Arduino Module/Shield Suggestions

General
1

I am considering starting a project which will involve performing the following tasks, I’ve included what I assume to be correct.

I am looking first observe a signal coming through a wire.

0-10 mV Peak to Peak
0-1.5 mV rms
0-500 Hz Varying frequency range
50-150 Hz Dominant frequency range
If possible, I would like to be able to remove noise from the line and clean up the information.

20-500Hz (Maybe 20-400Hz)
12dB/oct
I would like to be capable of recording the signal.

I hope to be capable of amplifying the recorded signal and sending this same signal back out.

Any common signals would be filtered, while differential signals would be amplified.
A minimum of two channels will be necessary, I may even need somewhere close to thirty channels later, scalability
At the moment it is not necessary to able to perform these from an external battery but this would be an end goal…

Which modules/shields would you suggest for these activities, As a budget cap I would prefer to stay under $100 but willing to go to $250 if I’m using well documented modules/shields which would be quick and easy to setup. I am willing to use multiple modules independently or all in one modules if they simplify the process.

My preference is to minimize the coding necessary on my part, so well developed modules are preferred. What are your thoughts?

2

For cleaning up the signal, one of the more effective ways I can think of is a differential pair. This would also give you room to amplify the 0-10mV to the more standard 0-5V for higher resolution when reading the signal.
You can also throw a couple passive, or active, filters in the path to make a bandpass around 20-400hz which wouldn’t be too hard.
As for specific arduino shields that do this, I’m not sure. But it would only take a buck or two of components and solving an equation or two to do yourself.

3

I don’t think your going to find a shield for this, pretty specific. However, some sort of audio input shield might work. If it has input, it will at least have a front end for the A/D converters.

I’m not sure what you’re trying to do, but you’re talking about rather low signal analog here. The AD converters on an arduino measure a range from 0 to 5V at 12 bits. You’ll need to amplify your signal with gain of 500 to get full scale conversion.

For the frequency range, you’re talking about a filter. You’d also have to eliminate DC or AC couple the signal. You’d need Opamps. Low noise, high impedance jfet input, probably. ‎You’d also need a bipolar supply or a floating bias – with gain like this, I’d suggest bipolar (2 9 volt batteries can be set up this way).

Anyway, sounds pretty intense for arduino-land. Maybe start by using a separate pre-amplifier‎ for the signal and then think about arduino.

L



From: Timothy Gregg
Sent: Saturday, May 26, 2018 12:26 PM
To: Hive13 Hackerspace
Reply To: cincihackerspace@googlegroups.com
Subject: [CHP] Arduino Module/Shield Suggestions

|

  • |

I am considering starting a project which will involve performing the following tasks, I’ve included what I assume to be correct.

I am looking first observe a signal coming through a wire.

0-10 mV Peak to Peak
0-1.5 mV rms
0-500 Hz Varying frequency range
50-150 Hz Dominant frequency range
If possible, I would like to be able to remove noise from the line and clean up the information.

20-500Hz (Maybe 20-400Hz)
12dB/oct
I would like to be capable of recording the signal.

I hope to be capable of amplifying the recorded signal and sending this same signal back out.

Any common signals would be filtered, while differential signals would be amplified.
A minimum of two channels will be necessary, I may even need somewhere close to thirty channels later, scalability
At the moment it is not necessary to able to perform these from an external battery but this would be an end goal…

Which modules/shields would you suggest for these activities, As a budget cap I would prefer to stay under $100 but willing to go to $250 if I’m using well documented modules/shields which would be quick and easy to setup. I am willing to use multiple modules independently or all in one modules if they simplify the process.

My preference is to minimize the coding necessary on my part, so well developed modules are preferred. What are your thoughts?

4

Thanks for input, I went on a brief hunt for a preamp with this much gain and I wasn’t finding much off hand.

I suppose the project doesn’t have to stay in arduino territory, I was simply hoping to throw something together to do this.

It looks like I may have been a bit too ambitious in my thoughts, your information has been very helpful.

Thanks for all of your help!

5

What is the signal? If I know that I can help.

10mv Pk-Pk is not that small. The bandwidth you specified is also pretty narrow, which makes it easier. You’ll probably need to play with some decent opamps and do some gain staging. A gain of 500 is a lot for one opamp circuit, but a gain of 25 followed by another at 20 is very doable.

L

6

The signal is an Electromyogram.

7

This just popped up in my twitter feed:

https://hackaday.com/2018/05/28/muscle-your-way-into-music/

Uses something called a MyoWare sensor. See the tutorial and blog “tabs” here:

https://www.sparkfun.com/search/results?term=MyoWare

I don’t know anything about it.

Matthew

8

Thanks again for your input, I had found this before and did consider but the follow became my concerns.

https://www.sparkfun.com/products/13723

The sensor from the videos of this in use is too restrictive for the intended purposes.
If the sensor input was a cable instead of not directly placed on the body, it might work.

Hacking the input for the sensor is a probable solution.

The second concern came that I’m not sure how many times I can stack this. I may potentially want close to 30 channels. Even if this can be stacked the budget gets well squashed by the number of this sensor I’d need to purchase.

So if this had a hacked input, cheaper price and could be stacked many many times this would perfectly suit the intended goal(I think).

9

Okay. Bio-potential measurement is doable. I’d look at the stuff based on TI ADS1299 and its cousins. http://www.ti.com/product/ADS1299?keyMatch=biopotential&tisearch=Search-EN-Everything. There are a lot of offshoots, 16 bit, 2 channel, 6 channel, etc.

It’s an IC with EMI filter, programmable gain amp, and 24 bit ADC. A lot of folks have been using this or similar for EEG or BCI devices. Simply put, though, getting EMG signals through the skin is doable with any number of EEG/ECG marketed thingies.

For instance, OpenBCI has some nice ones. They work really well. I have a 16 channel 32 bit version I assembled for a project when they first started. We compared it with a couple commercial consumer eeg/ekg’s – it was far superior. Also, the needed common mode rejection channel and other stuff is set up for you…

Lorin