AD620 circuit problem //Output

Thread Starter

sohamkul

Joined Jun 24, 2016
43
Hello everyone,

I am building a data acquisition system,i am using AD620 (instrumentation amplifier), have read data sheet thoroughly and i have used the following circuit which i obtained from internet (image attached ). I have also use two zero pots to offset zero on the inverting terminal of AD620.(images attached)

Problems:
1) even when the differential signal is not applied the output of AD620 jumps to 3.5 to 3.87volts
What is that i am doing wrong ??. i have tried various methods even then i am unable to amplify the differential signal
should i add resistors on both input terminals of AD620? please help .

p.s. I have grounded the reference pin
Thank you in advance.


(Internet image)
.Capture2.PNG IMG-20160703-WA0006.jpg Capture.PNG

Thank you in advance.
 
I can make a few comments only:
1) Your zero source is yucky and is high Z, fed from power rails
2) Type of power supply? (linear/switching?
3) Bypass caps (power rails, zero source)?
4) Does your signal have a PATH to ground.
 

Thread Starter

sohamkul

Joined Jun 24, 2016
43
1) i am using an SMPS
2) i have also used decoupling capacitors on each IC (0.1micro f)
3)sorry i do not have much knowledge about analog electronics can you please help me out to refine the circuit .

Thank you
 

dannyf

Joined Sep 13, 2015
2,197
1) even when the differential signal is not applied the output of AD620 jumps to 3.5 to 3.87volts
What is that i am doing wrong ??.
Could be nothing, all depending on what you meant by "not applied", and where the offsets are.

And the way you adjust the offsets is funny. Re-read the datasheet.
 

crutschow

Joined Mar 14, 2008
34,285
You cannot leave an input open (floating) to the amp. It needs a path for its input bias current, otherwise the input floats to some arbitrary voltage.
Connect a resistor across the two inputs if it's not connected to the signal source.
 

Thread Starter

sohamkul

Joined Jun 24, 2016
43
That was exactly the problem !! I connected both input terminals to ground and the circuit works like a charm !! Thankyou !
 

Thread Starter

sohamkul

Joined Jun 24, 2016
43
Exactly i didn't get it the first time you said it(dumb of me) but that was the problem ! now i have connected both inputs to ground and reference to ground as well
circuit works well !!
1)but can you please tell me as of to how should i connect inputs should i connect them directly to input pins or should i use resistors ?
2)Can you please suggest a zero pot assembly ?

Thankyou Man !!:rolleyes::rolleyes:
 

Thread Starter

sohamkul

Joined Jun 24, 2016
43
signal is differential and output is basically 0-2 mV and i want the output to be 0-2Volts.
what i want to do is using offset pot i want to set base of output at 1V so that output varies between
1v-3v i am going to connect this assembly to ADS1115 which has 3.3 volt input range only .


thank you !
 
Then use REF. Supply the REF terminal with a LOW Z source from an OP-amp.

Another comment I have to make is:
a) Is your A/D going to ratiometric?
1 and 3 is different than 1 and Vss of 3.3 V

When you make the reference of your A/D equal to the supply voltage, your pretty much assuming ratiometric operation. When you use a precision reference to the A/D, then you get stuff like 0-5V.

So as the power supply varies with load and noise for that matter 50% is always 50%*Vs. Sometimes Vs is say 5 and other times it's 5.1 and 50% is either 2.5 or 2.55 This is the way most automotive sensors work, but it always reads as 50% no matter what the supply voltage is.

So, if you set your A/D reference to the supply voltage, you want REF of the opamp to be a % of the supply.
e.g. A resistive divider amplified x1.

Then based on OP-amp selection, you can decide what to do with the smaller offsets like the IA and the REF amplifier.
 

crutschow

Joined Mar 14, 2008
34,285
As KISS noted, you want to use the REF input to control the output offset.
Use a resistive divider and an op amp follower to generate +1V for the REF input.
 

Thread Starter

sohamkul

Joined Jun 24, 2016
43
I am Extremely sorry for late reply guys (KISS and crutschow) i was busy building other components of the circuit .
Guess what the circuit worked like a charm !!!!:):):):cool::cool:
Thank you so much for your help ~~ :)
 
Top