I found several fans and mag units that use belts to drive them by one of the rear rollers. All of the add-on units would require more moving parts, belts that might not fit just right and moving parts that would eventually wear out with one exception. I came real close to purchasing a magnetic resistance unit like this one, but couldn't be sure that it would fit my trainer.
The unit uses magnets to provide resistance even though the aluminum isn't magnetic! The magnets don't come in contact with the roller, so how do they provide resistance?
Turns out that the unit is a simple Eddy Current Brake. Click here to read more about Eddy Current Brakes.
The magnets themselves have a magnetic field which when placed very close (a few mm) to the moving aluminum roller (a conductor) create an eddy current in the roller.
A little help from Lenz Law explains how the moving alu roller's eddy currents interact with the magnetic field in such a way to oppose the movement of roller and provide a smooth "braking" effect. You can see in the video how moving magnets can induce an emf in the aluminum by induction in this short video. Now, rather than move the magnet, we're moving the conductor (aluminum roller) and it will produce the same effect. The braking effect increases as the speed of the roller increases.
What this means is that you can use gearing to adjust the speed of the roller and increase resistance, or just simply increase your cadence.
The same principle is used in mag-lev trains and roller coasters for braking.
You can clearly see the neodymium magnets here used as an eddy current brake to provide resistance in this flywheel exercise bike. You'll find similar set-ups in consumer elliptical and exercise bikes.
So... I took some rare earth neodymium magnets and hot-glued them to a piece of scrap wood cut 1-2mm longer than the inside dimension of the roller frame. Wedge it between the rails of the roller frame and adjust to desired distance / resistance. Here's some shots of the simple setup.
Use a scrap piece of 2x4 which I hot glued the nedoymium magnets to. I actually used a paddle bit to counter sink them a little bit. Didn't want to make them flush in case I had to move them for some reason.
I actually alternately reversed the polarity of the magnets to see if this might add a bit more resistance, but I haven't spent the time to switch them back to see the effect. I do know that it works this way.
I cut the piece just a mm or two long to make a snug but not too tight fit. This way I can remove it easily or adjust the distance between magnets and rollers. The beauty of this is that you can cut the wood to length to fit any trainer. Here you can see I've just begun to wedge it into place.
Adjust to desired distance. This looks like a mm or two to me.
Set in place ready to roll!
A quick spin of the roller with and without the magnets demonstrates the effect. Spin the roller faster and you see the aforementioned increase in braking effect.
No resistance
With resistance unit (eddy current brake)
With resistance unit (eddy current brake)
The important thing is that it works... very well in my opinion. You can vary the resistance easily by moving to a larger or smaller ring. Move the magnets closer to the roller for even more resistance or remove them all together for an easy spin.
The only drawback which I have yet to really measure or deal with is heat production in the aluminum roller due to it's electrical resistance as a conductor. You are basically transforming kinetic energy into thermal energy.
Some dyno's used to test automobiles and motorcycles need coolers to remove the heat produced during testing.
How hot does it get? I don't know. I'm planning a few high-resistance workouts in the next week or two to see how it performs. I'll report back. Update: just finished 45 min Spinervals 1.0 workout and it got warm, but not even remotely "hot." We'll see how it goes for an 1.5+ hr. ride later.
http://www.launc.tased.edu.au/online/sciences/physics/Lenz%27s.html
http://en.wikipedia.org/wiki/Eddy_current_brake
http://www.physics.ubc.ca/~outreach/phys420/p420_96/bruce/copper.html
http://alex-cycle.blogspot.com/2008_02_01_archive.html
http://valorathleticsinc.com/store/images/D/AC-2_Magnetic_Resistance.jpg
http://en.wikipedia.org/wiki/Eddy_current
http://www.overstock.com/Sports-Toys/Valor-Fitness-AC-2-Spin-Bike/3228573/product.html
Great Post, ill definatley try this.
ReplyDeleteCan you tell me though, does each individual magnet have a different polarity on either side? ie is it '+' on one side and '-' on the other? Or do you have to get a '-' magnet/disc and a '+' magnet/disc? Sorry my magnet prowess is limited.
