Tricking NI 9219 with External Excitation

[Geank]

Hi all,

I need to take force measurements using full bridge load cells, but the ones I have need 10 Vdc and 5-15 Vdc excitation.

I know that the NI 9219 doesn't support external excitation, but I wonder if I could "trick" the board by selecting internal excitation and then connecting the ex. wires to another differential voltage supply. Assuming that I don't have grounding or voltage stabilization problems and I scale the measurements correctly, would it be any issues?

Regards,

Geancarlos Zamora.

[JÞB]

---

@Geank wrote:

Hi all,

I need to take force measurements using full bridge load cells, but the ones I have need 10 Vdc and 5-15 Vdc excitation.

I know that the NI 9219 doesn't support external excitation, but I wonder if I could "trick" the board by selecting internal excitation and then connecting the ex. wires to another differential voltage supply. Assuming that I don't have grounding or voltage stabilization problems and I scale the measurements correctly, would it be any issues?

Regards,

Geancarlos Zamora.

---

Just get a device that does support external Vex and avoid the headache and possible risk to your hardware.  the 9237 seems to be the better choice

[Geank]

That'd the ideal solution and we actually asked for that module in the university like a year ago, but since it seems like it's not going to arrive anytime soon I was wondering if I could the "trick". What would be the possible risks to the hardware, provided that the external supply is stable and without grounding issues?

[JÞB]

---

@Geank wrote:

That'd the ideal solution and we actually asked for that module in the university like a year ago, but since it seems like it's not going to arrive anytime soon I was wondering if I could the "trick". What would be the possible risks to the hardware, provided that the external supply is stable and without grounding issues?

---

Provided those caveats.  The risk is limited.  The next developer might not respect those limitations.  I would not recommend that trick.

Fair enough?

[Henrik_Volkers]

Seems that you need to make some measurements and time is also a factor 😉

I had a look at the user guide /spec of the 9219.  From what I found it seems that the 9219 doesn't have a constant voltage exitation. It depends on the load, so I assume, that the bridge supply is measured and the output (mV/V) is calculated acordingly.  So if you want to use the the full bridge mode to get the better specs, you need to supply your bridge ratiometric to the provided bridge supply voltage.

I would use a fully diffential OP (THS4505 or LT1994)  with a gain of 4 and tie the V_com to the center of a resistive 'half bridge' of 700 Ohm (2x350, results in 2.5V supply according to spec) , use a floating 15V supply again with a voltage divider to tie that center to V_com.

The documentation don't tell much about the internal wirering and common voltage ranges , maybe you can use a simpler setup with a floating 10V supply, 9219 set to full bridge mode, .... see picture:

Set the 10k pot so that you measure 2.5V between Ex- and Ex+ with a DMM

Disclaimer: Not proofed .. do it at own risk.... according to the spec nothing should burn 😉     maybe a NI eng. with more information on the input circuit can give a comment!

If you have the chance: Make a calibration of that setup!

Maybe you can find a bridge amplifier suitable to drive 10V with a amplified output for arduino (etc) boards for low cost (but questionable stability ??) and use voltage mode?

EDIT:  What loadcell are you using??  Real resistor based ones usually are linear and can be supplied with less than 10V. However piezoresistive ones need the specified supply voltage. Please provide the type and model (and a link to the datasheet?)