AD8226 Instrumentation Amp Not Outputting Correct Voltage

Thread Starter

mintaro

Joined Jun 17, 2022
5
I've been working on this for the past few hours and I'm at my wits end.

Basically, I'm trying to amplify an input voltage with a 100-ish gain using an AD8226 on a breadboard. I'm using a +/- 5V supply to power and op amp and using a 10 mV test for the input. I'm also using an oscilloscope to measure the output voltage, but the output reading is nowhere close to what I should be getting.

For example, a 10 mV input generates a 1.280 V reading, but a 20 mV input generates a 1.680 V outcome, and a 30 mV input generates a 1.760 V output and so on.

I've tried different AD8226 chips and I've been getting the same result. I'm thinking maybe my circuitry is wrong, but I haven't been able to find out what's wrong with it.

Any help would be greatly appreciated, this has been a gigantic pain. Here's a pic.

Red = S+
Black = S-
Green = ground
Light orange = Vin+
Dark orange = Vin-
 

Attachments

MrSalts

Joined Apr 2, 2020
1,840
I've been working on this for the past few hours and I'm at my wits end.

Basically, I'm trying to amplify an input voltage with a 100-ish gain using an AD8226 on a breadboard. I'm using a +/- 5V supply to power and op amp and using a 10 mV test for the input. I'm also using an oscilloscope to measure the output voltage, but the output reading is nowhere close to what I should be getting.

For example, a 10 mV input generates a 1.280 V reading, but a 20 mV input generates a 1.680 V outcome, and a 30 mV input generates a 1.760 V output and so on.

I've tried different AD8226 chips and I've been getting the same result. I'm thinking maybe my circuitry is wrong, but I haven't been able to find out what's wrong with it.

Any help would be greatly appreciated, this has been a gigantic pain. Here's a pic.

Red = S+
Black = S-
Green = ground
Light orange = Vin+
Dark orange = Vin-
Did you read page 20 of the datasheet and see how not to set the voltage reference? And how to set the voltage reference?
 

Papabravo

Joined Feb 24, 2006
18,806
The picture is not very helpful. A schematic diagram would be beneficial.
EDIT: This is the way I would do it in the absence of any additional information. It seems like either your gain setting resistor is wrong or the input voltage source is not well controlled and calibrated.
1655491295344.png
 
Last edited:

Thread Starter

mintaro

Joined Jun 17, 2022
5
Did you read page 20 of the datasheet and see how not to set the voltage reference? And how to set the voltage reference?
I did but I'm a bit confused as to what it is saying. Can I not just set the reference to ground, since it would essentially just be acting like a differential op amp?
 

Papabravo

Joined Feb 24, 2006
18,806
I did but I'm a bit confused as to what it is saying. Can I not just set the reference to ground, since it would essentially just be acting like a differential op amp?
The datasheet shows numerous examples of "correctly" using the part with REF connected to GROUND. What else is connected for the purposes of your test?
 

Thread Starter

mintaro

Joined Jun 17, 2022
5
The datasheet shows numerous examples of "correctly" using the part with REF connected to GROUND. What else is connected for the purposes of your test?
I have the +/-5 V supply connected to the S+ and S- terminals. I have a 0.1 mV DC supply voltage connected to my In- and In+ terminals, my 470 ohm resistor and my ref pin connected to ground.

Lastly I have an oscilloscope probe connected to my output and reference pin (which is ground connected). So far those are the only components connected.
 

Papabravo

Joined Feb 24, 2006
18,806
I have the +/-5 V supply connected to the S+ and S- terminals. I have a 0.1 mV DC supply voltage connected to my In- and In+ terminals and my ref pin connected to ground.

Lastly I have an oscilloscope probe connected to my output and reference pin (which is ground connected). So far those are the only components connected.
I've been trying to encourage you to provide a schematic, but you seem resistant to the idea. Is there a reason for that? Is your setup essentially the same as the simulation schematic I provided in post #3? If not you should try to make it so. IMHO the simulation and the real parts should agree to at least a 1st order approximation. You are there and we are not.
 

Thread Starter

mintaro

Joined Jun 17, 2022
5
I've been trying to encourage you to provide a schematic, but you seem resistant to the idea. Is there a reason for that? Is your setup essentially the same as the simulation schematic I provided in post #3? If not you should try to make it so. IMHO the simulation and the real parts should agree to at least a 1st order approximation. You are there and we are not.
Yes, the setup I have is the same one that you provided. The issue I am having is translating the schematic into a working breadboard model which is why I provided a picture of the breadboard.
 

crutschow

Joined Mar 14, 2008
30,116
I have a 0.1 mV DC supply voltage connected to my In- and In+ terminals,
Is that supply connected to ground through a resistor (It appears to be floating)?
The amp input input terminals cannot float, they need a path to ground for the small input bias current.
The inputs may be differential but they still need a path to ground.
 

Papabravo

Joined Feb 24, 2006
18,806
Yes, the setup I have is the same one that you provided. The issue I am having is translating the schematic into a working breadboard model which is why I provided a picture of the breadboard.
Yes, but the picture has no useful details that I can make out. AD provides a test jig for their part with LTspice. These are the results for a gain of 10 with an offset of 0.5V, that is 0.5V + 2 VP-P:
1655493428442.png
 
Last edited:

Papabravo

Joined Feb 24, 2006
18,806
Is that supply connected to ground through a resistor?
The amp input input terminals cannot float, they need a path to ground for the small input bias current.
The inputs may be differential but they still need a connection to ground.
This is clearly shown on the datasheet for inputs that are capacitively coupled. The DC path to ground is a universal requirement for opamp circuits.
 

Thread Starter

mintaro

Joined Jun 17, 2022
5
Is that supply connected to ground through a resistor (It appears to be floating)?
The amp input input terminals cannot float, they need a path to ground for the small input bias current.
The inputs may be differential but they still need a path to ground.
Yes, apologies for the messy circuit but the input voltages are grounded. The negative input is connected to a wire connecting it to the ground rail.
 

eetech00

Joined Jun 8, 2013
3,232
Yes, apologies for the messy circuit but the input voltages are grounded. The negative input is connected to a wire connecting it to the ground rail.
Instead of connecting pin 5(-VS) to -Supply , connect it to ground (single supply configuration).
Then ground the reference pin.

Connect pin 1(-IN) to ground. If you input 10m into pin 4 (+IN), the output should be 1v.

1655503889050.png
 

MrSalts

Joined Apr 2, 2020
1,840
I think you need to verify which through-hole pin corresponds to each surface mount pin on the chip before you go any further.
 

ericgibbs

Joined Jan 29, 2010
15,726
I have the +/-5 V supply connected to the S+ and S- terminals. I have a 0.1 mV DC supply voltage connected to my In- and In+ terminals, my 470 ohm resistor and my ref pin connected to ground.
Hi mintaro,
From your description of the mVolt source, it is highly likely your test ource voltage is not referenced to Ground/0v

Do the following check:
Use dual +/-5V supplies for the amp.
Ground the Vref pin.
Connect +In pin of the amp to 0V via 1k, do the same for the -In pin. ie: 1k to 0v

Measure the Vout of the amp, it should be close to zero volts.

E
 
Top