INA measurements done practically does not match datasheet in theory

Thread Starter

fbp

Joined Aug 9, 2023
30
I am using the INA849 IC: https://www.ti.com/lit/ds/symlink/i...15411&ref_url=https%3A%2F%2Fwww.google.com%2F

I have observed the following limitations in this IC when I test it practically on a breadboard.
- When I increase the power rails form +/-5V to +/-15V power rails, the IC gets super hot. I’m guessing it would get defective when the IC gets super hot. The datasheet mention on page 13 that at +/-15V power rails, at 500kHz the output can reach upto 12V pp but the IC starts getting hot when I start increasing the the power rails from +/-5V.

My requirement is that with an input of 500kHz and 4 Vpp to the +/- of the INA (INA+ to the sensor positive and INA- to the ground of the same sensor), I need to see the same values at the output. So, implementing a unity gain INA would give me what I expect. But it does not work.

For the schematic, I am using the EVAL AD8224 Board. I desoldered the Dual INA AD8224 IC, and have implemented the INA849 IC using small wires (since the IC pin numbers do not match) just for testing the IC.

Schematic:
EVAL AD8224: https://www.analog.com/media/en/tec...ation-documentation/854884503AD8224_EVALZ.pdf

Please assist me. Is there something I am doing incorrectly during the test setup? I have checked my connection multiple times and continuity across each line, but it always gives me the same results. Have tested it with two different ICs, both give the same results.

Any help would be highly appreciated.
 

Thread Starter

fbp

Joined Aug 9, 2023
30
Yes, it does work properly at +/-5V power rails. With 0 to 2.4Vpp input range, I get a clear sine wave at the output. After 2.4V input, the output sine wave starts getting distorted which does reflect the datasheet of Vcm operating input voltage: -2.5V < +/-Vin < +2.5V
 

crutschow

Joined Mar 14, 2008
33,358
My requirement is that with an input of 500kHz and 4 Vpp to the +/- of the INA (INA+ to the sensor positive and INA- to the ground of the same sensor)
If one input is connected to ground then why use an INA since you don't have a differential signal, or the common-mode rejection the INA provides.
All you would need is an op amp.
 

crutschow

Joined Mar 14, 2008
33,358
With 0 to 2.4mVpp input range, I get a clear sine wave at the output. After 2.4mV input, the output sine wave starts getting distorted which does reflect the datasheet of Vcm operating input voltage: -2.5V < +/-Vin < +2.5V
I'm confused.
You originally stated you needed a gain of one.
What gain are you running for this test?
 

Thread Starter

fbp

Joined Aug 9, 2023
30
I wanted to eliminate the common mode noise across both the positive and negative wires of the sensor. My circuit design is that I have a sensor connected to a 2.5m cable, to a centre tapped transformer for patient isolation and then to the +/-INA with gain 1.

I also choose the INA since it has better CMMR than a regular opamp.

Please correct me if I have mentioned anything incorrectly here.
 

Thread Starter

fbp

Joined Aug 9, 2023
30
I apologize for the confusion. I made a mistake in my value in the comment. I edited the values in the same comment. I posted the same comment below again.

Yes, it does work properly at +/-5V power rails. With 0 to 2.4Vpp input range, I get a clear sine wave at the output. After 2.4V input, the output sine wave starts getting distorted which does reflect the datasheet of Vcm operating input voltage: -2.5V < +/-Vin < +2.5V
 

crutschow

Joined Mar 14, 2008
33,358
I wanted to eliminate the common mode noise across both the positive and negative wires of the sensor. My circuit design is that I have a sensor connected to a 2.5m cable, to a centre tapped transformer
The transformer will eliminate any common-mode noise, so I see no reason for an INA.
Why is the transformer center-tapped?
Yes, it does work properly at +/-5V power rails.
Then I don't know what's causing it to overheat, other than a bad device.
Do you have more than one you can test?
Where did you buy it?

.
 

Thread Starter

fbp

Joined Aug 9, 2023
30
The transformer is center-tapped (1:1) since I wanted the same output as the input. My reason for including the transformer is only for patient isolation and common mode noise rejection.

I included the INA only for good CMRR, to keep an output voltage limitation by the power rails in case unexpected circumstances of overshoot in the input due to unknown reasons, and also for amplification is needed later in the future (for now it is only unity gain).

Yes, I do have tested three different ICs. All three give me the same input-output relationship. The IC overheats after +/-5V, but there is no effect in the output voltage increasing with respect to the input


I purchased from the distributor Mouser Electronics. The manufacturer is Texas Instruments
 

Thread Starter

fbp

Joined Aug 9, 2023
30
Would the error be because of the component or the circuitry I implemented?

I tried testing the IC by itself on a breadboard using jumper wires. It still gives me the same resuts.
 

crutschow

Joined Mar 14, 2008
33,358
I included the INA only for good CMRR
As I stated, with the transformer than is no common-mode signal that requires an amplifier with CMRR.
There is no common-mode voltage.
Would the error be because of the component or the circuitry I implemented?
Likey your circuitry.
Please post the schematic of the complete, exact, circuit you are testing, including any output loads.
 

Thread Starter

fbp

Joined Aug 9, 2023
30
I tested the IC on a breadboard independently as well. It gives me the same results.

For example, if I go to +/-10V power rails
Function generator input of 500kHz and 3Vpp

According to the datasheet, the Vcm operating input voltage is as below
1691621997760.png
So ideally, my output max range can be till +/-7.5Vpp with Power rails of +/-10V
But after 2.4Vpp, the output sine wave starts getting distorted.
 

Papabravo

Joined Feb 24, 2006
20,607
What is it about:
" Please post the schematic of the complete, exact, circuit..." that you do not understand? You are displaying an extreme level of entitled obstinacy!
 

Thread Starter

fbp

Joined Aug 9, 2023
30
Yes, I was going to post it here. I haven't created a schematic of the changes I implemented on a pre-existing Eval board. But once I make it I will send it here. Fow now, I can give the rough design as below.

Schematic:
EVAL AD8224: https://www.analog.com/media/en/tec...ation-documentation/854884503AD8224_EVALZ.pdf
AD8224 IC desoldered and INA849 IC implemented.
The addition on the pre-existing Eval board is the Transformer and the INA849 IC

Sensor signal as input ->Transformer -> INA -> Output to Oscilloscope
 

Attachments

Last edited:

schmitt trigger

Joined Jul 12, 2010
772
Soldering wires in a fine pitch (0.65mm) SMT footprint without creating any shorts or intermittent connections is, to put it mildly, very challenging.

Also, the datasheet is very adamant about the tight layout requirements.

Could you consider designing a simple, dedicated board instead?
Small boards from PCBway and EasyEda are only a couple of bucks.
 

Thread Starter

fbp

Joined Aug 9, 2023
30
I have checked the continuity across each line and the solder joints closer to each other on the PCB to ensure they are not shorting. I cross-checked only the IC (brand new one) on a breadboard separately using jumper wires to check the output characteristics v/s input characteristics for the same parameters as set on the PCB. Both the test methods give the same results.

I will anyways try it implement a custom PCB for the modified schematic.

Thank you everyone for your help.
 
Top