Which equipment for that ?

[hlod]

Is there any NI equipment which can do next thing on 60-100 channels:

- need to measure resistance, or just voltage (if i have constant current source)

- constant current source (need about 100 mA per channel)

What's the best solution for that. I need to log also all data.

[crossrulz]

You are going to need multiple cards.  As far as reading voltage, I have had good luck with the PXI-6225 which does 80 single ended AI channels.  I'm not sure at the top of my head what to use for a current source.  Probably looking at a power supply module.  I would contact my local NI rep to see what they come up with.  They have rarely ever failed me (and those few times we were asking for the world).

[johnsold]

Do you truly need a constant current source or do you just need to know the actual current?  What is the range of resistances you will be measuring? Are the resistances floating or connected to a common point?  What accuracy level do you require?  You mention logging.  How fast and for what time interval do you need to log the data?

Lynn

[hlod]

Yes for each channel I need constant current source (i need to measure resistance on connector, so there is no other power source). Measuring resistance is in range 5-15 ohms, with interval bellow 1000ns( I think the best for that is to use card with sampling rate 10 MS/s ???) Time logging is about 40 days, log to computer.

[johnsold]

That seems quite strange.  A connector with resistance as high as 15 ohms? Sampling faster than 1 MHz on up to 100 channels for 40 days? If I did the math right, that is more than 3E14 samples?

15 ohms at 100 mA dissipates 150 mW. 100 of them will dissipate 15 W.  Make sure your power source can provide that much.

Can any of the connections ever go open? My thought was to connect several connections in series and apply a voltage across the series combination of connections plus a known resistor.  Measure the voltage at each junction and subtract the voltage at adjacent connections to get the voltage across each one.  The voltage across the resistor gives the current. The current may not be constant but will change slowly enough that it could be considered constant for any one measurement.  Diodes could be connected across the connections to keep the series string conducting in the event of an open connection. This series string of resistances to be measured concept was published by NASA about ten years ago.

The inductnace and capacitance of the wiring to the connectors may also be an issue with MHz sampling rates.

Lynn

[Zody]

Dear Hlod,

Where are you from?

I can provide you the toll free number for NI TechSupport and we can discuss about what hardware will be the best for your application.

Kind regards,

Ion Rosca

Applications Engineer National Instruments