NI USB-6009 Problem: 1.4V measured at open input with DAQ Assistant

[johanneshoer]

Hello,

I use one NI USB-6009 device to measure voltages. I have connected the device via USB with the PC and all inputs and outputs of the device are open - nothing connected.

If I place a DAQ Assistant block in a vi open it and run the assistant it shows me voltages of around 1.4 V at each input. In the vi in a display wired to the DAQ Assistant output, it shows me 1.4V as well.

If I go into MAX and run the Testpanel of NI USB-6009, it shows voltages of around 0V.

What is the problem with DAQ Assitant? Why doesn't the Testpanel show the same voltage as the DAQ Assitant? (see attached pictures)

Another strange thing is: If I put pull down resistors at all open inputs, the measured voltage in the DAQ Assistant is reduced to almost 0V.

[Hillman]

Hi There,

Good Morning and I hope your well today.

I would not recommend leaving your inputs floating (disconnected). This will cause the voltage reading to fluctuate and be random. Your tests aren't accurate, as in the time it takes you to change from Testpanel to code (or vice versa) the floating value could very useful change. 

It is not strange that if you wire the inputs to ground, that both the testpanel and DAQ code read 0 volts, as this is the value at the inputs to the DAQ device. This would be the best way to test if they were reading the correct values. You could also test your code by connecting a signal generator and see if the testpanel and DAQ code read back a sinewave for example - you would generally expected the code to show the same results, if the DAQ code is written well. 

Can I ask why you wanted to read the floating value?

If you want to look for some examples of DAQ code then have you looked at NI Example Finder? Help>>Find Examples.. Hardware Input/Output>>DAQmx>>Analog Input..

Hope this helps, 

Message Edited by Hillman on 06-05-2009 11:31 AM

[johanneshoer]

Hello,

thank you for the quick response.

> I would not recommend leaving your inputs floating (disconnected). This will cause the voltage reading to fluctuate and be random. Your tests aren't accurate, as in the time it takes you to change from Testpanel to code (or vice versa) the floating value could very useful change. 

I know that open inputs can float, this is obviously the case here.

But I am wondering, why the Testpanel and the DAQ Assistant do not show the same voltage.

In the Testpanel it is always around 0V, and in DAQ it is always around 1.4V, not varying a lot, just a little rippling, but not floating from 1.4 to 0. Its definetly not the case, that the voltage floats from 1.4 to 0 V, while I change from DAQ to the Testpanel, and it is not floating back to 1.4 V, when switching back.

> Can I ask why you wanted to read the floating value?

I have connected path sensors to the NI USB-6009 to measure travels in a testbench. Sometimes the sensors have to be removed, so the disconnected sensors have to be replaced with a pull down resistor in this case.  

That resistor has to be quite low. If I use a Mega-Ohm resistor it does not pull down the voltage, it has to be lower k-Ohm. I don't like that, because the travel sensors are also in the range of k-Ohms.

Johannes

Message Edited by johanneshoer on 06-05-2009 06:23 AM

[Coq_Rouge]

If this is the problem you should implement a service mode switch in your application. If this setting is applied all daq output may be forced to a safe value.

[Tom__Sedlack]

The test panel is measuring in differential mode (the difference between ai0 and ai4 if I remember correctly) and DAQ assistant is running in single-ended (RSE) mode which is measuring AI0 with ground as a reference.  The differential shows ~ 0 because both ai0 and ai4 are probably around 1.4V which you read in single-ended mode.  Put either panel into the measurement mode of the other, and they will read the same.
 
I checked using my USB-6008 and differential is ~0V and RSE is 1.4V.  Why they measure ~1.4V, I can't answer.
 
Tom

Message Edited by Tom Sedlack on 06-05-2009 07:08 AM

[johanneshoer]

I have just checked, what resistor do I need to pull down the input.

I have connected the following resistors, one at each input:

AI 0: 1k

AI 1: 3.3k

AI 2: 10k

AI 3: 33k

AI 4: 100k

AI 5: 330k

AI 6: 1M

AI 7: open

Now see the result (attached picture): The higher the pull down resistor, the higher the measured value (1MOhm --> 1.2 V). Only the 1k pulls it down to almost 0 V. So the pull down resistor schuld be maximum 1k !???!

[johanneshoer]

How can I switch from RSE to Differential measurement mode in the DAQ Assistant?

If I iopen the Terminal Configureation pull down menu, it just shows RSE and >Let NI-DAQ Choose>, the rest is greyed out.

Does Differential mean, measuring the difference between two inputs instead of measuring between input and ground? If yes, then I have only 4 inputs (+ & -) instead of 8, but I need 8.

Johannes

Message Edited by johanneshoer on 06-05-2009 07:41 AM

[falkpl]

see app note:

http://digital.ni.com/public.nsf/allkb/E687933C5694AB00862570BD00593CA3

[johanneshoer]

@falkpl: Thank you for the link

In the USB-6008/6009 User manual it says on page 19

"When no signals are connected to the analog input terminal, the internal

resistor divider may cause the terminal to float to approximately 1.4 V

when the analog input terminal is configured as RSE. This behavior is

normal and does not affect the measurement when a signal is connected."

This is correct, if the signal is strong enough.

But if its a weak signal (voltage with high resistor), the floating voltage does affect the measurement! Measure the voltage in a potential divider and you'll get incorrect results.

[falkpl]

I had this issue with a USB 6008(9?) a few years ago (2006) and remember it was a stumper.  I originally thought it was a hardware problem but did get it to work with the help of some of the app notes.