NI USB-6008

[johnsold]

I made a mental math error. With 13 multiplexers you will not have enough DO lines to control them directly.  You could use a 4-line to 16-line decoder to manage the Inhibit lines.

But, the comment Dennis posted gave me another idea. I was not taking advantage of all eight analog inputs. If you use two levels of multiplexing you could put 16 thermocouple channels on each of the 8 analog inputs for a total of up to 128 channels and it would only require 4 DO lines for addressing.  For each AI channel use two CD4051s. The outputs of those go to one input pair of a CD4053. The output of that section goes to the AI terminal. Three DO lines go to the A, B, C address inputs of all the CD4051s. The fourth DO line goes to all the A, B, and C selector input of all the CD4053s.

With the multiple connections through different materials you will definitely need an AD595 per thermocouple. At 30 degrees C you only have 1.2 mV from the thermocouple at about 39 uV/degree. If the temperature of the multiplexers and wiring varied a few degrees, the errors could be too large to make the system usable. Plus, it would take twice as many multiplexers (with twice as much error) to switch both sides of each thermocouple. The error due to the on resistance of the multplexers will be about 0.3%, considering the low input impedance of the USB-6008.  This is much smaller than the errors due to the accuaracy of the USB-6008, the thermocouples, and the AD595.  The combination probably has total errors on the order of 5-6 degrees.

Lynn

[ubongakpabio]

thank you so much for your explanation, this is a lot more explicit. but i was thinking that since am measuring temperature and in this case taking contionous reading, i dont need the digital outputs and inputs.

most importantly, please list the total number of multiplexers CD 4051 and AD595 thermocouple amplifier that i would need to achieve this 100 channels and any other material if necessary to be purchased.

Thank you again for your time.

Part Two

you said i could use NI 9123 together with a chasis instead, but recently a friend told me about NI-PXI 2575 and NI- SCXI-1175 could this other ones be used to achieve the same 100 channels with proper signal conditioning, is it a better option to the NI 9123.

[Dennis_Knutson]

I'm beginning to think that you don't have any electronics background. Of course you would need the digital I/O. That is how you select a channel with the CD 4051. You would need one AD595 post thermocouple as has been stated a couple of times.

You can use the sxci our pxi if you also purchase an appropriate daq device to do the measurement and of course you need a chassis and controller (at least for pxi) and the total is going to be several thousand dollars.

Do you have any experience with schematic capture and pcb design and layout? The 6008/CD4051/AD595 option is going to require a pcb and not some flaky breadboard.

You should also think about renting some commercial test equipment such as the Agilent/Keysight 34980. I recently used one to acquire data from 100+ thermocouples. It even has software that you can use instead of writing your own.

[ubongakpabio]

yes, I dont have a solid background in Electronics, my specialty is power engineering, just delve into a bit of the electronics for my masters. thank you very much. i would figure out the rest things for myself. I appreciate your patients in handling my issue.

[ubongakpabio]

your advice really helped am getting to study on the levels of multiplexing and its quite challenging since my field is not electronics. you said for each of the 8 analog input i require 2 CD4051 and that the ouput goes to a pair of CD 4053. this therefore implies that i need to purchase 16, CD4051 and 8 CD 4053?.. i have a good knowledge of the PCB

[SteveD123]

30 years ago I was involved with a project multplexing K type TCs into an ADC. I strongly advise against multiplexing the raw TC inputs because of the interference and settling time between separate TCs. I advise an AD595 for each channel, followed by whatever multiplexer arrangement you choose.