If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TIDA-00366: Regarding ADC resolution in current sensing

Part Number: TIDA-00366
Other Parts Discussed in Thread: AMC1301, OPA320

Dear TI team,

Good morning.

We have been referring and using TIDA-00366 current sensing application circuit involving AMC1301 in our project(Motor current sensing).

But we have been facing below issue,

For example,

Scenario1:

we are measuring current 10A which is maximum current.

a. Reference voltage/level shift voltage is 1.65V .

b.  At maximum current,we have designed ADC voltage  to be 3.3V (peak to peak)

c. The offset voltage from ground level to negative peak of sine wave is 10mV(at op amp output)

d.  Here the offset 10mV doesn't affect ADC resolution and easy to design and understand.

Scenario 2:

We are measuring minimum current 1A

a. Reference voltage/level shift voltage is 1.65V .

b.  At minimum current,we have got ADC voltage as 0.5 V(Peak to peak)

c.  The offset voltage from ground level to negative peak of sine wave is 2.8V (at op amp output)

d. Here the offset 2.8V affect ADC resolution because from ground to 2.8v its empty.

How to offset the offset voltage at minimum current sensing?

Do we need to convert signals from time domain to other domain to counter loss of ADC resolution?

Please guide and provide reference design and papers.

Thank you,

Girish Kantharaju

• Hello Girish,

Thank you for refering to our reference design and reaching out to us on this issue.

According to your description on the 1A measurement scenario, the OP Amp output is within 2.8V~3.3V range (2.8V + 0.5V = 3.3V) , thus the bias is really (2.8+3.3)/2 = 3.05V, not 1.65V. Is this true? Have you measured the static state output of the OP Amp at 0A case?

Could you please share your current sensing section's SCH to us here?

Because, if the 10A scenario is correct, the 1A case's OP Amp output low level should be 1.65V - 1A * (1.65V - 10mV) / 10A = 1.486V.

Best regards,

Jerome Shan

• Girish Girish,

thank you for using our TIDA-00366 reference design.

The AMC1301 outputs a fully differential analog signal with an offset of 1.44V and linear full-scale differential voltage of +/-2V, for an input voltage of +/-250mV. Please refer to the AMC1301 data sheet, figure 24.

The TIDA-00366 use a 5mOhm shunt. That means a +/-50A current will yields a +/-250mV drop.

For single-ended input ADCs with a 3.3V unipolar input voltage range, the TIDA-00366 use a differential to single ended amplifier, please refer to the TIDA-00366 design guide section 4.4.3.3. The gain of the amplifier (OPA320) is 0.797 (I use 0.8 in below calculation). The output voltage offset will be 1.65V, which fits unipolar 3.3V input ADCs, where 1.65V equals 0A.

That means a +/-50A current (assuming a 5mOHm shunt), will be convert like: 0.05V=1.65V-(2V*0.8) for -50A and 3.25V for 50A. 0A yields 1.65V output voltage.

A 10A current with the TIDA-00366 settings (see schematic) will create a output voltage of 1.65V+0.4*0.8 = 1.97V.

It seems you measure different output voltage vs. your phase current?

Could you validate your design against the TIDA-00366 schematic for shunt, AMC1301 and OPA320 and ensure your design matches the TIDA-00366 schematic?

If you like to change the full-scale current range, please refer to above example calculation. If you like to change for example to +/-25A full-scale range, you may either increase the shunt to 10mOhm (ensure power ratings are met) or change the gain of the OPA320 from 0.8 to 1.6.