Non-contact revolution counter

dudara

I've been searching on the net, but I'm not finding anything that quite suits my needs.

I have two small dwarf hamsters who spend ages running in their wheel. I want to count the amount of revolutions (preferably using a non-contact method as these are quite small little dudes) so that I can calculate the average distance travelled.

I initially thought that a bicycle computer/odometer thing might do, but they mostly seem to cater for larger circumference wheels (the hamster wheel is 54 cm in circumference). I can't find any that has a simple revolution counter setting.

Likewise I looked at laser tachometers, but they mainly count RPM rather than counting the total no of revs.

I think that a magnetic counter based on a Hall sensor would probably be best, but I'm open to any suggestions that anyone may have. Thanks in advance for any responses.

bushy...

A few i can think of :

A mechanical or electronic pulse counter is what you need really , lots of types around

http://radionics.rs-online.com/web/search/searchBrowseAction.html?method=searchProducts&searchTerm=pulse+counter&x=0&y=0

might look like some kind of experiment you're carrying out on the little guys though


Or if you leave on your pc all the time anyway , you could just hook it to that

or hook it to one of these http://www.picotux.com/ and put them on the internets

.50 (MOA)

The bicycle spedometer will work even though its made for larger diameter wheels.

when you are setting it, set the wheel size to a any wheel diameter, then work out the difference between this size and the actual wheel, then when you check how far the hamster has gone divide by this value to get the actual distance.

you say the wheel is 54cm in circumfrence (make sure this is measured to the inside face where the hamster actually runs). instead set the unit to a value of say 600mm diameter this will mean the unit registers 1.885m per revolution. simply divide by the difference in this case it is:

1.885/.54 = 3.4888888

so if the odometer reads 260m then the hamster has gone 74.522m in reality.

one bit of advice, the magnet used with the sensor will work best if placed futher out on the wheel, this will unbalance it so you'll need to put something the same weight on the opposite side of the wheel to balance it again (bluetack maybe- likely they'd eat it though). with this the wheel will be balanced but still be harder to turn because of the greater mass, nothing you can do to stop this though

dudara

.50 (MOA) wrote: »

The bicycle spedometer will work even though its made for larger diameter wheels.

I was thinking this, I only really want to know the revs and not the distance.

one bit of advice, the magnet used with the sensor will work best if placed futher out on the wheel, this will unbalance it so you'll need to put something the same weight on the opposite side of the wheel to balance it again (bluetack maybe- likely they'd eat it though). with this the wheel will be balanced but still be harder to turn because of the greater mass

Dwarf hamsters are quite small, I'd guess maybe 30 - 50g depending on size/age, so obviously I don't want to attach magnets that are too heavy, especially as the moment of inertia will increase the futher the magnets are placed from the centre of rotation. (double this effect if I use a counterbalancing weight).

As for the blu-tack - it'll be gone into their tummies in about 30s - they're not the brightest.

bushy... Thanks for the Radionics link - must check our Farnell as well. Gone are the days when I used to 'acquire' all my gear from the undergrad lab.

coolwings

One simple and dirty solution (there are many more) :

Attach a threaded bar, or long threaded machine bolt to the axle of the hamster wheel, so that it projects out in a horizontal direction.
eg an M4 or M6 bolt of as long dimensions as you can get.

Heatshrink tubing, or a drinking straw might be a simple way to attach it.

Then take a simple thin metal tag shaped somewhat like a drink can ring pull.
Lay it on the threads and let the weighted end hang down.
For every revolution the tag will move out one thread along the bolt.

Use a ruler, using the pitch of the thread chosen, "x" threads per inch/cm.

BTW. Consider the direction of revolution before picking the threaded bar/bolt. If you have an RH instead of an LH, either you have to ask the hamsters to run backwards, or your hanging tag will move inwards.

.50 (MOA)

i've got it, why don't you attach 2 light strings to the axle of the wheel and count the twists put in them.

attach the strings to a weight with no twists between them.
set the weight on the same surface as the cage.
attach the strings to the axle, with a bit of glue.
each turn of the wheel will put a twist between the strings.

use enough string that it will take a lot of turns for them to tighten up and make the wheel difficult to move.

when you want to find the no. of rotations take off the string carefully hold up the weight by the strings and count the rotations needed to remove all the twists, you might need to keep slowing it down as you count and use a weight that you can easily tell that it has made a full turn, something with some obvious feature to count when it rotates past each time as you count.

this isn't really non contact but if the string is light enough and long enough for the number of rotations in a day or whatever period you want to count for the force needed to move the wheel, shouldn't increase by much.

a simple solution i feel, with no outlay required, win- win

spideog7

Would a light beam work. A simple infrared led setup sending a beam across the top of the wheel. Assuming the wheel is made up of a series of non-transparent spokes you should be able to count the resulting pulses using a fairly simply circuit. The number of pulses divided by the number of gaps on your wheel will indicate how many revolutions have occurred.

