What am I doing that causes USB-2008/2009 channels to die?

[mtngirl]

I frequently lose 1 channel on USB-2008/2009 boards and need to have them repaired.  I always disconnect the DAQ from my computer and power source when removing or installing wires.  I re-connect it to the computer while the computer in ON.  Is this a problem?  I also often leave the DAQ connected to my computer when not using it.  Anything else that might be causing damage to the DAQ?

[AnalogKid2DigitalMan]

Hi mtngirl:

Did you mean 60008/6009? Nothing you mentioned should be a problem. What is that 1 channel connected to? I have a hunch it may be voltage spikes or high current loads killing the channel.

-AK2DM

[mtngirl]

Yes, I did mean 6008/6009 DAQ.  I am using all 4 differential AI channels, 3 of which are connected to amplified photodiode signals (max 5V) from a custom analog board, and 1 is connected to a thermistor/voltage divider and follower circuit to increase input impedence; the thermistor is powered by the +5V from the DAQ.  I also use 2 AO channels, which are amplified and used to run peristaltic pumps.  All circuit boards are powered by Shenzhen Mastech DC power supplies. Who can I contact to find out whether my DAQ is still under warrantee?

[John_P1]

You should contact National Instruments to check the status of your warranty or if you want to send the boards in for repair.  However, we still would like to know what's causing the damage to prevent this from happening again in the future.  Is the problem with your analog input channels or analog output channels?  The AO channels on the 6008/6009 are capable of sourcing 5mA per channel, while the 5V output on pin 31 can source up to 200 mA.

-John

[mtngirl]

The problems are with the AI channels; it is a different channel each time.

[John_P1]

How do you have your signal source connected and grounded?  Are you using RSE or Differential?  It sounds like there is some sort of issue with your input source that is causing this to happen repeatedly.  Could you attach a digram of your connections including part numbers and any other relevant information?

In the meantime, I recommend that you Contact NI by phone to check the status of your warranty and arrange for repairs if necessary.  You can also speak with a support engineer who can help further troubleshoot your issue.

-John

[mtngirl]

I use differential programming.  The schematic for the "analog board" that I use is attached.  My board is modified in that it does not have the regulator circuit shown; instead I use +/-13V power supply. 

The amplified signals from 3 photodiodes (J1: 2, 3, 4) are attached to the NI AI0+, AI1+, AI2+; the ground from the analog board (J1: 1) is split 3 ways and attached to the NI AI0-, AI1-, AI2-. 

The +5V signal and ground from the NI are attached to a follower circuit (with OPA27GP) and resistor (10K) which go to a thermistor; the output signal from the thermistor go to AI3+ and AI3-.

The AO0 and AO1 signals go to separate amplifiers (OPA551PA) that control 2 peristaltic pumps; the bottom ground on the NI board, left side, is split 2 ways and connected to the input terminal of these amplifiers.

It is always the AI channels that I have problems with, but is is a different channel each time.    

[John_P1]

Hi mtngirl,

Could you clarify what you mean when you say a channel goes bad?  What does the data look like coming in on a bad channel (i.e. is it 0 volts, does it rail to +10V, or some other behavior)?  When a channel goes bad, are you able to take RSE measurements on it (or its negative channel pair)?

-John