Hello all,
I am experiencing a problem with a NI 6225 card measuring the signals of several pressure transducers. To sum up, the measurement of a channel is slightly modified by the value measured on another channel. This NI 6225 card is multiplexed and I think that this problem is related to its "settling time".
I am using 2 pressure transducers (Keller PAA23). These pressure transducers are made of strain gages and include signal conditioning which provide a 0-10V output. I do not have the details about this signal conditioning as this is embedded in the sensor.
These 2 transducers are connected in differential mode to 2 channels of the NI 6225 card. The voltage output of the first one is 10V. The output of the second one is around 1V. Consequently, the voltage difference between the 2 channels is large: around 9V.
When the first transducer is not connected to the NI 6225 card, I read a value of 1.000V on the second channel connected to the second transducer.
When I connect the first transducter, and when the "interchannel delay" of the NI 6225 card is set to the minimum value of 4 microseconds, I read 1.006 V for the second channel. It means that I have an error of 6mV.
When I increase the interchannel delay up to 100 microseconds, I read 1.001V. The error is still 1mV.
I must increase the interchannel delay up to 200 microseconds, so that I read again 1.000V. No error.
I read the datasheet of the pressure transducer and its output impedance is supposed to be lower than 10 Ohm.
I repeated this test with a different make of pressure transducer supposed (according to its datasheet) to have also a low output impedance and I obtained more or less the same results: errors of several mV for low interchannel delays.
What I do not understand is that for a low impedance (10 Ohm), according to the NI 6225 datasheet, the settling time should be very low: lower than 10-15 microseconds. In my case, it is much larger: 100 or 200 microseconds.
1) Has someone already experienced this problem with pressure transducers used with multiplexed DAQ?
2) Can someone explain me the discrepancy between the actuel settling time (100-200 microseconds) and the theoretical one given by the datasheet (10-15 microseconds)? Is it related to the type of signal conditioning included in the pressure transducers? Why the manufacturers are giving a low impedance value if it is not really the case? Is it related to the NI 6225 datasheet that is too optimistic (by a factor 10!).
3) What can I do to solve the problem, except increasing the interchannel delay or using a voltage follower set-up?
Thanks for any help.
