Creating a Velocity Measurement Tool

[bcouilla]

Hey Everyone!

I am new to the forum and was hoping to get some help on a project I am working on.

The idea is this:

1.) Floor Unit houses a string/cable that attaches to a bar about 6-7 ft high off the ground
2.) Bar then moves load up and down for 1 repetition.
3.) The movement of the string generates a distance and time of the repetition
4.) Unit relays information to device that outputs (displays) velocity (preferably average and peak velocities)

I was thinking of using a String potentiometer but my thoughts tell me that how does the differentiate between repetitions.

Thanks!

 

[Gryd3]

Hey Everyone!

I am new to the forum and was hoping to get some help on a project I am working on.

The idea is this:

1.) Floor Unit houses a string/cable that attaches to a bar about 6-7 ft high off the ground
2.) Bar then moves load up and down for 1 repetition.
3.) The movement of the string generates a distance and time of the repetition
4.) Unit relays information to device that outputs (displays) velocity (preferably average and peak velocities)

I was thinking of using a String potentiometer but my thoughts tell me that how does the differentiate between repetitions.

Thanks!

What is this device... The cable must connect to something other than the bar?

 

[shrtrnd]

Accelerometer?

 

[bcouilla]

This is what I was thinking of. See attached

shrtrnd, that could work how accurate can they get? I want it to be consistent and repeatable as possible.

 

[bcouilla]

This is what I am trying to make. Not nearly as elaborate obviously....

http://www.tendosports.com/TENDO-Units.html

 

[Gryd3]

Accelerometer?

How accurate could you do velocity with an accelerometer?
It won't be able to sense motion as such... just changes in motion.
You would need to use a little calculus to determine the velocity of the object when given an acceleration and time, but the acceleration won't be constant... It can be done, just nervous about the required accuracy, and what you may end up with an accelerometer.
If it's going to be suspended from this bar without a reference point then this will be as good as it can get.
If you can use a reference somewhere... ie, using a ping sensor to measure distances from the floor, then you could more easily calculate velocity to a more accurate level.

 

[bcouilla]

The unit i would make would be on the floor. But i get what youre saying about ping sensors. how would you approach using those?

Thanks!

 

[Arouse1973]

Look at altimeter modules. They are used in lifts and can determine the distance between floors of a building. I saw one sense a few cm of movment of a desk. You might be able interface with one of these.
Adam

 

[Gryd3]

The unit i would make would be on the floor. But i get what youre saying about ping sensors. how would you approach using those?

Thanks!

A ping sensor would merely be attached or suspended from the bar and pointed at the floor (or ceiling) so that when the bar moved, the distance reported back changes.
Using the Altimeter mentioned above would help as well...
Simply get one distance, wait, get another distance and calculate speed.

 

[bcouilla]

Could a string potentiometer be used to where the revolutions measures time and distance measured by string displacement?

 

[Gryd3]

Could a string potentiometer be used to where the revolutions measures time and distance measured by string displacement?

I'm sure you could use that.
You could also put the 'string' on a rotary encoder.

You would need a little math to translate rotations to linear distance, but the steps afterward would be the same as if you used an altimeter or ping sensor (or any other distance sensor for that matter)

 

[Gryd3]

What is this being used for?

Will it be attached to the bar on a bench press?

 

[hevans1944]

I visited the TendoSports web site and looked at some the videos posted there. It seems to me that the primary thing their sensor measures is velocity, not displacement. Displacement is fairly easy to derive from velocity by integration. A suitable velocity sensor can be as simple as a homopolar generator, a conductive disk (usually copper) rotating in a magnetic field applied perpendicular to the plane of rotation, with brush pick-offs at the axle and outer edge. An optical disk essentially measures displacement, so differentiation (a noisy process) is necessary to determine velocity. Either approach is viable though.

I have used string potentiometers to measure displacement, and they perform pretty well... not as good as LVDT transducers, but it is hard to make an LVDT that responds to large displacements. I have never tried to differentiate the displacement signal to obtain velocity. This was way before microprocessors and inexpensive analog-to-digital converters, so this may be a "no brainer" today. I would try the string potentiometer approach first and see how difficult it is to derive velocity data using the Arduino. You can always substitute a small DC motor with a string pulley and re-wind mechanism for the string potentiometer. Or take the string potentiometer apart and add a small DC motor to the string potentiometer shaft to get both displacement and velocity information from the same sensor. Either way, it sounds like a fun project. The TendoSports solution is probably patented, but that shouldn't stop you from making one for your own personal use.