I suppose i just want to make sure i get the correct magnets.
Thanks
ML
Brilliant! You just saved me $80. Thank you.
ReplyDeleteSorry for late replies... not receiving notice that there were comments...
ReplyDeleteThe magnets I picked up were 3/4 in diameter x 1/8 in thick, Rare Earth Neodymium Disc Magnets for less than $1 a piece and have the familiar +/- polarity you are talking about...
Used the rollers this morning... good resistance, not super-strong... but enough for what I'm looking for on rollers...
Great posts,
ReplyDeleteIt looks like the magnets will decrease the spin down(coasting) time. Any plans to add a flywheel?
Just stumbled across this and tried it with my rollers. I'm amazed at the amount of resistance this added. I used 7 magnets spaced fairly evenly across a piece of wood similar to what is shown above. I put small flat head screws through the middle of washers (about the same size as the magnets 3/4") to give the magnet something to "stick" to as well as hot glue around the edge. Because the magnets are not drawn to the drum, the wood doesn't necessarily have to be wedged between the frame. If you maintain the proper height you can create a slide to allow you or someone else to move the magnets closer or further away depending on what you are trying to accomplish. I did notice that the drum got warmer but I certainly wouldn't say it got hot.
ReplyDeleteHi - have you experimented any more with alternating the polarity vs having them all the same?
ReplyDeleteI intend buying magnets with a countersunk hole through the centre, which is convenient for mounting. I need to decide whether to buy them all with the countersunk in the same sense, or buy half with the countersink on the nth face and half with the countersink on the sth face. I'm taking a punt that 4 'N45' 20mmx5mm magnets should add enough resistance. I may mount the magnets off a single arm from one side of the trainer, so the magnets will be quite close to each other rather than distributed across the width of the rollers. Thx.
Hi Graeme,
DeleteThere is no noticeable difference with reversing polarity of the magnets - at least that I can detect without any real scientific measurements / tools.
You might try prototyping it simply to test out a few ideas first, then invest a bit more time and resources in a more refined setup. Good luck with your design!
Matt
Hi Matt, thanks so much for this great post! I've got my magnets and getting ready to mount them, but have a question for you. In your picture it appears that you have two magnets mated up and then glued to your wood. Is this correct or did you just glue one (1) magnet at a time to the wood? Does it matter which side of the magnet faces the AL roller?
ReplyDeleteHey Jeff, thanks...
DeleteI used a spade bit to countersink one of the magnets and mated the magnets together. Then I just put a bunch of hot glue in the hole and squished both magnets down in the hole. The top magnet appears sticks out while you can barely see the other magnet in the photos. I imagine you could fasten a strip of metal or something to the wood, and then "stick" the magnets onto the metal strip attached to the wood. That's the great thing about all of this... just experiment and see what works for you! I DID alternate the polarity of the magnets... Good luck!
Matt
OK, great. Thanks so much for the reply. Just just to make sure I'm understanding you correctly, one doesn't *need* to have two magnets stuck together per position to make the Eddy Current active, right? If I have one (1) magnet per position on the plank of wood that will have the same effect no? I did this already, and I think that it provides resistance, but want to make sure it isn't just psychological!
DeleteWell... more/larger magnets might mean more field interactions between the magnets and the conductor (roller) and subsequently more resistance... so I would personally try a configuration that would allow you to easily add more magnets or arrange them differently to see the effect...
DeleteGreat. Thanks so much for your time and patient interaction Matt.
ReplyDeleteNo worries... this is the beauty of the 'net!
DeleteHello, I have a spinbike that uses a resistance pad and I'd like to modify it so it uses magnetic resistance with permanent magnets. It has a 31lb flywheel but it is not aluminum (I think it's a chrome plated cast iron flywheel). With this type of metal can a eddy currents still be induced to create a magnetic resistance mechanism?
