INA253: The contradiction of INA253A1 between the theoretical and practical value

Willie Neoh

Part Number: INA253
Other Parts Discussed in Thread: TINA-TI

Tool/software:

Hi,

Theory Part

I have tried to run a test by using the software called LTspice. Below is the construction of my circuit. As observed below, the model "U1" that I use is INA253A1. This "U1" is downloaded from Texas Instruments page under the part of design & development.

After I click "Run" in my simulation, the diagram below is the result.

The desire outcome that I want to get for my "Vout" is 1 V when 10 A is supplied on the "I1", while "Vout" is -1 V when -10 A is supplied on the "I1". To achieve that, I have added in network resistors as shown on the diagram below.

After I click "Run" in my simulation, the diagram below is the result. This is the outcome that I want.

Until this part, above explanation concluded for the Theory Part.

Practical Part

I tried to build exactly the same setup based on my simulation on my practical PCB setup. However, the results that I have obtained are different with the "Theory Part".

I can get Vout = 1 V when I supply 10 A, while I can get Vout = -1 V when I supply -10 A without adding in the network resistors. This scenario shows a contradiction between the Theoretical value as well as the Practical Value. Can I know why I will get these reading with explanation. I look forward for the reply.

Thank you.

Regards,

Willie

4 months ago

0 Peter Iliya 4 months ago

TI__Mastermind 32395 points

Hey Willie,

I am looking this over and will respond shortly.

Best,

Peter

0 Peter Iliya 4 months ago in reply to Peter Iliya

TI__Mastermind 32395 points

Hey Willie,

Would you please provide a schematic of your PCB circuit and instruments?

Also I think there could be an issue with how the model was imported into LTSpice, or how it is being interpreted. The INA253A1 is specified with gain of 100mV/A, so your real-world circuit is working as expected. I tested the model in TINA-TI Spice and it shows the correct gain given your circuit conditions. The only thing I could think of is that your input current source (I1) is floating. Maybe you need to bias it to ground (or some other valid Vcm voltage level).

Sincerely,

Peter

0 Willie Neoh 4 months ago in reply to Peter Iliya

Prodigy 30 points

Hi Peter,

Can you show me how do I add the bias to the ground by using LT Spice? This is because I am using LT Spice to run my simulation.

Thank you.

Regards,

Willie

0 Peter Iliya 4 months ago in reply to Willie Neoh

TI__Mastermind 32395 points

Hey Willie,

I’m sorry about the weekend delay.

I’m not exactly sure, but all you need to do for a connect and input pin (IN+/IN-) to an allowable common mode voltage which includes GND.

Best,

Peter

0 Willie Neoh 4 months ago in reply to Peter Iliya

Prodigy 30 points

Hi Peter,

After doing so, I am still unable to get the expected outcome that I want on the simulation.

Regards,

Willie

0 Peter Iliya 4 months ago in reply to Willie Neoh

TI__Mastermind 32395 points

Hey Willie,

It shouldn’t matter but try connecting GND pin to the ISN pin only.

You are shorting the INP and INN pins which is going to just make the device not see any signal. So remove the short of INP and INN and the reposition V2 to ISN only.

Sincerely,

Peter

Amplifiers

Amplifiers forum

INA253: The contradiction of INA253A1 between the theoretical and practical value