AD8609 Thermal drift issue

am using AD8609 OP-Amp for load cell interface, the problem i have is the ADC output is fluctuating due to temperature rise (if we set it for 1000 the value very from approx.900 to 1300 range, this values are direct averaged output of ADC ).

The reason for ADC value fluctuation is, Opamp output voltage is decreasing while the temperature is increasing, the same way the temperature is decreases the voltage is increasing.

if i kept the board is off state in few hours and turn on the AD8609 U15C output voltage is going high above the setting voltage.

i use to set the U15C output voltage is around 1.211(To set ADC output as 1000 counts)

AD8609 Supply voltage AVCC = 4V this is given by LDO

Please find the attachment for schematic.

Note:

am not doing any special temperature test, its happening under normal working condition of the board.

Putting some emphasis on correct technical terms, we should talk about temperature drift as in (a systematical effect) as in the question title in contrast to random fluctuations.

In the present circuit, resistor T.C. can be expected to be the most severe drift source, at least factor 10 larger than OP offset drift. It's promoted by a unsuitable instrumentation amplifier circuit that misses the chance to reduce the effect of common mode error. It should better refer to a standard true instrumentation amplifier circuit with the gain located in the first amplifier stage.

I presume that you separated amplifier from load cell drift in your measurement by shorting the load cell output terminals.

FvM said:

Putting some emphasis on correct technical terms, we should talk about temperature drift as in (a systematical effect) as in the question title in contrast to random fluctuations.

In the present circuit, resistor T.C. can be expected to be the most severe drift source, at least factor 10 larger than OP offset drift. It's promoted by a unsuitable instrumentation amplifier circuit that misses the chance to reduce the effect of common mode error. It should better refer to a standard true instrumentation amplifier circuit with the gain located in the first amplifier stage.

I presume that you separated amplifier from load cell drift in your measurement by shorting the load cell output terminals.

Click to expand...

FvM,

"Thanks for your comments"

The resistor used in this circuits are 0.1%, 25PPM, thin flim resistors,
if i short the load cell output terminal from the opamp side the opamp U15C output stays at 1.258V, our ADC wont measure at this voltage because the ADC max is 1.25V only,

please find the attachment for complete Analog section circuit

if i short the load cell output terminal from the opamp side the opamp U15C output stays at 1.258V, our ADC wont measure at this voltage because the ADC max is 1.25V only

Click to expand...

So you are saying, that zero load cell differential output is outside the measurement range? That's very unusual. I don't understand how a load cell amplifier could be calibrated under this conditions. I would always try to implement a bipolar input range, at least including 0.

I'm not sure about your ADC counts calculation. According to my calculation, a change of 400 counts corresponds to 6 mV ADC input change which is 6 uV input related. Related to 1 uV/°C typical input offset drift a moderate amount.

What's the T.C. specification of your 0.1% resistors? Even if it's oustanding, R37 variable resistors can easily dwart it.

- - - Updated - - -

I correct the input referred "400 count" voltage to 7.6 uV, still the same problem.

How do you set R37? It should be adjusted for maximum common mode rejection. If you use it or zero adjustment instead, common mode rejection will most likely bad and a small drift of bridge comon mode voltage cause a large offset.

We are setting Load cell input amplifier as virtual ground level of 1.25V, i adjust R37 trim pot to get U15C output less than 1.25V usually set it in to ADC output as 1000 counts

the U15C output reduces the ADC count will increase, ADC count 1000 is averaged value of 16 internal counts of ADC