ReplyDeleteI have tried mounting one 1"x1"x0.5" neodymium magnet (magnetized through the 0.5") on the frame facing the center of the flywheel from the outer edge on the perimeter (just a few mm away from the flywheel, poles in the plane of the wheel) but I haven't noticed any change in resistance. I will try two of these magnets tomorrow, but instead oriented such that they are facing each other on opposite sides of the perimeter of the flywheel (poles perpendicular to the plane of the wheel).
Assuming eddy currents can be induced with this kind of metal, I have no idea how strong these magnets need to be in order to create a noticeable level of resistance in the 31lb flywheel.
Well, after some jerryrigging with scrap wood, I was able to get the two magnets on opposing ends of the flywheel stably (just a few mm away, tbh these magnets are so strong that it this was a bit nerve racking). While I cannot *feel* any significant difference in the resistance when pedaling, the magnets are definitely having an effect: without the magnets, pushing down hard on the pedal with my arm produces ~22 rotations, but with the magnets there are only ~9. It's a rather imprecise way of testing, but I've repeated this test many times without much variation in the number of rotations.
DeleteIf the change in resistance was actually noticeable, I would've just purchased more magnets, but the effect here was so negligible that I'm having doubts about this solution. It's looking like I'll be purchasing a different bike.
Thanks for the update! Sorry for my late reply...
DeleteI wonder if the increased inertia (due to increased mass of the flywheel) is doing some of the work for you while pedaling (at constant speed) at higher rpm.
Spin bikes utilize the flywheel to simulate realistic feedback (the pedals pushing back on your feet) during acceleration of the flywheel. This feedback is supposed to simulate what you might feel if you were climbing a hill.
The harder you push on the pedals, the greater the pedals push back on your feet. The added resistance of the heavy flywheel to a change in its motion (inertia) causes the change to happen over a longer period of time than if the wheel were made of a much lighter material.
The initial effort to get the heavy flywheel spinning (hence your 22 vs 9 rotations) would be noticeable at low rpm due to the fact that the flywheel is not moving at all when the force is applied.
Once the heavy flywheel gets spinning at high rpm, momentum increases due to higher speed of the wheel and keeps the wheel moving in spite of small variations in cadence and effort (the purpose of the heavy flywheel).
This same principal reminds of the guy who pulls the passenger jet on the ground, with a huge effort required at the beginning, but with less effort once the jet is rolling. Smaller bouts of effort keep it moving.
Using an equally sized wheel made of a much lighter material would more easily be influenced by the eddy currents and would require more effort on your behalf to keep going with natural variations in cadence. These slight changes in cadence would give more feedback with a much lighter wheel.
This doesn’t mean the overall resistance on the wheel provided by the mechanical brake and the eddy current brake provided by the magnets isn’t working, just that maybe the perceived effort using a heavy flywheel with very few magnets doesn’t give the feedback you might expect at higher rpm (even though eddy current resistance increases as speed increases).
So… maybe use…
1. More magnets for a heavy flywheel
2. Less magnets with a lighter wheel
3. Add another mechanical brake
Thanks for your comments!
Thanks for the response! I ended up just selling the bike and getting a new one (the "Sunny SF-B904". It's far from high quality but it seems like it'll do the trick). The bike has a substantially thinner flywheel (maybe a third of the size of the other one) and maybe over a 6th of the wheel's outer radius has a strip of magnets overtop of it. Assuming I'd have to buy 3x as many magnets to get the same effect, well, I would have ended up spending quite a bit of money. In fact, in the long run I may even end up adding more magnets to this one.
DeleteAfter about an hour of usage, the wheel will get quite warm though!
Hi
ReplyDeleteThinking of trying this with electro-magnets and a switch / dial resistor to adjust them on the fly from the handlebars- any thoughts on this ?
Why not? If you have the time and determination... not sure how large of an electromagnet you'll need? The extra hardware might be a little cumbersome... but there's nothing like a good 'ole DIY project! Let us know how you make out!
Deletehttp://ichi2.net/28-years-later/electromagnet.jpg
DeleteSo I do the same many years before, but as you known the energy never lost just only change and in the mangnet and aluminium becomes in hot. in fact, too hot too fast...
ReplyDeletemy solution was to place one magneto in each roller this disipate the temperature