temperature ratings for the NI SHC68-H1X38 for PXI-6551

[ajog]

Hello,

Does anyone know the absolute maximum operating temperature limits for the flying lead cable that accompanies the PXI-6551?  Its part no. is NI SHC68-H1X38.  I have reason to believe that this could be functioning abnormally below -25 deg C.  Please advise.  

Thanks.

Anand

Message Edited by ajog on 08-17-2009 11:17 PM

[Chris_W.]

Hi Anand,

What is the actual problem that you are seeing. I would think that the issues you are viewing are likely due to the operating temperature of the the 6551 and not the cable. If you take a look in the specifications document for the 6551 you'll notice that that the operating temperature for the 655x cards is 0-55ºC, so you are likely seeing some incorrect behavior due to the fact that you are far out of the normal operating temperature range of the card. 

Chris W

[ajog]

Thank you Chris,

The 6551 is placed outside an oven in which the DUT is placed.  The flying lead cable connects the two.  The PXI chassis is at room temperature.

We observe an incorrect READ below -25 degC.  I will try different cable channels for data i/o and see if there is any difference.

Does the flying lead cable also have operating temperature limits.  If so, what are these?  Please advise.  Thanks!

Anand

[Chris_W.]

Hi Anand,

There are no temperature limits for that cable that I am aware of where there is a degredation of of signal due to a working temperature of the cable. However, one thing that could possibly be causing some problems is the thermal barrier between the -25ºC of the oven (or wherever that temp. is being reached) and the room temperature of the second section of cable and 6551 in the chassis. The resistance of conductors do change due to changes in the temperature of the conductor and the thermal expansion/contraction of the conductor. I do not actually know how much of a change occurs, as I have not done any calculations, but this could be a cause for some error.

If the resistance of the cable on the room temperature side of the cable is much more than the resistance of the cable on the cold end, then there could be some reflection of the signal that is occuring, which would not allow the full amplitude of the signal to pass through that thermal barrier, which could potentially cause some false 0s in the data read by the 6551. I am not sure what the amplitude of the signals that you are generating from your DUT, but if they are something like 3.3V outputs, then this is worth testing for. If you can get your hands on a single flying lead that you have run from the DUT through the wall of the chamber that is -25ºC and check the amplitude of the signal that is being input to the 6551. This would show you whether the problem is due to changes in the resistance of the cabling.

More importantly, what do you mean when you say you "observe an incorrect READ below -25ºC?"

Chris W

[ajog]

Thanks Chris, very helpful. I will investigate this issue in greater detail now.

By incorrect READ, I meant that the DUT sends out a different bit pattern to a stimulus signal at -25 deg C than at room temperature using the same hardware setup (DUT, PXI-6551 etc.).  The problem could be on the DUT side, but I don't observe this when a non-PXI setup is used to communicate with the DUT.

Anand

[Ryan_M]

Anand,

A quick test you can run on the cable would be to disconnect it from your dut in the chamber and connect the end points in a "wrap back" configuration.  We ship a LabVIEW example called "Dynamic Generation and Acquisition-Source Synchronous.vi" which connects a source synchronous generation task to an acquisition task.

There's a note about how to run it:

"Before running this example, please use equal length cables to make the following connections
on the terminal block:
DDC CLK Out to Strobe
Trig 1 to Trig 2
Data 0 - 7 to Data 8 - 15"

You can scale this to larger channel counts and data patterns.  For example, I think it uses a count pattern on 1000 samples so it will only ever toggle 9 channels.  You may want to use a random pattern instead.