Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
NI 9502 ADC conversion
5 views
Skip to first unread message
menzi74
Jul 20, 2008, 2:10:10 PM7/20/08
to
I have the following short questions regarding the NI 9205 module, which I use in connection witht the RIO: - Floating input, how to connect COM?
In the manual of the NI-9215, a bias resitor of about 1M (connected to AI-) was suggested to provide a defined common mode voltage if the source was floating. This resitor was even in-built in the version with the BNC housing. This makes sense, but it was not mentioned inside the NI-9205 manual; can I use the same method to connect floating inputs?- Sampling rate.
The sampling rate was mentioned to lay inbetween 200kS/s and s250kS/s. In the FPGA, module setup, I can adjust a time between conversions between 4us and 8us. I am not quite sure about the precision issue. If you sample multiple channels at <8us, probably the precision will decrease, due to the settling time. However, If I sample only a single channel, I could utilize the full 4us/250KS/s without loss of precision, right?
In the manual of the NI-9215, a bias resitor of about 1M (connected to AI-) was suggested to provide a defined common mode voltage if the source was floating. This resitor was even in-built in the version with the BNC housing. This makes sense, but it was not mentioned inside the NI-9205 manual; can I use the same method to connect floating inputs?- Sampling rate.
The sampling rate was mentioned to lay inbetween 200kS/s and s250kS/s. In the FPGA, module setup, I can adjust a time between conversions between 4us and 8us. I am not quite sure about the precision issue. If you sample multiple channels at <8us, probably the precision will decrease, due to the settling time. However, If I sample only a single channel, I could utilize the full 4us/250KS/s without loss of precision, right?
Eirikur
Jul 24, 2008, 7:40:08 AM7/24/08
to
Hi menzi74 The Measuring Floating (Nonreferenced) Sources section of the <a href="http://zone.ni.com/devzone/cda/tut/p/id/3344#toc2" target="_blank">Field Wiring and Noise Considerations for Analog Signals Guide</a> gives an excellent explanation of when and how to use bias resistors. Basically yes, a bias resistor is recommended when measuring differentially to keep the common mode voltage level within the input range of the device. Bias resistors should be used when measuring floating signal sources in DIFF and NRSE configurations no matter the device. Failure to do so will result in erratic or saturated (positive full-scale or negative full-scale) readings.
Regarding your other question you are right.
Regarding your other question you are right.
Eirikur
Jul 25, 2008, 3:10:10 AM7/25/08
to
You can take a look at page 4-17 of the <a href="http://www.ni.com/pdf/manuals/371022j.pdf" target="_blank">M-Series User Manual</a> for a discussion of the resistor value
that should be used when connecting AI- to AI GND.
that should be used when connecting AI- to AI GND.
0 new messages