Author Options:

How to make a cheap myoelectric sensor like the ones they use on the exoskeleton Hal-5? Answered

I'm working on building an exoskeleton like Hal-5. I have almost the whole thing planned out, but I don't know how/where to make/buy a cheap myoelectric sensor that will measure the faint bio-electric signals coming from the brain to the muscle. Any information would be apprieciated.


Hi. I too have your dream lol. Like many people I would love to have my own iron man powersuit. haha, oh well, dreams are free. I tried making a small myoelectric unit a few years ago for a science and technology competition. It was ideally designed as part of a much larger project to aid disabled people. I used disposable sensor pads, such as you would see on hospital patients. In total i used 3 pads. Two pads attached to the skin over the bicep muscle, and the third (a ground or 'neutral') attached to the abdomen since it was an almost unrelated area of the body. The two bicep pads (each with only one wire connecting out) connected to the two inputs of a common op-amp IC. I amplified the signal from the op-amp with a series of op-amps. Then I connected the op-amp's output with a comparator IC. This would compare the signal created by the contraction of the bicep muscle with a preset resistance value so that when the signal peaked above the set level it would trigger a relay. From there you would be able to switch anything you like on or off, such as pneumatics or hydraulics. In the end I couldn't amplify my signal high enough to switch on the relay. However, I haven't attempted to fix this. I would love to continue with this in the future. If you are interested in my project any further I can provide more information. Cheers, Dan


I think this is what you need. Sparkfun also sells a neoprene garment to carry the sensors.

May be a little late, but recently has been developed something like you need, the MYO, https://getmyo.com/, although not available yet, it will be in the year.
Am also interested in the development of a basic exoeskeleton, I hope we can help each other to achieve.

José Luis

Hi Gearhead88 I would love to hear more about that project. I am using lego mindstorms nxt for my experiments, and the only thing I can't figure out is the sensor. Any info would be awesome. Thanks, Sam

Hi again, I will endevour to post an instructable on this in the next month or so. Thanks, Dan

There is an open source eeg project on the web somewhere. The important component there is the active electrodes they use, amps are put right on the electrodes so the signal is amplified before it gets the chance to pick up too much noise.

I have a similar desire to build a powered exo-skeletal aparatus, but instead of using sensors to detect muscle impulses I use tactile switches to respond to minute physical movements of the pilot.


9 years ago

They used 20 years and millions of dollars to make the exoskeleton. Dont expect that you can easily make one yourself.

not quite true, i have to disagree, i myself have a plan for a reasonable exoskelleton, one that is made with easilly acquirable material, such as old nokia cellphone batteries and electric window motors. scraps and an old back pack, 14 to 20 gauge wiring. i have to admit my circuitry skills are at begining level, but my mechanical and physics skills are quite impressive if i don't say so myself, so unless i could get good circuitry help i would have to base my exoskelleton on switches straight from batteries to the motors, the beauty would be more in the design of it. as for the sensors i would use for such a project would be something simular to a switch that closes the circuit once pressure is applied to it, meaning that if placed in the right place when you move your fourarm for instance it would apply pressure to the switch moving it in the same direction that your arm moves in increasing the force of your movement. trust me it's not really difficult. not at all.

No, to develop it. Not make it. It probably only to a few thousand dollars to make it. They have developed it so now he just has to copy it.

A worthy comment, but it's still fun to mess around small scale in such things

@Gearhead88, You are doing it wrong. If a signal can be read, it can be input into a computer. From the computer a very simple program can redirect the signal to a micro-controler, including the powerlevels. It in turn can operate the device meant to do the work. I.E. hydrolics, pnumatics, or servo-mechanism.

Measure the range of motion of the aparatus, and set 0 at full extension and 100 at full contraction. Measure the voltage output from the person, and input the data into your computer. Viola! You have a readily made system that you can tweak and modify yourself to suit your needs.

Remember to test it before you try to use it. Strapping yourself to an exoskeleton than has a greater range of motion than you do may result in serious bodily injury, and death. If you should happen to ignore the warning and do it anyway, please post a video so I can watch! (Death and dismemberment have always made me laugh!)

its pretty easy, you only have to have a instrumentation  amplifier then a signal filter (high pass between 15 - 20 hz and a low pass with cutout frequency at 200 hz) then a noise removal filter (notch filter i don't know in usa i believe its 50 hz, here in Mexico is 60 hz)  then amplify and take the signal to a ADC (maybe in a arduino)

Instead of myoelectric which has problems with skin galvanance and nasty problems such as sweat, how about an array of pressure switches based on anti static foam( the black foam they sometimes send ICs in? When that stuff is squeezed, its resistance goes way down. Put an array inside the suit (package the foam so sweat won't get into it. The harder you push against the inside of the suit, the lower the resistance of the sensor array. This could be used to activate a cascade of relays to power up a motor or activate a solenoid valve.