I get too much noise and cross-talk between input and output with usb-6009

[mtngirl]

I am re-creating a sensor that was originally programmed with a TXI tattletale A/D board.  I am using a NI USB-6009 to output 4 digital signals that go through a power board that turns on/off a 12V signal to 3 pumps and a board with 3 photodiodes.  The USB-6009 also collects analog (voltage) signals from 3 photodiodes.  My system sees >10X the noise (with signal averaging to 1 sec) of what the original system saw in the analog inputs (I get ~20 mV noise at 1000 Hz, which averages out to ~2mV noise at 1 Hz; I need ~0.1-0.2 mV noise at 1 Hz).  I also get cross-talk between digital outputs and analog inputs, e.g. with I am pulsing a solenoid pump (0.15 sec on/0.35 sec off), I see ~0.5 sec waves (~30 mV amplitude) on my analog signals. 

What is the best way to eliminate the cross-talk and reduce the noise?

I also have a DAQ 6062E and BNC 2110.  Would this perform better as far as noise since it is shielded and accepts shielded cable?  What about the cross-talk? 

[AnalogKid2DigitalMan]

What was the resolution and voltage range of the TXI board? The 6009 has 13 or 14 bits of resolution depending on whether single ended or differential mode is used. Could be the 6009 has more resolution and is thus detecting noise that the TXI could not resolve? The 6009 also has low input impedance (not usually a good thing) when compared traditional DAQ cards- only 144Kohm. Could this be affecting your photodiode signals? What amplitutude range are the photodiode signals and have to set teh 6009 to the proper range?

Regarding cross talk- are your analog signal wires routed away from other noisy electronics- pumps, valves, and all other wiring? Have you tried differential mode on the analog inputs on the 6009? Are you analog signal wires shielded? Proper grounding techniques followed?

Sorry for all the questions, troubleshooting noise issues in someone else' system remotely can turn into a can of worms

[mtngirl]

I have not completely worked out my noise issues, and have some related questions.

1) I want to be sure that I have the system correctly grounded:  I have 3 analog signals coming from a circuit board (our analog board; all signals are tied to the same ground).  In RSE mode, I have the ground from the analog board connected to the left-most ground slot on the analog side of the USB-6009.  I then have the signals connected to AI0, AI1, and AI2.  Is this sufficient or do I need a ground for each signal?

2) I have also added to my system a YSI thermistor.  I am measuring the V(in) and V(out) from the thermistor at AI6 and AI7 without an additional ground (the V(in) signal comes from a circuit board (our power board) which is grounded to the analog board which is grounded to the USB-6009 as mentioned above).  Do I need to ground these inputs separately?  Are all the ground slots on the USB-6009 connected?  If not, it is unclear to me which ground belongs to which signal input.

3) If I were to use differential mode, would I wire each signal plus the ground to the + and - terminals?  e.g. would I wire signal 1 to AI0+ and ground to AI0-?  Then what is the purpose of the ground slots?

4) If I were to use differential mode for my first three analog signals, could I use RSE mode for my thermistor V(in) and V(out)?

Thanks, Sorry for all the questions  

[Laura F.]

Hello,

---

1) I want to be sure that I have the system correctly grounded:  I have 3 analog signals coming from a circuit board (our analog board; all signals are tied to the same ground).  In RSE mode, I have the ground from the analog board connected to the left-most ground slot on the analog side of the USB-6009.  I then have the signals connected to AI0, AI1, and AI2.  Is this sufficient or do I need a ground for each signal?

I would not use RSE mode if your analog signals are considered a grounded source (which it seems like they are).  You could be introducing ground loops.  For more information, please check out the tutorial Field Wiring and Noise Consideration for Analog Signals.

2) I have also added to my system a YSI thermistor.  I am measuring the V(in) and V(out) from the thermistor at AI6 and AI7 without an additional ground (the V(in) signal comes from a circuit board (our power board) which is grounded to the analog board which is grounded to the USB-6009 as mentioned above).  Do I need to ground these inputs separately?  Are all the ground slots on the USB-6009 connected?  If not, it is unclear to me which ground belongs to which signal input.

The inputs do not need grounded separately.  All analog grounds are connected inside the USB-6009.

3) If I were to use differential mode, would I wire each signal plus the ground to the + and - terminals?  e.g. would I wire signal 1 to AI0+ and ground to AI0-?  Then what is the purpose of the ground slots?

Yes.  The ground connections are for the RSE measurement configurations.

4) If I were to use differential mode for my first three analog signals, could I use RSE mode for my thermistor V(in) and V(out)?

Yes.  You would use all of your analog input channels - 6 for the three analog signals and the remaining two for each of the thermistor measurements.

---

Hope this information helps!

Laura

[AnalogKid2DigitalMan]

Laura:

Can you mix differential and RSE configuations on the 6009? I was under the impression it had to set all channels to one or the other.

[Laura F.]

I know that I can mix channel configurations under NI-DAQmx 8.3.  Perhaps it was different for older driver versions...

Regards
Laura

[mtngirl]

I tried differential programming, and I do get lower noise, on channels AI 1 and 2, but I get the incorrect voltage readings on channels AI 0 and 3.  To get to the source of this problem I put a direct 3.5 V current in from a power supply, and found that in differential mode, channels AI 0 and 3 give oscillating voltage readings between 2.5 and 3 V.  In RSE mode, channels AI 3 and 4 give incorrect voltage readings (2.7 and 2.8 V, instead of 3.5 V), which explains why differential channels 0 and 3 do not work.  This DAQ board is new.  What could cause it to malfunction??

[Laura F.]

Hi mtngirl,

It seems to me like you could be experiencing some sort of grounding issue.  When you measure the power supply using AI 3 and AI 4 in RSE mode, are all other wires disconnected?  If not, how are the rest of the channels connect to the USB-6009?  Can you ground AI 3 and AI 4 and see what happens?  Do you experience any oscillating? 

Answering these questions will help us figure out what is happening.

Thanks,
Laura

[mtngirl]

When I measured the voltage from channels 3 and 4 in RSE, I used a grounding slot to ground to my analog board, sam procedure as for the other 6 channels.  Nothing else was connected to the analog side of the board, but I do have digital connections.

[Laura F.]

Hello mtngirl,

If you can take that same voltage and apply it in RSE configuration to another channel and it reads correctly, but incorrectly on channel 3 or 4, it seems that those two channels could be damaged.  You should contact our technical support department to determine if you should send the device in for repair. 

Regards,
Laura