Controller Change slows down the PXIe-1062Q

[User002]

History: 10 years ago I built a PXIe-1062Q Chassis with a PXIe-8130 Controller running Win-XP. 4 PXI-6542 were installed.

              3 years ago I rebulit a similar system with a PXIe-1082 Chassis, an external Rack mount PC and 4 PXIe-6544 running Win-7

Now:    The XP controller was replaced with an external WIN-7 Rack mount (same type, same OP, same HSDIO 1.7.4., same LabView 2013 programm)

So far, so good, but:

• A testprogram, loading pattern into the on board memory of the HSDIOs runs on the old XP in about 200ms 
• On the newer system a bit quicker (as expected) but on the rebuilt system it tooks about 2 Seconds (only pattern load, without playing out) .... 😕

Does anybody has an idea what could be the reason for this behaviour? 

[santo_13]

Maybe for some reason, Win7 system is overloaded in terms of CPU/RAM usage and slowing down the automation?

Is the rack mount PC exactly same in HW specifications?

[User002]

Hello Santhosh,

yes, the PCs are the same and I changed then with each other -> no effect.

The strange thing I found by changing components was that the PXI-6542 card will go fast when it is mounted in a PXIe-1082 Rack instead of a PXIe-1062Q. But only when the hardware revision of the Backplane is K. In a PXIe-1082 with revision J the PXI-6542 will also work at low speed.

Firmware version of both Racks are the same.

!! When mounted a PXIe-6544 HS-DIO in all sorts of Racks they will work fast as expected. !!

So I found a configuration of both that will work for me, but how can I be sure, that in the future a spare part would also work when it is revision dependent?

[santo_13]

Definitely, there is a bandwidth difference 1062 (3GBb/s) and 1082 (8GB/s)

[User002]

Yes, I know,

but the quicker 1082 (8GB/s)  behaves slower than the 1062 (3GB/s) 😕

and the 1082 rev.J behaves slower than the 1082 rev.K 😕😕