Gym equipment project

sully1071

Joined Nov 8, 2011
5
I am looking at adding a sensor to the piece of gym equipment. The motion of the player is forward explosively and holding the weight on the way back to his starting position. Currently I have been advised to look at using an LVDT (linear variable differential transducer) to measure the displacement. I need to be able to plot this against time.

Any help with regards to how i could do this would be of help as my knowledge of sensors and signal conditioning is limited....

wayneh

Joined Sep 9, 2010
16,513
There are many ways to approach this and I'm no expert, but maybe a sensor that merely measures ANGLE would be a good choice here. My hunch is that this would be easier to deal with than a linear approach.

Is your ultimate goal to send the data to a computer screen? You could use data acquisition by something like a LabJack U3HV (I have one and love it) under control of a laptop running Excel, for instance, to plot your graph in real time.

Georacer

Joined Nov 25, 2009
5,182
A picture of the exercise of gym instrument would be very helpful.

Do you want realtime data acquisition or can do the analysis later?

The easiest way would be to have a camera shoot vertically against a scaled grid and have the subject perform the exercise between them. With the camera FPS known, you can plot travel distance vs time.

wayneh

Joined Sep 9, 2010
16,513
I've used the video camera approach to plot the acceleration curve for my boat. It's time consuming to analyze the footage, but pretty darn precise and a great choice for something you don't do often, or can't easily arrange in the field. I mean, I could hold the camera with one hand and push the accelerator with the other. Third hand on the steering wheel.

sully1071

Joined Nov 8, 2011
5
The piece of gym equipment i will be using is similar to

http://www.myoquip.com.au/MyoTruk.html

I will be measuring the displacement linearly. I need to acquire the data so i can find the velocity and acceleration.

I will be putting my information into a formula to find the power the player is exerting. (Power = mass x acceleration x velocity). My plan is to be able to show the power thay are outputting on a screen over the piece of equipment as well as recording there results on the computer.

SgtWookie

Joined Jul 17, 2007
22,220
You might want to measure the G-forces directly using a MEMS - then you can extrapolate velocity and acceleration from there.
Example:

You'd need a microcontroller to interface it with, but if you used a uC that you could program in a version of the Basic language, it wouldn't be too tough for a beginner. The player could wear the item, and the measurements taken directly from the players' movement. It could be built pretty small, like perhaps pedometer-sized.

Georacer

Joined Nov 25, 2009
5,182
Sgt, wouldn't it be even better if the device was planted on the gym equipment? That way it couldn't be mishandled by the player and measure the exact displacement of the moving part of the machine.

SgtWookie

Joined Jul 17, 2007
22,220
Georacer,
Yes, that's possible. The whole G-force sensor thing might be overkill though.

Safety related: after looking at that machine again, it seems to me that the rectangular beams the truck rolls on could be a safety hazard; if someone lost their balance while using the machine, they could fall into one of the ends of the beams, and cause some damage to themselves. Yes, there appear to be some black plastic caps on the ends of the beams, but those types of things frequently get lost. It should probably be some kind of rounded cap that wouldn't gouge or puncture someone.

Anyway, while searching for some LVDS options, I came across this which seems interesting:
http://www.activesensors.com/products/contactless-linear-position-sensors
Basically, a contact-less system that reads out like a potentiometer, and operates from a wide range of voltages.

It seems that the travel on LVDS and the PLS systems max out at around 1m/40". I don't know exactly what distance is needed. The costs of LVDS seem to be running around \$400 or so per unit. It seems that one has to contact the company to get pricing information about the PLS systems.

It just occurred to me that a Hall-effect sensor picking up the movement of teeth or pins somehow attached to the truck would be considerably more economical than one of these sensors. It could simply be a stamped steel form that had an evenly-spaced series of holes, serrations, waves, etc. that a Hall effect sensor would pick up.

Georacer

Joined Nov 25, 2009
5,182
Is there a multi-turn trimmer that could be fixed to a string moving along the athlete? The ones I know are too flimsy to do the job.

SgtWookie

Joined Jul 17, 2007
22,220
I thought of that - they wouldn't take the wear & tear.

Might use something like a metal or some sort of fiber/Kevlar? tape that wrapped around two rotating spools in an S-curve; the tape tensioned on one end. The spools could have holes, slots, etc that could be detected by a Hall-effect sensor. The trouble would be trying to conceal something like that. You'd probably have to have a slot cut the length of one of the box beams. Yecch - yet another impractical scenario.

I don't know what kind of cost cap/budget that sully1071 has for this feature.

wayneh

Joined Sep 9, 2010
16,513
The piece of gym equipment i will be using is similar to
Having the athlete push off a load cell under their feet would tell you everything you need to know also (when combined with timing).

Like the ideas of an accelerometer or measuring the angle at the top pivot, I still think it will be better to get away from trying to measure linear position against time. Moving parts, long length of travel, etc. make it tough to do that.

[EDIT] Looking again, it appears there's a pulley under the machine, probably to route the cable that pulls the assembly back towards the athlete. Maybe put magnets on it and count blips?

Last edited:

Georacer

Joined Nov 25, 2009
5,182
How about some optical encoding? Some kind like this:

Attachments

• 33.2 KB Views: 41

wayneh

Joined Sep 9, 2010
16,513
Is that a mirror for the athlete to see the grimace on his own face?

sully1071

Joined Nov 8, 2011
5
I like the idea of the optical linear encoder and am looking into it. I think i will have trouble if i use the lvdt as the movement of the machine will be restricted to half the measurement distance?

would it be easy to plot distance vs. time using the linear encoder? as i need to calculate acceleration and velocity..

wayneh

Joined Sep 9, 2010
16,513
...would it be easy to plot distance vs. time using the linear encoder? as i need to calculate acceleration and velocity..
Once you have the data of position (or force, or angle, or whatever) against time, it should be easy to calculate whatever you want. Getting the data is the hard part.

But you need a computer of some kind. Manual calculation or a hand calculator will likely not be satisfying, except maybe if you have some physics students. You could use a microprocessor perhaps, but you'll need to code it and use a display. My preference would be a smartphone or a cheap laptop to accomplish the calculation, data storage and display chores.

No matter what you choose, you'll have to solve the problems of getting the data into the device (tough), making the calculations (easy) and displaying results with a human-friendly interface (as tough as you want to make it).