cRIO 9205 input errors

[Oley_Mich]

Hello!

I've got an NI cRIO 9012 controller installed on an 8 module backplane.

2 x 9205, 1 x 9263, 2 x 9422, 2 x 9585 modules.

I'm having trouble reading an LVDT DC input.  Set up as +/- 10 VDC and differential.

My meter says it's at +10 VDC (should be near +30623 counts) , but reads around 18,000.

Even more disturbing is the fact that as the sensor moves through its stroke, the counts rise appropriately, but near the end of travel, counts seem to dissappear and go down to the 18,000 level.

Is there any possible explanation for this?

I can provide software if necessary...

Thank you!

[muks]

Will this help? 

[Ipshita_C]

Hi Oley,

How much time are you running the acquisition for? Also, can you tell me how much the voltage reading is off by (you have mentioned the counts but not the voltage offset). Also, have you tried calibrating the module?

Ipshita C.

[Oley_Mich]

Hello!

The acquisition is pretty much continuous.

With an input of 10 VDC, the counts returned start around 30623.

This lowers itself to 18950, equalling around 3.821 VDC difference, becoming steady state.

Every time I apply power to the LVDT, the readings start near 30623 and drop to the 18900 level in a few seconds.

Remove power, signal counts go to near 0, apply power, same results.

All the while, a DMM measuring signal out put always goes to +10 VDC with power applied.

Seems crazy to me!

[Ipshita_C]

Hi Oley,

Just to verify:

The drop in counts is arising only from the readings through the 9205 module.

The DMM is placed at the LVDT output signal--that is where you are measuring the 10V DC.

If both of the above are true, it would be interesting to try the following:

We need to isolate whether the power supply affects the reading at code level or module level. you can make a simple FPGA VI to just read the voltage from the module and output it to an indicator. Place this inside a loop (you can make it a single-cycle timed loop for cleaner code). If these voltage readings are skewered, your 9205 module may be at fault. If not, then you have to find how changing the power supply affects the FPGA.

Ipshita C.

[Oley_Mich]

Thank you for your time!

That is correct.

However, I have solved the problem (just recently in fact).

I have connected the COM pin (17) on both 9205s to my power supply ground.

This has corrected any issues I was having.

So now I have my + sig and -sig connections from the LVDT, PLUS, my power supply ground to COM.  I don't think I saw that in any documentation anywhere...

Seems rather strange.

Again, thank you for your time! 

[Ipshita_C]

Hi Oley,

Thanks for letting me know. I will make a note of this.

Ipshita C.

[Bassett_Hound]

Hi Oley,

The latest version of the NI 9205 manual should state the following note:

"If you leave COM unconnected, the signals float outside the working input range of the NI 9205. This may result in unreliable measurements because there is no way to ensure that the input signal is within 10 V of COM."

It could be that you have an older version of the manual so you may want to grab the latest version.

Hope that help direct you to the correct documentation.

Thanks,

Basset Hound