We have to measure 80x SG (strain gauge) channels with 20x NI-9237 modules in 3x cDAQ-9188 chassis. The CEng guy wants to use a passive SG for temperature compensation in all channels (although not needed because we use self compensating SGs in their temperature range). So he will glue the active SGs on the tested model, and he will glue the passive SGs on pieces of same material AWAY FROM THE MODEL - NO THERMAL CONTACT WITH THE MODEL!
This configuration for me is QUARTER Bridge Type 1 since the passive SGs (dummies) are just resistors unaffected by the seismic activities on the model.
However, he insists on Type 2 guided by its definition ("Strain Gauge Configuration Types" Tutorial) which sounds like his comprehension of his bridge. I stay with Type 1 because the dummies are not in a close contact to the tested model - I see them just as bridge completion resistors.
Can someone with expertise in SG conditioning and measurements help in solving this dilemma?
Thanks in advance,
If you are going to use the dummy SG then I would recommend Quarter Type 2. Please see the attached images and please note that for Q2 the “dummy” gauge is not bonded to the specimen and therefore does not change resistance with deformation of the specimen only with temperature. Hope this helps!
Thank you again Carisa!
I have made my DAQ work but am lacking info on the bridging. I have already adopted the Q2 also because I was told Q1 is inapplicable for NI-9237? Also that with Q2 poisson ratio is ignored, and that with RemoteSensing Rlead value is ignored.
Can you please explicitely confirm the following since there is no info I could find:
1. If I wire RemoteSensing (the RS+ and RS-) NI-9237 will automatically compensate for the leads resistances..?!
2. If I use the "DAQmx Perform Bridge Offset Nulling Calibration.VI"with my external HALF completed I DO NOT need to measure the initial voltage...?! Is this balancing suffcient to take into account the inherent voltage offsets?
OK, let me paraphraze my questions.
NI-9237 documentation says RemoteSensing is meant for HALF and FULL bridges, and for QUARTERs we should use shunt calibration. Since I plan to use Q2 which is not a HALF bridge will RemoteSensing be performed by the NI-9237 or not? If NOT, will measuring lead resistance do the job instead? I find it very awkward that this crucial info is missing from the NI-9237 documentation!
2. Brdge balancing
NI-9237 documentation says this module needs to know the "initial voltage" measured on disconnected ETERNAL part of the bridge to compensate for offset. I have used the "DAQmx Perform Bridge Offset Nulling Calibration.VI" to balance all my bridges in COMPLETED status prior to any acquisition. Is this doing the job so I do not have to measure the "initial voltages"?
The tasks of interest are the 3x cDAQ. The PXI task needs some IEPE current sources to be switched on for 60sec prior to the acquisition so that the PCB accelerometers can be initialized.
Thanks in advance,
1. For quarter bridges it is recommended to use shunt calibration which can correct for errors from the resistance of both the excitation wiring and wiring in the individual resistors of the bridge. Remote sensing only corrects for resistances from the EX pins on the NI 9237 to the sensor. Since you are essentially creating a half bridge by connecting the external resistor, you can perform remote sensing if you'd like.
2. Can you please point me to where you found documentation on "initial voltage"? I was unable to find that phrasing in the 9237 Operating Instructions and Specifications.
1. Initial bridge voltage
"Initial bridge voltage" is an input control within the "SG bridge information" cluster. AFAIK, this is the voltage measured (I guess betwen AI+ and AI-) in an OPEN configuration that is afterwards used to balance the bridge. I'd use the "DAQmx Perform Bridge Offset Nulling Calibration.VI" instead which works on COMPLETED bridge prior to acquisition. Before tests I always scan for offsets on all of the sensors and substract them afterwards when calibrating the signals - isn't this the same? - Should I acquire zero after balancing a bridge channel?
2. Lead resistance
This is also a control within the "SG bridge information" cluster. Is this ignored when RemoteSensing is wired? - It seems redundant to the RemoteSensing compensation.
Look forward to hearing from you,
I am working on a project to identify the strain acting on the landing gear of Cessna 150, is there any chance I can use the Rosette rectangular type strain gauge and wire it with one module - 9237. If yes, then I would have to use the quarter bridge configuration. But I am not sure if 3 gauge arrangement can be made with 9237 module. Any advice or assistance will be appreciated.
Yes, you can use one NI 9237 module to measure a rectangular strain gauge rosette. The NI 9237 module has four input channels, so you can connect all three gauges to one module.
And as of NI-DAQmx 9.5 you can specifically select to create a AI Rosette Strain Gauge task. Prior to DAQmx 9.5 you could still acquire from a rosette, but you had to manually perform the necessary rosette calculations in software. Now the driver will make the SW calculations behind the scenes. If you are using LabVIEW 2012 there is a new shipping example that demonstrates acquiring from a rosette.
If you have an older version of DAQmx and are programming in LabVIEW, something like this Rosette Strain Gauge Example VI can be used to calculate the 2D plane strains from the three strain gauges.