NI 9205 versus 9206 - What is the max voltage difference between channels?

[J_Cherng]

I am trying to compare the compactDAQ NI 9205 with the NI 9206 for a fuel cell testing application.  It seems to me that the main difference between these analog input modules is the amount of bank isolation they have.  The 9205 has 60VDC, where as the 9206 module has 600VDC.  However, I am more concerned about the maximum voltage difference between channels that the modules can tolerate.  For example, each channel will be used to measure an individual cell which is less than 2V.  The cells are connected in series with each other and therefore, the voltage difference between channel 1 and channel 8 could be as high as 16V.  Does either module have enough isolation to withstand such voltage differences?  If not, what is the maximum that the module can tolerate?  Any advice would be much appreciated.  Thanks!

[Jeff_T.]

Hi J_Cheng,

If the difference between ch 1 and ch 8 is only 16V then I believe that both the 9205 and 9206 will work.  An excellent wiring diagram can be found on page 13 of the 9206 operating instructions.  Notice that no channel is further than 10V from the isolated ground even though you might have many fuel cells stacked on top of earth ground.  To ensure this condition, you can connect the isolated ground (or COM) to one of the middle fuel cells.  For the 9205 though it is key that none of your input channels are more than 60V from earth ground.  Let us know if you need any further help as you develop this application.

Best Regards,

Jeff Tipps
Applications Engineer
National Instruments

Message Edited by Jeff T. on 01-09-2007 01:07 PM

[MCMC]

Hello,

This is an old topic but I had to reply because we have a problem with the channel-to-channel voltage restriction of NI 9206. What we try to do is measure the voltage of twelve series connected li-ion cells in a battery stack, each cell to a seperate channel in differential configuration. The open circuit voltage is 4.2 V for an individual cell when it is fully charged, and 3.7V nominal. There seems to be no problem if the number of cells is less than 6, however it is impossible to measure the voltage properly if we connect more than 6 batteries, therefore we can't use the 9206 with this battery stack configuration. This channel-to-channel voltage limit of ~20V or so doesn't clearly exist in technical specifications, it took a long while to figure out where the trouble was. We thought 9206 would be the best choice for our test system after carefully comparing the specs of all available products. Is it really possible to achieve what we need without saturating the device? Would you mind guiding us on how to solve the problem?

Thank you very much in advance,

MCMC

[Steven_K.]

Hi MCMC,

The limitation you are running into is the maximum working voltage. Here is the relevent spec from the NI 9206 Operating Instructions and Specifications:

 

6 batteries at 3.7V technically violates this spec since the total voltage is 22.2V. Even if you put COM at the middle of 11.1V, you would still be over the 10.4V limit. This is probably still close enough for the device to make a measurement, but it is still out of spec. It makes sense that adding another battery results in an incorrect measurement.

As for options for measuring 12 batteries, there are a couple ways you could go about this. One way is to get 3x 9206. Each 9206 would then handle 4 of the batteries, so it does not violate the working voltage spec listed above.

The most cost effective way would be to use the NI 9221. It has a +/-60V range. The downside is the resolution is only 12 bits and it is spread across a 120V range so the resolution is not as good as the 9206. The overall accuracy is an order of magnitude worse than the 9206 as well. You would need 2x 9221 to get the channel count for your application. Also, this is a single ended device.

[MCMC]

Hi Steven,

Thanks a lot for your answer. I think we will try to build identical resistive voltage dividers for each channel in order to decrease the overall

voltage, but i'm afraid the accuracy will suffer. If this doesn't work then 3x 9206 or 2x 9221 seems to be the only solution.

All the best,

MCMC

[Steven_K.]

Hi MCMC,

You want to be careful when using voltage divider to step down voltage for measurement. Most of our cards are multiplexed devices (including the 9206 and 9201), so there is a RC circuit in line before the ADC. By having too high of a resistive load, the time to discharge the capacitor increases, which can cause ghosting. Here is a good article on multiplexing and ghosting:

How Do I Eliminate Ghosting From My Measurements?

[engahmedtg]

Hello,

I know this is an old topic but I have a concern about the same issue. I have a small electrical network and i'm collecting arbitrary measurements from PTs and CTs. Is NI 9205 ok and will it work properly or should I go to 9206?

Your help is appreciated in advance. 

[amezam]

Hi engahmedtg,

What would be the requirements for your application?