Author Topic: Instrumentation Amplifier gain accuracy problem  (Read 1612 times)

0 Members and 1 Guest are viewing this topic.

Offline filssaviTopic starter

  • Frequent Contributor
  • **
  • Posts: 433
Instrumentation Amplifier gain accuracy problem
« on: May 23, 2016, 09:15:59 pm »
As a part of a larger project I'm developing a load cell front-end, the first circuit I've prototyped is the one attached where I have the ADC connected to I2C to an arduino uno taking the measurements, the signal from the load cell  is at 10mV and has to be amplified to meet the ADC's (ADS1115) full scale input range, so I used an istrumentation amplifier (AD8236) with suitably selected precision gain setting resistor.

The problem I'm having is a gross gain error, so basically the amplifier gain is only 7 instead of 25 as it should be; the theories i'm coming up with are as follows:

1) even though the in-amp is marked on the datasheet as Rail-to-Rail input and output has some limitation in input and output so it cant gain so much with reference voltage grounded
2) since the prototype board is through hole (Inamp and ADC are SMD on breakout-boards) wired point to point with 30 AWG wire wrapping wire and the wire (not more than 10 cm)  between sensor ground and inamp reference-ground (which are soldered toghether really close) is enough to cause gain accuracy issues as stated in the datasheet
3)something else I haven't thought off

so what do you think is the issue here?

UPDATE 1: I need a gain of 25 to get from the 10 mV FS of the load cell to the 256 mV FS off the ADC

If the first is the likely culprit my solution to that would be lift the reference pin to mid supply and use the ADC as differential bethween vref and vout of In-amp, can it work or i need a fully differential amp and have a real differential front-end
« Last Edit: May 23, 2016, 09:37:46 pm by filssavi »
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2850
  • Country: se
  • Country: Broken Britain
Re: Instrumentation Amplifier gain accuracy problem
« Reply #1 on: May 23, 2016, 09:32:07 pm »
The problem I'm having is a gross gain error, so basically the amplifier gain is only 7 instead of 21 as it should be;

G = 5 + 420 k?/RG

I make the gain with 21K 25. Are you sure the 21K is not 210K that would give a gain of ... 7.
.  That took much longer than I thought it would.
 

Offline filssaviTopic starter

  • Frequent Contributor
  • **
  • Posts: 433
Re: Instrumentation Amplifier gain accuracy problem
« Reply #2 on: May 23, 2016, 09:36:27 pm »
That is exactly the point (i could have been clearer now that I think about it, i'm sorry)
the gain i need is 25 (10mV cell output, 256 mV ADC input range) instead i get only 7, i already measured the gain setting resistor and it is the correct value (i'll double check tomorrow but the RS catalog number is correct the value stamped on the resistor is the correct one)
 

Offline tszaboo

  • Super Contributor
  • ***
  • Posts: 7938
  • Country: nl
  • Current job: ATEX product design
Re: Instrumentation Amplifier gain accuracy problem
« Reply #3 on: May 24, 2016, 12:24:22 pm »
How much is the VREF? What generates it?
 

Offline filssaviTopic starter

  • Frequent Contributor
  • **
  • Posts: 433
Re: Instrumentation Amplifier gain accuracy problem
« Reply #4 on: May 24, 2016, 12:44:19 pm »
Right now i have it directly connected to the In-amp ground and this is connected throug a 30awg wire to the load Cell ground

In the proposed solution i will set it to mid supply (2.5 V) by using a buffered (to get low impedance) eesistive divider from power supply
 

Offline tszaboo

  • Super Contributor
  • ***
  • Posts: 7938
  • Country: nl
  • Current job: ATEX product design
Re: Instrumentation Amplifier gain accuracy problem
« Reply #5 on: May 24, 2016, 02:16:59 pm »
Well both of them should work. I suggest to measure R1 again just to be sure.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf