08-25-2008 02:06 AM
Hi all,
I would like to know the inputs (keyed in by the user on the front panel such sensitivity/ gain etc..) prior to acquiring data from a bunch of accelerometers and the calculations involved (subtracting offsets) in order to obtain accurate voltage measurements .
regards
daq newbie
08-25-2008 05:42 PM
I am not an expert. Have you looked at the NI-Scope examples? I used the NiScope EX Differential.vi to get started on my last test using accelerometers. I believe your sensitivity/gains will be adjusted with the amplifier you are using with the accelerometers then your NI digitizer card will record the output of the amplifier.
The last test I used a PCI-5105 to collect the accel data from the amplifier and used a Bruel & Kjaer 4294 to check the calibration at 1g with the accelerometers. So I am not sure about the VI you are using that needs sensitivity/gain inputs. I have also found that it is easier for people to help when you can post the code you are having problems with.
Maybe someone else can help more.
08-26-2008 04:25 PM
Hi,
I would like to clarify exactly what your goal is. If I understand you correctly, you would like to record front panel control values from your LabVIEW front panel. One option would be to write that data to a measurement file above each data set as a Header. Do you have the Report Generation Toolkit?
Can you explain how knowing those inputs will help you obtain accurate voltage measurements? Do you mean that you'd like to programmatically process your data using those values, post-acquisition?
Thanks,
Dan Richards
08-26-2008 09:58 PM
Hi Mr Richards,
Thx for your reply. Basically I would like to perform some numerical processing when acquiring the raw data from the accelerometers.
I have got 10 accelerometers which go through a signal conditioning unit. I adjust the voltage offset and the gain for each one of them using the signal conditioning unit knobs so as to get max sampling range.
I then was thinking of maybe sampling all the connected data channels to calculate their bias. I would also get the user to input on the front panels: the sensors sensitivity (from datasheet) and the gain (from the sig conditioning unit)
finally, I would do then calculate the corrected value of each sensor ( Updated_data = (Raw_ data - bias) *10^3 (mV) / (Sensititvity*Gain)
Im using a pcie 6259 and an SCC 68 break out box by the way
What would be the best way of obtaining the bias and to implementing the above.
regards
08-28-2008 08:40 AM
Hello,
As far as the best way to obtain bias through your DAQ board, I will have to do a little bit more research.
Processing your data, post-acquisition, in such a way that it will not delay your DAQ or slow down your program could be tricky. You will need to empirically determine that. The solution to processing waveform or dynamic data, however, is essentially straightforward.
1. If you are reading to Dynamic Data, use the From Dynamic Data function to pull out one array or multiple arrays of data. If using Waveform Data, use the Get Waveform Components function to get the data array(s).
2. Wire the array into an Array Size funtion to get the size of the array (might output an array whose elements are the sizes of each array dimension - e.g. 100, 2).
3. Index the Array Size output if necessary and wire the length of the array into the "N" terminal of a For-Loop. If you know what size the array will be (e.g. if you are taking only 100 samples per loop iteration), then you could also just wire a constant to the For-Loop.
4. Create a wire branch from the data array and wire that into the For-Loop. Right-click the "tunnel" where data is passed into the loop and be sure that Indexing is enabled. Within each iteration of the For-Loop, you can do your processing on the value being passed through. Pass each processed value to the border of the For-Loop, again being sure that indexing is enabled and the output array will be an array of processed numbers. If using more than one channel per array, you may have to use one or more Index Array functions to pull out individual elements, in which case you would need to disable indexing where the data enters the For-Loop
Please let me know if this information helps you and if you have anymore immediate questions. I will look into determining gains from 6259 inputs.
Regards,
Dan Richards
08-29-2008 05:05 PM
Hello,
To my knowledge, the right way to handle bias would be to calibrate your device in Measurement & Automation Explorer (MAX) and let MAX handle the bias that way. Do you need your program to handle the bias in real time?
Please do let me know if this information helps.
Thanks,
Dan Richards
09-01-2008 02:59 AM
Hi again Dan
sry for the late reply, my internet was down was like a week. Oh I see so I dont need to sample each channel for its bias. Thats great then, that would save me a lot of time. would there be any help or documentation about how to calibrate my accelerometers anywhere on the ni website.
regards
don
09-02-2008 02:53 PM
Hi DAQ_Newbie,
The information on how to calibrate your accelerometers is likely to be available on the accelerometer manufacturing company website. There is a calibration button in your virtual channel configuration tab at the right side. You can click on it and follow through the instructions. This creates a NI-DAQmx scale for the channels, not the accelerometers. To calibrate the accelerometers themselves look at their product manuals. Hope this helps.
Ipshita C.