Some crude drawings, ignore the fullstops they are only for formatting.
Side view of the wheel:


Top off cage

X:::::.(..).:::::::X <

---

Light beam including sensors (x)
.........(..)
.........(..)

---

Axle
.........(..)
.........(0) <

---

Hamster

Floor of cage


That should work, I think, the number of pulses you measure should be equal to the number of gaps on your wheel (between the spokes) on each revolution, so a wheel with four spokes (like a pie divided in quarters) will have 4 pulses for each revolution.

Just thinking of the counter weights for the earlier situation, I think you would actually need four (3 + the magnet) to balance the wheel properly, other wise it would still need to rotate a possible quarter turn after the hamster had stopped running and he would have to lift the weight a quarter turn too.

.50 (MOA)

spideog7 wrote: »

Just thinking of the counter weights for the earlier situation, I think you would actually need four (3 + the magnet) to balance the wheel properly, other wise it would still need to rotate a possible quarter turn after the hamster had stopped running and he would have to lift the weight a quarter turn too.

The wheel would still be balanced with just one balancing weight and the magnet.

If the hamster has to lift the weight a quarter turn then there is also the magnet which is falling a quarter turn balancing it out, as i said and the OP picked up on, the moment of inertia would be increased, making it harder fo the little guy to turn the wheel, this might mean that he's getting more exercise in less time so the results would be affected because he is running less- kindof like running up hill.

Capt'n Midnight

Bike meter the easiest , simply divide by 10.

Bike wheels of 700C would scale down to 7cm ( 700mm/10) if you move the decimal point
next trick is to set it to mph instead of kmph so you now get another factor of 1.6 so would read KMPH when the wheel is ( 7cm * 1.6 = 11.2 cm ) in diameter

Check the largest wheel size on the bike , should go up to ~860mm which would be 13.76 cm if you use the mph trick.


put the magnetic near the axel to reduce the inertial and put a counterweight opposite it so the wheel balances and doesn't rock back and forth when the magnet is in the lowest position.

dudara

I liked the simple approaches especially the one with the thread, but unfortunately the hamsters run in the both directions, so any system I come up with has to be capable of both.

I thought about a small Hall sensor and magnet and some form of data logger for the laptop, but that's getting a bit fancy.

I think that after considering all these options, I'll go with Capt'n Midnight and find a bicycle computer with a small light magnet. Time to eBay I think.

Thanks everyone for all your ideas. Much appreciated.

Capt'n Midnight

dudara wrote:

Time to eBay I think.

check out the german discount stores they have them for around a fiver every so often

Harpz

bike sensor in aldi. around 5€ from today

If you want revolutions to be measured just input (dunno if it requires radius/diameter/circumference) whatever parameter to give you 1 metre per revolution. Then your metre will read the number of revolutions (as metres)

dudara

Thanks for the tips Harpz and Cap'n Midnight. I'll take a walk home past Aldi this evening and get myself sorted.

The hamsters cover some amount of ground every time they're up and I'm really eager to quantify it somehow.

Turbulent Bill

dudara wrote: »

I liked the simple approaches especially the one with the thread, but unfortunately the hamsters run in the both directions, so any system I come up with has to be capable of both.

The bike computer idea should work nicely for this as it only measures the number of magnetic switch actuations per unit time to work out the speed (and the distance from the given wheel diameter), not the direction of travel (I think). This means that both clockwise and anticlockwise travel is measured. The only issue might be where the magnet is passing to and fro in front of the sensor if the wheel is rocking, making it look like the hamsters are pegging it around. This should be a small enough problem though unless your Blu-Tack eating furballs are exceedingly smart and devious...

Nice to see so many different ideas being proposed, plenty of innovative people on this forum!

.50 (MOA)

Turbulent Bill wrote: »

The only issue might be where the magnet is passing to and fro in front of the sensor if the wheel is rocking, making it look like the hamsters are pegging it

this is why i said to put the magnet as far as possible from the axle. for the same angular displacement the linear displacement is greater and the magnet is less likely to switch the sensor more than once for any small movements.

finding a perfect balance between far and near would be advantageous

dudara

Turbulent Bill wrote: »

This should be a small enough problem though unless your Blu-Tack eating furballs are exceedingly smart and devious...

Let's just say that those lads definitely won't be taking any steps along the evolutionary ladder anytime soon.

Turbulent Bill wrote: »

Nice to see so many different ideas being proposed, plenty of innovative people on this forum!

Most definitely - great varied responses - lots of food for thought.

spideog7

Using the magnet is definitely easier alright, any magnet should work so you could try to get one as small as you can and glue it on instead of the big bulky ones with the screw that normally come with the bike speedo.
Best of luck with it, let us know how it goes.