SCXI-1128 Long Warm Up

[ecbarnum]

SCXI-1128 32 Channel Solid State Mux driving SCXI-1125 isolation amplifier being read by PCI-MIO-16XE-10 Data Acquisition Card.

We use the mux cards to scan approx 50 inputs, 32 on one card, balance on other.

When powered up at room temperature it takes several minutes for the system to start reading right. 

When powered up at cooler temperatures (say 50 deg F) it has taken as long as 45 minutes to start reading correctly. 

While it is warming up it looks like the multiplexing card action is not coordinated with the DAQ card reads.  You can see various voltages flashing on screen common to various sections of the 50 signals travelling thru different mux cards paths.   Eventually there will be brief flashes of correct readings and then finally the bad readings stop appearing and all is well from then on.

However, seems like solid state system should not need to warm-up.  Any ideas on what the problem might be?

[Logan_K]

Your setup is not a standard setup, so I have some basic questions about your system because I am not clear on how you are using the SCXI-1125 and SCXI-1128 together:

How are you connecting the SCXI-1128 to the SCXI-1125?
Are you scanning just one channel on the SCXI-1125 which is connected to the output of the SCXI-1128 MUX? Or, are you scanning multiple channels?
How are you synchronizing the SCXI-1128 to the DAQ/SCXI measurements?
Finally, right now does the problem look like it is in the SCXI-1128 or somewhere else?

Regards,
Logan K

[ecbarnum]

I learned that the mux's are acting independantly- 32 channels combined by the first 32 ch mux goes into one channel of the isolation amp and the remaining 10 channels combined by the second 32 ch mux goes into a different channel of the isolation amp.   Then the separate isolation amp outputs are routed to separate channels of the PCI-MIO-16XE-10 card. 

So it could be the mux or the a/d card that is slow to start working since both would have same symptom: if the mux wasn't switching the card would read all channels the same, or if the card was scanning without waiting for the mux to switch all the channel readouts would be the same.

See "scanadvd" wires and "ext trig in" wires on schematic.  What's basic principle of synchronizing the mux card switching with the A/D card sampling? 

A/D card's PFI4 pin is connected to multiple (three) mux scanadvd and a/d card's GPCTR1 drives multiple (three) mux's ext trig in's.  Can these be temperature sensitive?

[GDE]

I’m curious, if you don’t use your multiplexing scheme, and just read in one channel through the 1125 do you notice the same “warm-up” timing issues. It would be good to rule out the possibility that it is the 1125 that is taking the time, and instead is something with the multiplexing setup.

-GDE

[ecbarnum]

When its wrong it is way wrong - for example normally each channel has its own steady and distinct reading around 4 volts.  When initially warming up it'll mostly read all identical 0.0's or sometimes flash identical 1.3's on all channels with intermittant hits of actual data.  Although now that you mention it, if the isolation amp was messing up I guess that could create the problem just as easily as the mux or scanner messing up.  Looking like I'll need to get an oscilloscope to get to the bottom of it...