INA250EVM: INA250

Hey Latha,

Here are the requested waveforms. I also attached the DC sweeps for load = 0A to 5A for both VREF conditions. As you can see, making your VREF =3.3V will seriously diminish the full-scale range of unidirectional current you can measure. If you want to measure unidirectional current, then you should consider keeping VREF a lower voltage. However, the one key benefit to providing a reference voltage is that you always keep the output in the linear region of operation even when load = 0A. This is important if you are measuring switching currents that can start from and fall to 0A and you need the device to react as quickly as possible to avoid overload recovery delay (which happens with any amplifier when its Vout is saturated into a supply rail and it needs to get out of this saturation). Any VREF >=100mV will work to bias the output into linear region and keep device from saturation at load=0A and thus prevent overload recovery delays.

Remember the following equation is always true: Vout = Load*Gain +Vref.

Hope this helps.

Best,

Peter

Hey Latha,

Concerning the simulations, nothing went wrong and amplifier model is behaving exactly has it should. They are displaying the point I am making about setting Vref voltage too high. The output voltage of the INA250 (and all other amplifiers) are bounded by their supply voltages. Vout can never exceed supply Vs. When making Vref = 3.3.V, this means the amplifier's Vout can only move from 3.3V to 5V for positive currents, thus output range is diminished. Thus, Vout saturates into Vs rail (minus ~100mV of datasheet swing) at around +3.2A.

As for the SPI digital isolator, I believe one of our digital isolators can help here. You can view all digital isolators here:

https://www.ti.com/isolation/digital-isolators/products.html

You can learn plenty about these products here:

https://www.ti.com/isolation/digital-isolators/technical-documents.html

You can also get more information and ask questions to the experts at the following links:

https://e2e.ti.com/search?q=SPI%20isolation&category=forum&group=355

https://e2e.ti.com/support/isolation-group/isolation/f/isolation-forum/991655/faq-how-to-connect-spi-mcu-to-multiple-isolated-spi-nodes?tisearch=e2e-sitesearch&keymatch=SPI%20isolation#

Sincerely,

Peter

Hey Latha,

I am a little confused with the question, but I think I know how to clear that up.

I assume that your "Actual" measurements are made with respect to the 2.5V reference voltage, correct? A schematic that correlates to the table of data could be helpful.

Anyway, the data you present makes sense and demonstrates basic functionality, although there is some gain error in your circuit/instruments, possibly from the DC source, ammeter measuring current, or the volt-meter measuring the output.

With all that being said, to answer your question directly: whenever you drive a reference voltage into the REF pin, the output voltage will become biased up to the DC voltage. So Vout = Gain*Load + Vref (this is with respect to ground). Usually it is helpful to measure the differential output voltage with respect to the REF voltage. The primary advantage in doing this is that the absolute error in the REF voltage is completely negated. So ideally the REF voltage should be 2.5V, but in reality it might be 2.492V. So in order to make sure there is no 8mV offset error to the measurements, the ADC should measure the differential output ( Vout with respect to VREF) so it never sees this 8mV difference.

Hope this makes sense.

Best,

Peter