 

[bcouilla]

I visited the TendoSports web site and looked at some the videos posted there. It seems to me that the primary thing their sensor measures is velocity, not displacement. Displacement is fairly easy to derive from velocity by integration. A suitable velocity sensor can be as simple as a homopolar generator, a conductive disk (usually copper) rotating in a magnetic field applied perpendicular to the plane of rotation, with brush pick-offs at the axle and outer edge. An optical disk essentially measures displacement, so differentiation (a noisy process) is necessary to determine velocity. Either approach is viable though.

I have used string potentiometers to measure displacement, and they perform pretty well... not as good as LVDT transducers, but it is hard to make an LVDT that responds to large displacements. I have never tried to differentiate the displacement signal to obtain velocity. This was way before microprocessors and inexpensive analog-to-digital converters, so this may be a "no brainer" today. I would try the string potentiometer approach first and see how difficult it is to derive velocity data using the Arduino. You can always substitute a small DC motor with a string pulley and re-wind mechanism for the string potentiometer. Or take the string potentiometer apart and add a small DC motor to the string potentiometer shaft to get both displacement and velocity information from the same sensor. Either way, it sounds like a fun project. The TendoSports solution is probably patented, but that shouldn't stop you from making one for your own personal use.

As I study the string pot more and more. Would a rotary encoder, fishing line (or something similar), and an ID badge retractor be a viable alternative? The costs of string pots are pricey. Thanks!

 

[bcouilla]

What is this being used for?

Will it be attached to the bar on a bench press?

Yes it will be. Also for squat (which could be up to a max of 7-8 ft of string. The big concern for me is tension on the string (max im looking for would 3-6 ounces of tension as not to affect the loading pattern of the lifter).

Thanks!

 

[bcouilla]

By the way this is my first undertaking at a project relating to electronics (if that is not apparent in my babble). All this help is much appreciated. I am looking to build this for under 100 bucks if possible so a string pot is out of the question i think.

My goal is to go with:

Digital Output of Velocity
String length of 8 ft
Minimal String Tension 3-6 oz
Arduino coding and interface
Data to export into Excel for reporting/graphical

The coding I'm thinking will have to involve a premise that the "go" command will be after the string is hooked to the bar and if the string (attached to the bar) moves more than +- 3 inches than the rep has started and rep concludes after it returns to +- 0 inches.

Does this seem possible? Seems overwhelming right now haha. Yes, it will be personal use and i will be checking with a friend of a friend in the patent office to see whats held on this device.

Thanks everyone!

 

[hevans1944]

If you connect the string to a velocity transducer (try a small permanent-magnet DC motor for starters), you can tell when the rep begins (and ends) by looking at the velocity profile. With a DC motor you could use an analog comparator to trigger data acquisition when a certain (low) velocity was exceeded during the lift, and another comparator to stop data acquisition when the velocity signal, which will be the opposite polarity when the string is retracted, drops below a certain (low) velocity threshold when the weight is lowered.

The main problem you will run into is maintaining sufficient tension on the string to retract it when the weight is lowered because this occurs much more quickly than the time to lift and extend the string. One possible solution is a rotary "constant tension" spring that provides a constant string tension as the barbell is lifted and lowered. A better solution might be a torque motor, but these are a bit pricey. For your first crude prototype, consider mounting a small pulley (one inch diameter or so) to the shaft of a PM DC motor, wrap one turn of the string around it, and then connect the free end (not the barbell end) to a longish extension spring, such as were used to close screen doors in the early part of the 20th Century. Or connect the free end to a take-up drum attached to a torque-motor shaft. You might even be able to use a small PM DC motor as a torque motor although the stall current will be high. Some trial-and-error would be required to avoid "burning out" the ersatz "torque motor". I would use steel "leader line" for the string to avoid stretching and inaccurate velocity readings. IIRC this is what model airplane builders used to use to connect servos to airplane control surfaces. It may also be available at fishing supply stores.

Hop