FES Drop Foot Project

[nyoungblood]

Hello eevblog forum! Long time listener, first time caller.

I am in the planning stages of a project that I am making for my dad. He has drop foot from a stroke that happened when he was 15 and has to use a brace to walk. Recently, he tried a device similar to what they have here: http://www.walkaide.com, but it isn't covered by insurance and costs a lot (several thousand). It uses electrical stimulation to actually contract the muscles which could reverse atrophy in his leg. As a reasonably resourceful person with a bit of electrical experience, I think I could make a system that is comparable for a fraction of the cost. The main design parameters are: the device needs to be portable, about the size of a smart phone, and last for at least a day between charges. Looking at some papers on FES, the electrical stimulation uses 50-100 mA current spikes that last for ~350 microseconds at a frequency of 40 Hz. Since the resistance of skin/tissue is high, the voltage needed is 100-200V. If I can create a highly efficient boost converter, I should be able to run this and the control circuitry for at least 8 hours with a cell phone battery or two.

Now to the actual question: I am wondering what type of boost converter would be best for this project. Would it just simply be a transistor, inductor, capacitor, and some resistors? Or is there a special kind of boost converter designed for portable applications? I was thinking of maybe using a similar design to those disposable camera flash circuit boards.

Thanks in advance for your ideas / help!

[HalfSpace]

Hi nyoungblood,

The simple circuit below is from Bing.com it’s a muscular bio-stimulator using a 555 timer. You will also need a tilt sensor and adjustable delay/duration circuit (maybe another 555) to trigger the bio stimulator circuit. I can do a simulation of a full circuit in LTSpice if this is the type of design you are looking for.

HalfSpace 

[nyoungblood]

Hi HalfSpace,

Thanks for your response! That circuit looks a lot more straight forward than some of the others I've seen in papers. I am thinking about incorporating a low power accelerometer and a teensy board to control the output. It might make more sense to have a dedicated 555 timer like you suggested to adjust the delay though. Is it possible to estimate the power consumption of the circuit with Spice? I haven't tried simulating anything more complicated than a resistor in Spice...

[VK5RC]

To contract a muscle you need good skin conductivity,  this makes it potentially dangerous. Be careful to ensure conductive paths are well away from the heart,  a bit of reading re medical electronic devices and the classes would be essential.

[tggzzz]

I am thinking about incorporating a low power accelerometer and a teensy board to control the output.

The result of a malfunction could be quite "entertaining". I believe you can sell videos of pratfalls to TV stations for $250

I wonder what the "product" liability would be if your device injured an uninvolved third party? Or whether your insurance would cover it?

[HalfSpace]

Hi nyoungblood,

The circuit below has been simulated in LTSpice using 555 timers. Average current drawn during output with 9V supply is 15mA. V2 is a trigger pulse which would be replaced with a tilt switch/sensor. The first 555 provides an adjustable delay between trigger point and start of output signal. The second 555 provides an adjustable duration length of the output signal. The third 555 provides the output signal with an adjustable intensity of the output signal. For the output transformer I have used in the simulation L1 and L2 with a coupling factor K1 as I do not have an LTSpice transformer model that works and R4 on the output is just a dummy load.

This is a simple and cheap circuit to make up on a breadboard for experimenting but please note it can produce very high voltages and all precautions must be made not to injure yourself or anyone else with this device.

As a starting point maybe build up a circuit that does not have a high voltage output but a LED or buzzer to indicate when the leg muscle would contract to resolve any trigger/timing issues etc.

HalfSpace 

[nyoungblood]

Very cool. Thanks HalfSpace! I think I will start a post in the "Projects" part of the forum when I have some results worth sharing. Thanks again for your help!