SCXI-1102 T type Tc °F shift adjacent channel connection

[Scottyd]

SCXI-1102, Type T Tc, channel reading shifts up to 3 °F when Tc connection to an adjacent channel is made. Issue began with a Dell mother board that had capacitor damage. The PCI video card and PCI-MIO-16xe-10 were damaged, but the replacement has this issue. This is a stand alone application, virtually (calibration contants are different) the same software is running with no issues on 4 other test stations.

Did the SCXI-1102 get damaged as well? Could the pulse from the capacitor overload damage the SCXI-1102 amplifiers? We tried a laptop and a 6024E connected to this chassis and it showed it had the same shift issue, so the SCXI module appears to be the component damaged, yet when a different newer "loaner" PCI-MIO-16xe-10 was installed into this computer the shift didn't occur.

[Logan_K]

When channels on the SCXI-1102 are left disconnected, then ghosting of the large open thermocouple signal from one channel to an adjacent channel is a common problem. The root of the problem is the settling time of the connected data acquisition board, and the nature of the problem can change from board to board; which might explain why some DAQ boards exhibit the issue and some don't.

Take a look at this knowledgebase for details on investigating this issue further.
http://digital.ni.com/public.nsf/websearch/C9D0BD68E7CB7D6F862570760057F72D?OpenDocument

Logan K
SCXI R&D
National Instruments

[Scottyd]

Logan, thank you for the response.

I am having some difficulty selling my Boss with the explanation of the PCI card settling time being "suddenly" the issue. 

4 of the 5 test stations with the same clock speed, sampling rate, and software application do not have this issue.

Is there another possible cause for this within the SCXI-1102 hardware, IE: if it was damaged by the computer motherboard capacitor failure? Could all of the SCXI-1102 amplifiers be overloaded or damaged? Could that pulse have damaged the module circuits that convert mV to V?

Thanks Scottyd

[Randall_T]

Scott,

 I hope you have your problem solved by now. I am sorry for weighing in on this so late. I was searching the discussion boards and ran across this posting.

I have experienced this when reading ground-referenced thermocouples. In this case, I was using the scxi 1303 terminal block. To resolve this I substituted a 10M-ohm resistor for the factory shipped 10-ohm resistor in the Bias Resistor network. The 10-ohm is only good for floating signals; most thermocouple I use will connect to a grounded test unit.

[Scottyd]

We used and are using 10 M ohm resister packs.

We also switched packs from another TBX1303, just to see if there was a difference, there was none.

it is still an open isssue.

[Jennifer O]

Hello Scott,

I noticed that your temperature offset has not yet been resolved.  I was hoping to get some information from you to look into this a little more.  Would it be possible for you to post the revisions of the different boards that you have been working with (the two PCI-MIO-16xe-10 boards and the DAQCard 6024E)?  This information is noted in the part number by the letter preceding the dash (-). For example, 183742H-01 would be revision H of the PCI-MIO-16xe-10. The part number is generally found on a small white label affixed to the board.

Are you still open to troubleshooting this issue?  Does a decreased scan rate or increased interchannel delay make any difference in the size of the offset? Are you currently using any work-around?

Regards,

Jennifer O.

[Scottyd]

PCI-6030 board (s/n 11E1F7E)? 

old PCI-MIO-16XE-10 boards (s/n AD9C97) 

 loaner PCI-MIO-16XE-50 boards (s/n 1058A9A). 

SR# 7115230.

Their old 16XE-10 failed to work, so they quickly bought the 6030E for short-term replacement so they could keep running while they RMA'd the failed one. It turns out that the 6030 had some other measurement offset issues that only showed up with their SCXI systems and after being RMA'd, so did the repaired 16XE-10. The only board we could get to work in the system was our loaner, PCI-MIO-16XE-50.

This was the guy working on the solution.

Justin Mintzer

Applications Engineer

National Instruments

this person may be able to help you with the rev level issue, the boards are at NI.
Scott Diedrich

Manitowoc Ice Inc.

[Jennifer O]

Hello Scott,

Thank you for the information.  I looked into the SR to get complete details on the case.  I noticed that you had a few working systems while just one exhibited the voltage offset. Did you ever rule out the possibility of there being a problem with the SCXI system and cabling by switching only the SCXI system(and cabling) between a working and non-working system?

You mention that this is still an open issue and I want to make sure you have a long-term solution.  Please let us know if there is anything we can do .

Regards,

Jennifer O.

[Scottyd]

Yes  we switched TBX1303'S, cable's, modules, connections at SCXI, PCI slots..etc Same O BS

even whole chassis, Same stuff

two different systems had issues,   #4   #5

5 test stands running fine, two systems had smoked cap on pc's video cards, then problem appeared.

other 3 still OK

change to the different style board, OK

change back "repaired' board, Bad

Install "new" board, Bad

Intall different (borrowed board), OK

?????

?????

all systems running now with 2 different boards in #4 and #5. had to change software to reflect sampling rate of the different boards to make them run within their spec.

Scottyd