Question about Designing EMG data acquisition

[Tom.G]

Hi Mo_Tuk,
Could you post a photograph of the assembled amplifier and another photograph of showing it connected to you? Please show where the leads connect to you and to the amplifier. With only the wiring diagram, it is very hard for us to guess what may help to get it working.

The waveform you posted earlier (https://www.physicsforums.com/attachments/tek0003-jpg.232273/) looks like there is a DC power supply nearby that is making some electrical noise, maybe a battery charger or computer, or maybe even an LED lamp!

Cheers,
Tom
edit: fixed typo

 

[jim hardy]

I hope this screenshot of the circuit would be clear,

we trust you tied the junction of your batteries to circuit common

for now I just want to see clear EMG signal from my bicep muscle

and shielded your signal wires

i added those features to your drawing

 

[Mo_Tuk]

Hi Mo_Tuk,
Could you post a photograph of the assembled amplifier and another photograph of showing it connected to you? Please show where the leads connect to you and to the amplifier. With only the wiring diagram, it is very hard to for us to guess what may help to get it working.

The waveform you posted earlier (https://www.physicsforums.com/attachments/tek0003-jpg.232273/) looks like there is a DC power supply nearby that is making some electrical noise, maybe a battery charger or computer, or maybe even an LED lamp!

Cheers,
Tom

Thanks @Tom.G
do yo mean photograph of the circuit connected on the breadboard??
and about the electrical noise I think the laptop was near to my circuit and plugged into charger.

 

[Mo_Tuk]

we trust you tied the junction of your batteries to circuit commonand shielded your signal wires

i added those features to your drawing

View attachment 232407

Thanks @jim hardy
for the connection you made of electrodes A&B to reference, would that help to reduce noise or something else??

 

[jim hardy]

for the connection you made of electrodes A&B to reference, would that help to reduce noise or something else??

The red dashed lines do NOT represent connections to the electrodes,
instead
they represent shields around the wires bringing signal from the electrodes to your circuit board.

The purpose of a shield is to physically surround the signal wire so the can be no capacitve coupling into it.
Recall from basic physics that no E field can exist inside a closed conducting surface - google "Faraday Shield".

Using a twisted pair to route both signals together will reduce magnetic coupling.
That's why "twisted shielded pair" is so popular.

Your 50 hz noise is probably a combination of pickup that's
capacitively and inductively coupled into the input wires ( and gets amplified 1100X) ,
and
capacitively coupled across the power transformer in your oscilloscope's power supply.

Shielding the input wires will help .

old jim

 

[jim hardy]

looks like there is a DC power supply nearby that is making some electrical noise, maybe a battery charger or computer, or maybe even an LED lamp!

Emphasizing @Tom.G 's comment"Driven Shield" is a method that's used in precision equipment .
It's shown in the INA111 datasheet

the cylinder surrounding the signal wires represents a shield that is driven to nearly average voltage of the two signals
which minimizes voltage between the shield and signal wires
minimizing capacitive coupling of current into the signal wires.

INA111's input impedance is so high that your signal wires act like an antenna .
Shielding hides them from electric fields by creating a Faraday Cage.
Twisting them reduces their sensitivity to magnetic fields by reducing the area(hence the magnetic flux) they encircle.
That's why we EE's like twisted shielded pair for signal wires..

Try grounding shields first to see if it helps. Driven shield should give even better results.

old jim

 

[Tom.G]

do yo mean photograph of the circuit connected on the breadboard??

Yes.

 

[Tom.G]

about the electrode cables would you suggest any specific type of cable that less noise produced ?

Belden 2221 or 11700A would be good choices but will be hard to find and the 2221 costs about $2 to $3 USD per foot ($6 to $9.3 USD per meter). A little bit cheaper if you buy 1000 feet (330 meters).

I suggest you first try whatever shielded wire you can get and twist the wires together, about 3 or 4 twists per foot (10 per meter) should help.

A possible source is: https://www.nassauelectrical.com/pr...ingle-pair-bc-stranded-30x40-pvc-jacket-cable

 

[Mo_Tuk]

Okay, there are two things that you should approach this with, IMO:

• You should look more at shielding and balancing your probe lines to cut down on the AC Mains noise pickup -- if you post your full circuit diagram including Earth ground connections and shielding, we should be able to help with that
• You should consider digitizing the waveform output from your circuit and doing DSP processing to filter out the AC Mains noise. You can do much higher order filters and adaptive filtering with DSP routines. Do you have experience with programming microcontrollers (uCs) in C or Assembly code?

I have also another type of electrodes (Covidien Kendall Q-Trace Diagnostic Tab Electrodes Resting ECG Electrodes) , is that type better than (BioProtech)?

 

[berkeman]

I have also another type of electrodes (Covidien Kendall Q-Trace Diagnostic Tab Electrodes Resting ECG Electrodes) , is that type better than (BioProtech)?

As long as you are using disposable medical grade electrodes and not something homemade, it should be fine.

 

[Mo_Tuk]

Screenshot of the circuit that I connected on Breadboard, I hope it's clear.

 

[Mo_Tuk]

This is screenshot of the activity of my bicep muscle (Resting and Contraction)

 

[Tom.G]

WOW! You did it! Congratulations.

Can you post the details and schematic of what you did?

Cheers,
Tom

 

[Mo_Tuk]

View attachment 233125

WOW! You did it! Congratulations.

Can you post the details and schematic of what you did?

Cheers,
Tom

Thanks @Tom.G
I will post the schematic, now I need to know how to get data from the oscilloscope so that I can deal with it using Arduino, the oscilloscope used is "Tektronix TDS 2024B" and I just used Print button to capture the signal.

 

[Mo_Tuk]

The attached image is the schematic for EMG circuit that worked, and I have question about DC level shifter circuit should I add it ??
so that (e.g. the output will be from 0 to 5V max), because Analog to Digital Converter "ADC" of the arduino can only accept this range.
any Suggestion ...

 

[Tom.G]

The attached image is the schematic for EMG circuit that worked, and I have question about DC level shifter circuit should I add it ??
so that (e.g. the output will be from 0 to 5V max), because Analog to Digital Converter "ADC" of the arduino can only accept this range.

For your particular use of detecting different muscle activity you would not need a level shifter.

However the converter on the Arduino board may not like a negative-going signal. I would say try connecting to the Arduino to see if you get reasonable results, if the results are bad then add a level shifter. (Unless of course you want to add a level shifter as part of learning more -- often a good idea. )

Cheers,
Tom

 

[Mo_Tuk]

Hello

I added the dc lavel shifter and the circuit works fine with arduino using ADC of the arduino, but now I have to do two things:

1) comparing the result with other built-in EMG shield by using correlation technique "Auto-Correlation"
2) add some realistic application to prove that the system is practical, any idea ??