I'm no expert on EMG, instrumentation amps or filter design, but I would have thought that, at the very least, you would need a deep notch filter to attenuate the local mains frequency drastically. Perhaps a switched-capacitor filter?
It's unfortunate that mains frequency and its low harmonics fall right in the middle of the useful EMG spectrum.

 

I'm no expert on EMG, instrumentation amps or filter design, but I would have thought that, at the very least, you would need a deep notch filter to attenuate the local mains frequency drastically. Perhaps a switched-capacitor filter?
It's unfortunate that mains frequency and its low harmonics fall right in the middle of the useful EMG spectrum.

Thank you, sir, for your reply. But can you please check my circuit and tell me whether it is correct or not.

 

Here's a simulation of the frequency response of the filtering part of your circuit :-

 

Here's a simulation of the frequency response of the filtering part of your circuit :-
View attachment 137822

I think This is correct!
Can you please tell me the name of tool that you have used for simulation?

 

LTspice. Many members here use it. A free download from Linear Technology.
Yahoo has an LTspice User Group where you can get help and free downloads of model libraries.

 

LTspice. Many members here use it. A free download from Linear Technology.
Yahoo has an LTspice User Group where you can get help and free downloads of model libraries.

Is my circuit correct for basic measure of EMG signal?

 

Try it and see: but as I said in post #21, I think you'll need a steep notch filter. Other members may have a better informed view.

 

Try it and see: but as I said in post #21, I think you'll need a steep notch filter. Other members may have a better informed view.

Thanks a lot, sir, I will try it.

 

Here is my final amplifier circuit with transient characteristics.
Please check it and suggest me changes.
Will replace INA106 with AD620 in the PCB design.

 

Apart from the filter adequacy and output limitation issues already raised, I do wonder if opamp offset errors might be a problem because of the DC coupling between stages.

 

Apart from the filter adequacy and output limitation issues already raised, I do wonder if opamp offset errors might be a problem because of the DC coupling between stages.

How to keep a check on input offset voltage in TL072.
In case of 741 we use a trimpot input voltage compensation but what should I do for Tl072?

 

How to keep a check on input offset voltage in TL072.
In case of 741 we use a trimpot input voltage compensation but what should I do for Tl072?

You can avoid the problem by AC coupling the stages.

 

You can avoid the problem by AC coupling the stages.

What value of capacitor should i use? will a capacitor of values 1uF will be enough at the last stage of amplification i.e before U4.

 

If you use a cap to AC couple to U4's non-inverting input, don't forget to provide a path for the input's bias current, e.g. 100k or so to ground. 1uF should be more than adequate for the cap.

 

If you use a cap to AC couple to U4's non-inverting input, don't forget to provide a path for the input's bias current, e.g. 100k or so to ground. 1uF should be more than adequate for the cap.

Can you please give a circuit diagram that can help me get a better understanding of how to place the ac coupling?

 

I'm no expert on EMG, instrumentation amps or filter design, but I would have thought that, at the very least, you would need a deep notch filter to attenuate the local mains frequency drastically.

Actually, not so much. Besides the amplitude response issues, a notch filter injects lotsa phase distortions for a couple of octaves on each side of the notch. Since the distortions are frequency-dependent, different harmonics of the baseband signal are delayed by different amounts, altering the waveform after the filter compared to before. Proper wire harness design, a well-tuned IA front end, and a linear phase lowpass filter outside the signal bandwidth work well together.

ak

 

The input to U2 is already AC coupled and as U2 and U3 have DC gains only slightly above unity there should be no problem with the input to U4. However U4 has a DC gain of 100 so its output could have a significant offset which as the circuit stands would be passed on to U5 so I think that it is the input to U5 that needs the AC coupling.

 

I think that it is the input to U5 that needs the AC coupling.

Good point.

 

The input to U2 is already AC coupled and as U2 and U3 have DC gains only slightly above unity there should be no problem with the input to U4. However U4 has a DC gain of 100 so its output could have a significant offset which as the circuit stands would be passed on to U5 so I think that it is the input to U5 that needs the AC coupling.
View attachment 137982

Thanks a lot sir!

 

How do I use the ceramic capacitors? and how to make a ground. should i connect it to the ground in Arduino or leave it open?