TLV2472: TLV2472 op amp gain different from ideal gain

Dani Rustandi

Prodigy 70 points

Part Number: TLV2472

Hi,

I want to create amplifier, to amplify voltage from DC R Shunt. the voltage from r shunt is 0-60mV.

i want to amplifiy to 0-4.8V (80-81x gain).

i use TLV2472IDGNR, single supply +5V, and use R1 120K ohm R2 1.5Kohm (0805 1% tolerance)

configuration of op amp is non inverting.

The result of my trial, the gain was unstable, the gain is different between low voltage and higer voltage.

when i inject 5mV the gain was 60x, and the gain increase when the input signal increase. I try to inject 60mV the gain is about 77x.

could you please to help me, why it was happened, i try on ti web bench the result is like the calculation and the gain is stable.

over 5 years ago

0 Paul Goedeke over 5 years ago

TI__Mastermind 30600 points

Hi Dani,

Can you share some scope-shots of what you are seeing? How are you calculating gain? I would also be interested in knowing if your input current/voltage is changing rapidly or if it is a slow signal.

I put together a schematic in TINA, can you verify that it is correct and matches your system? I've simplified the input stage and used a 10mOhm resistor and an ideal current source - this measures input currents from 0A to 6A.

Thanks,

Paul

TLV2472_CurrentSense.TSC

0 Dani Rustandi over 5 years ago in reply to Paul Goedeke

Prodigy 70 points

Hi Paul,

Thanks for fast reply.

I calculate the gain refer to non inverting opamp equation (gain=1+Rf/Rg). I attach the schematic and measurement result.

for the test i used the mV generator, the module not connected yet to Rshunt, i want to test it first by DC mV generator.

i download the tina software first then i will open file from you.

Thanks,

Dani

test result.pdf RShunt sig cond sch.pdf

0 Ron Michallick over 5 years ago in reply to Dani Rustandi

TI__Guru*** 140146 points

Dani,

Gain can not be measured using just one data point because the transfer function is VOUT = VIN * Gain + Offset
Gain is not equal VOUT/VIN unless offset is zero. Gain is (Vout1 - Vout 2) / (Vin1 - Vin2)
Offset comes from Vio * 81 plus CMRR of your resistor matching * sense resistor common mode voltage.

0 Dani Rustandi over 5 years ago in reply to Ron Michallick

Prodigy 70 points

Hi Ron,

Thank you for your explanation, i just calculate my test result by your equation and the gain is quite stable.
my another question, i used same resistor value, same brand for 2 TLV2472, but the result is different.
so i have to do trial and error to change the resistor value to make the output near equal on all opamp, It make the production process longer.
do you have advice? Than you

Regads,

Dani

0 Ron Michallick over 5 years ago in reply to Dani Rustandi

TI__Guru*** 140146 points

Dani,

The regulator in the schematic is 3V, not 5V.
The difference amplifier is not DC balanced. The inverting pin ground (negative input) resistor is 1k and the non inverting positive input resistor is 2.1k; this is not a big issue if the sense resistor is terminated to same voltage potential ground as R2, R10, R19 connect.

TLV2472 Vio at room temperature can vary +/-2.2mV per data sheet.
The gain is about 83 so the output can vary 83 * +/-2.2mV = +/-183mV from ideal output.
Do you see device to device variance within this range?

0 Dani Rustandi over 5 years ago in reply to Ron Michallick

Prodigy 70 points

Hi Ron,

I was wrong to write designator in the schematic. Actually i'm using TPS70950DBVR as regulator and the output voltage is 4.999V.
base on your explanation, my output was vary and still in range of vio tolerance.
i create 8 amplifiers using this opamp and there was variance between output, even the input is the same voltage, but i think still in the range +/-183mV. this is my first project using op amp, so i hoped the result is near ideal. i think i have to learn more about op amp. i'm looking for precision design for current sense but the cost is not too expensive, it will be use on my rectifier to sense DC current and voltage. Thank you

Regards,

Dani