INA125 Incorrect SPICE Model

[beedees]

Hi everybody!

I'm simulating a circuit with INA125 that is an Instrumentation Amplifier with Precision Voltage Reference. I'm using it with 5V single supply, so V+=5V and V-=0V. I've noticed that if input differential voltages are negative the INA amplifies the signal and if input differential voltages are positive it doesn't amplify. It's a strange behavior! If should be the other way!

I'm using the PSPICE Model provided by Texas Instruments and according to me there is something in the model that is wrong.

I tried to simulate INA125 with a simple circuit too but the result is the same.

According to me a possible error in SPICE Model could be the output (21) of the A1_125 SUB-CIRCUIT and the output (5) of the A2_125 SUB-CIRCUIT.

What do you think about? Have you ever tried to simulate this component? What is your result?

Maybe I was mistaken posting this thread here, if so I'm sorry. But I didn't find a best section.

Thanks all!

 

[Harald Kapp]

I've noticed that if input differential voltages are negative the INA amplifies the signal and if input differential voltages are positive it doesn't amplify.

What exactly do you mean by this?
IS this Vin+ - Vin- > 0 and Vin+ - Vin- < 0?
What are the absolute values of Vin+ and Vin-? Are they within the common mode range of the INA?

Also note that the datasheet specifies that it is possible to operate the INA125 from a single supply, but it is designed to be used preferentially with a dual supply since its outputs are not truly rail-to-rail. Try to simulate with a dual supply and compare the result to the single supply simulation.

 

[beedees]

What exactly do you mean by this?
IS this Vin+ - Vin- > 0 and Vin+ - Vin- < 0?

Yes, I meant this.

What are the absolute values of Vin+ and Vin-? Are they within the common mode range of the INA?

The absolute values are 1.7V and then above the limit reported in the datasheet.

Try to simulate with a dual supply and compare the result to the single supply simulation.

I have already done this simulation but the result was the same. For this reason I think the model is wrong. It's a strange behavior, right?
Surfing the internet I found this. Look at the last post. I think it's the same problem, isn't it?

 

[Harald Kapp]

Looks like you're right. Have you tried TINA as suggested in the link?

 

[beedees]

Yes, I have tried and but it works at the same manner. Unfortunately there isn't the possibility to report this problem to Texas Instruments. I would like to resolve this issue and help others who might waste time like me!