How to limit bandwidth of an instrumentation Amplifier

Thread Starter

fieryfire

Joined Feb 14, 2017
75
Hello All,

Could someone explain how to limit the bandwidth of an in-amp. With Op-amps all you need to do is to insert a capacitor between the input and the output in order to lower the bandwidth and thus reduce the noise over the unnecessary bandwidth. But can someone please help me understand how to do this with an in-amp?
 

Thread Starter

fieryfire

Joined Feb 14, 2017
75
Hello Eric,

I in fact tried to simulate this with TINA TI. But for some reason this method doesnot seem to limit the bandwidth of the in amp. If you take a
look at the following circuit. The difference filter is cut of at 300Hz. But my output frequency does not cut off at 300Hz

upload_2019-7-8_12-55-50.png
 

Thread Starter

fieryfire

Joined Feb 14, 2017
75
hi,
To get a low frequency cut off the the filtering is extreme, wil require low leakage caps.
What is the full apllication.?
E
Hello Eric,
But then again, the simulation needs to take into account of a perfect cap. but it does not seem to be so.

My application is to use the in-amp for signal conditioning quarter bridge strain gauges of 350Ohm and to have the noise as low as possible. I chose 300Hz as the bandwidth cut-off inorder to be able to measure dynamic frequencies measured by the frequencies which happen at under 100Hz.
 

Analog Ground

Joined Apr 24, 2019
169
Hello Eric,
But then again, the simulation needs to take into account of a perfect cap. but it does not seem to be so.

My application is to use the in-amp for signal conditioning quarter bridge strain gauges of 350Ohm and to have the noise as low as possible. I chose 300Hz as the bandwidth cut-off inorder to be able to measure dynamic frequencies measured by the frequencies which happen at under 100Hz.
Looks like the same IA and application as your other post. The AD8428 has special filter connections which other IAs do not have. Are you aware of this feature on the AD8428? If not, it is covered in the application section of the data sheet. The filtering is after an initial gain of 100. If you are going out of range after a gain of 100 then you will need another technique.
 

Thread Starter

fieryfire

Joined Feb 14, 2017
75
Looks like the same IA and application as your other post. The AD8428 has special filter connections which other IAs do not have. Are you aware of this feature on the AD8428? If not, it is covered in the application section of the data sheet. The filtering is after an initial gain of 100. If you are going out of range after a gain of 100 then you will need another technique.
Hello Analog, :)

Great to see you here again. Yes i am aware of the special filter pins on AD8428. But i did the simulation test on INA2128 another instrumentation amplifier that i used. The INA2128 does not have the special filter inputs.

The thing is when i use the TINA TI, to simulate a bandwidth limit for an op-amp, it works. But i do not know if it applies to the in-amp also.
 

Analog Ground

Joined Apr 24, 2019
169
Hello Analog, :)

Great to see you here again. Yes i am aware of the special filter pins on AD8428. But i did the simulation test on INA2128 another instrumentation amplifier that i used. The INA2128 does not have the special filter inputs.

The thing is when i use the TINA TI, to simulate a bandwidth limit for an op-amp, it works. But i do not know if it applies to the in-amp also.
Analog Devices has a SPICE model for the AD4828 but I'm sure you already know of it. I would have a go at running the model in TINA-TI but I am an LTspice user. Perhaps someone from the forum can weigh in on using the model with TINA-TI?

As an aside, if TINA-TI does not have strong noise analysis capability, I suggest switching to LTspice which has very useful noise analysis. Also, as a blatant recommendation for AAC, there is recently a series of articles on using noise simulation in LTspice.

Edit: I just saw your post on the other thread about using LTspice. Let me digest it.
 

Thread Starter

fieryfire

Joined Feb 14, 2017
75
Analog Devices has a SPICE model for the AD4828 but I'm sure you already know of it. I would have a go at running the model in TINA-TI but I am an LTspice user. Perhaps someone from the forum can weigh in on using the model with TINA-TI?

As an aside, if TINA-TI does not have strong noise analysis capability, I suggest switching to LTspice which has very useful noise analysis. Also, as a blatant recommendation for AAC, there is recently a series of articles on using noise simulation in LTspice.
Hello Analog,

I am an LTSPICE user myself. In a way i do a little both.

But it appears to me as though TINA TI has more features with respect to noise analysis. For example. i can generate the following graphs from it which is not very easy to do so in LTSpice. OR maybe i could be wrong.

for ex: Its the total noise that gives a good information on noise analysis and to understand ways to minimize noise. But the feature of total noise does not seem to be available in LTSpice

upload_2019-7-9_12-59-13.png
upload_2019-7-9_12-59-26.png

upload_2019-7-9_12-59-46.png

upload_2019-7-9_12-59-57.png
 

Analog Ground

Joined Apr 24, 2019
169
Here are the AAC LTspice articles. LTspice can compute total noise which is fundamental. The first article below has a section on total output noise. The method in the article is the simplest. It can also be done with a .MEAS directive which is used extensively in the third article.

https://www.allaboutcircuits.com/technical-articles/Noise-Analysis-Using-LTspice-Tutorial/

https://www.allaboutcircuits.com/technical-articles/how-to-perform-transient-analysis-noise-simulation-LTspice/

https://www.allaboutcircuits.com/technical-articles/using-ltspice-for-amplifier-noise-measurement/
 
Top