Non-linear scale issue

[ceilingwalker]

Hello. I am trying to scale the input from an amplifier that I thought was linear but it isn't. I have a 0-5V signal from an NI DAQ card (USB-6002) that is driving a 0-200V amplifier. This amplifier has what is labled "Monitor Voltage" BNC port on the front of the amplifier. Also, on the front of the amplifier is a label that reads "10mV/1V". These are the raw numbers, from LabVIEW:

1V=0.20V

2V=0.58

3V=0.96V

4= 1.35V and

5=1.73V.

I need to scale this so that:

1V=40V

2V=80.6V

3V=120V

4V=161V and

5V=201.4V

Thank you for your help.

[James.M]

Can you share the device information for us? It seems like a 2V ouput on your DAQ device should be generating the max 200V on the device.

Where are you getting these values from? Was the "1.73", "1.35", etc. read with a voltmeter?

[altenbach]

---

@ceilingwalker wrote:

Hello. I am trying to scale the input from an amplifier that I thought was linear but it isn't. I have a 0-5V signal from an NI DAQ card (USB-6002) that is driving a 0-200V amplifier. This amplifier has what is labled "Monitor Voltage" BNC port on the front of the amplifier. Also, on the front of the amplifier is a label that reads "10mV/1V". These are the raw numbers, from LabVIEW:

---

You can ignore the numbers on the left (since they match between the two sets) and simply do a polynomial fit of sufficient order (e.g. 2) between the right values of the two sets. Having the polynomial coefficient, you can use "polynomial evaulation" to calculate the V (right)  from set two for any V ( right) of set 1.

[James.M]

This is a linear correlation, or at least incredibly close to one. Here's a plot of those points and the linear fit is y=40.32x-0.36 with a mean square error of 0.1.

A polynomial fit comes up with 0.24+39.806x+0.085x^2 at order 2,

-0.6+40.986x-0.364x^2+0.05x^3 at order 3

[ceilingwalker]

---

@James.M wrote:

Can you share the device information for us? It is an NI DAQ (USB-6002)

It seems like a 2V ouput on your DAQ device should be generating the max 200V on the device. It is a 0-5V signal from the DAQ device that drives the amplifier. 5V out should drive the amp to its max 200V. So, at 1V out the amp is 40.21 output, and the "Voltage Monitor", what I am trying to scale, is at 0.19mV, according to data in from the DAQ device as well as a Simpson analog meter that I have on it.

Where are you getting these values from? Was the "1.73", "1.35", etc. read with a voltmeter? These are all values taken from the DAQ device, confirmed by the Simpson meter mentioned above.

---

[altenbach]

is (0,0) a forced point or could there be an offset?

[ceilingwalker]

---

@altenbach wrote:

is (0,0) a forced point or could there be an offset?

---

I don't know how to tell.

[altenbach · Accepted solution]

Well, given your data, here's how to scale the values after obtaining the polynomial coefficients. (you can delete the rest of the code after playing with polynomial order, etc.)

[Bob_Schor]

I think you might be a little confused.  You mentioned that your amplifier (with its monitor port) can deliver up to 200V.  What are the characteristics of the amplifier?  Is it a power amplifier, designed to deliver a lot of current to a load?  It might only have a voltage gain of 1 (or less), but still have the characteristics you describe.

The 6002 is well-suited for monitoring this amplifier, as its A/D inputs can accomodate ±10V, which means when your amplifier is maxed out at 200V, only 2V will be presented to the 6002, well within its tolerance.

Bob Schor

[James.M]

---

@Bob_Schor wrote:

I think you might be a little confused.  You mentioned that your amplifier (with its monitor port) can deliver up to 200V.  What are the characteristics of the amplifier?  Is it a power amplifier, designed to deliver a lot of current to a load?  It might only have a voltage gain of 1 (or less), but still have the characteristics you describe.

The 6002 is well-suited for monitoring this amplifier, as its A/D inputs can accomodate ±10V, which means when your amplifier is maxed out at 200V, only 2V will be presented to the 6002, well within its tolerance.

Bob Schor

---

He mentioned "input from an amplifier", so it seems like he is trying to scale a reading of 0-5V to 0-200V within LabVIEW.