already have done a iso & cont. check on the terminal block with the wiring, and it seems like it was okay.

Another question I would have, is how strong of a interference would we have if the suggested method is used?


Thanx