Dumbell exercise counter

But sensors of accelerometer are always expensive.Is there any low cost solution?

Have you tried using an off-the-shelf pedometer?

Richard

Most dumbbells can be trained to carry out simple tasks such as recording start and end times of exercise. Maybe this is a psychological problem more than an engineering problem.

try here

martin

I have tried using 4 photocell as a matrix to sense the direction of motion. But it failed when the image is plain and the error is large when the background light intensity is dim.

That's the usual scenario. Either it is easy and costs a lot or you need to build it at rock-bottom cost. In that situation you have to design your own accelerometer so that it costs just pennies. It can be done but requires lots of engineering.

Hint: Think how older electronic scales used to work.

Regards, Joerg

Photocell.

Pressure.

Did you try the pedometer suggestion Richard made up-thread? With careful adjustment I'd have thought that would give an accurate reading?

But your spec is rather vague anyway. Are you trying to count the number of times a dumbbell is raised? Counting interval started and ended how? From where to where? During what types of exercise? (Press, jerk, curl, single-arm bench press, etc?) Exact count? Only dumbbells, not barbells? What distance qualifies as a 'lift'? Etc...

Pressure sensors does not have enough fast response time.

Walmart carries them for $3.88/ea. Just strap it to your arm and it should do what you want.

There are pressure sensors with a BW of a several Hz or more. I guess even Arnold couldn't sustain a training regimen this fast.

Regards, Joerg

Yes. Each time you lift the dumbbell (was your mother a weightlifter?), add an integer, i.e., "one, two, three, four, five... " and so on.

When the numbers get too high for your musclebound brain to remember, write them down, e.g. "...seven, eight, nine, TEN!" Make a mark on a piece of paper or other writing medium. Then start the verbal/oral counting sequence again, "one, two, three, four, etc..." or maybe even "eleven, twelve, etc..." - after the twenties, it gets much easier - (Twenty) one, two,... (Thirty) one, ... and so on.

Or, a piece of string, a plumb bob, and a microswitch?

Or hire someone? A Helper? A Friend? A Spotter?

Good Luck! Rich

Betcha thisn's a thread-stopper! ;-P

Cheers! Rich

The response curve is not linear. For so small pressure change the response time to get stable is rather slow. In many datasheets I get the response time is less than 1ms, but there is a comment that the testing condition is from 10% to 90% of the sensor's pressure range.

Actually I have evaluated one pressure sensor and the pressure sensor has response time of 0.1ms (10% to 90%) in the datasheet.

Ok - if you can't use optical, or accelerometer, or pressure, use a mercury switch.