data aquisition

[rizo59s]

Hi everybody,

I would be glad if someone could assist me in solving this problem we are facing here with the data aquisition system. We are using the NI-SCXI 1001 chasis, for measuring strains, custom voltage with excitation for various measurements related to civil engineering projects. For now we seem to have also figured out a program in LABVIEW to measure all the parameters important to us, but the problem is, we cant get the initail readings to be zero before we start our aquisition. In our program we have about 17 channels with 3 Daq-mx assistants. Everything is working fine and giving output as we like it but the fact remains that we cant seem to initialise the readings to zero for the first measurement. I am also attaching a copy of our LABVIEW program, i know its a little too unsophisticated but works for us. Any help on the solution would be greatly appreciated. Thank you all.

Hardware used:

NI-PXle-1062,

Chasis: SCXI 1001

Modules: SCXI 1520, 1540

Terminal Blocks: SCXI 1314, 1315

[devchander]

Could not open your vi, coz i have LabVIEW 8.0 in my system.

But here are a few suggestions form making your initial readings zero.

For strain channels: you could perform offset nulling as illustrated in this example vi 'Cont Acq Strain Samples (with Calibration).vi' that you will find in examples shipped with LabVIEW

For voltage channels, there is no such feature as 'offset nulling' and they will measure the potential that is present across the terminals of that particular channel.

However,if absolutely necessary in you application, you could perform an offset nulling by measuring the value of first sample and subtracting this value from all subsequent value

[rizo59s]

Hi,

Thank you devchander for ur input, we have tried the offset nulling option for strains and it seems to be working but for the other channels which measure custom voltage with excitation, as you had advised to subtract the first value from subsequesnt values, i had a question, is there any way of doing it in LABVIEW program itself. Or do we have to do it later in an Excel file.

Also is there any limit for the buffer before it gets filled up and we gotto transfer the data before going on? Coz we got to run an application that might last for an hour. Also we are using DAQMX assistant for creating the channels and obtaining the data and not the NAQDMX task creater present in the functions pallette.

Thank you.

Rizo.

[devchander]

---

is there any way of doing it in LABVIEW program itself. Or do we have to do it later in an Excel file.

---

One more comment I might add here.

I suggest you do this only if you want to ignore the offset that your voltage source has.

For example: If your voltage source has an initial offset which you want to ignore, you can do this sort of nulling

But if you observe some offset ( with reference to what an additional measuring instrument like DMM measures for that voltage source) in a particular DAQ channel's reading, you might be better off trying to find the reason for this induced offset and correct it, rather than nulling the offset.

Which would your case be?

Regarding your query on buffer, DAQ assistant vi's are built using DAQmx functions. This link should guide you on that

[devchander]

---

is there any way of doing it in LABVIEW program itself. Or do we have to do it later in an Excel file.

---

One more comment I might add here.

I suggest you do this only if you want to ignore the offset that your voltage source has.

For example: If your voltage source has an initial offset which you want to ignore, you can do this sort of nulling

But if you observe some offset ( with reference to what an additional measuring instrument like DMM measures for that voltage source) in a particular DAQ channel's reading, you might be better off trying to find the reason for this induced offset and correct it, rather than nulling the offset.

Which would your case be?

Regarding your query on buffer, DAQ assistant vi's are built using DAQmx functions. This link should guide you on that

[rizo59s]

Hi Devchander,

Once again thanks for ur help, from what i have found out, our case seems to be the first one ( we have an initial offset which we want to ignore). So how to go about doing that in the LABVIEW program itself.

Your reply on the buffer problem was also very helpful, thanks.

Rizo.

[devchander]

Just perform a finite acquisition before starting your continuous acquisition ( either a single point acquisition or an average of a multi point acquisition)

next store this offset value in a local variable or wire it into your continuous/main acquisition vi.

Now, subtract this offset value from all acquired values

Hope this attached vi helps

Note: at present, I do not have a DAQ card and could not test this vi.

So, do run it and let me know if it serves your purpose

Regards,

Dev

[rizo59s]

hi devchander,

Thank you very much for ur prompt replies. They have been very usefull and i think we are on our way. The VI that u sent with the last message was working just fine and we could incorporate the methodology in our VI.

Once again thank u very much for ur help. Will contact u if we need anything else.

Thanks.

[rizo59s]

Hi,

I had one more query though that i wanted to clear before we proceeded with the final test.

1) I am using three DAQ assistants in my labview program for different devices, is it ok to have three different "while loops" and consequesntly three "STOP" controls, for the DAQ assistants or do we need only one loop and eveyy DAQ assistant should be inside that one loop?

2) I dont know if you are the right person to ask this but does it matter if the cables that we use for connecting the LVDT and the potentiometers and the accelerometers and very long. Actually we are tyrying to do a blast and would want the apparatus to be atleast 50 meters from the blast site. Since we are using them to measure the "custom voltage with excitation" we dont have any bridge type (eg. quarter bridge or full full bridge) associated with them.

Thanks.

[devchander]

For your first query, If each of it is continuous acq, you better keep them in seperate loops itself.

In fact, since your 3 devices are not synchronized its best to keep the three DAQ assistants in seperate Loops.

Note: probably you could synchronise the 3 loops using semaphores, and use a local variable of a single STOP boolean( mechanical property: 'Switch' when pressed) to stop all 3 loops at same time

Regarding second query, yes the distance does matter!

But to what extent, I really cannot determine and tell