My Own Trainer Project

That's a cool pic I just noticed it is a wind resistance machine.

Alex Simmons said:

I've decided to have my own indoor trainer built up. I'm not doing it myself but have a mate who does this sort of stuff for fun. He's done a prototype which I've looked at and it's pretty cool.

Normally I wouldn't bother but with my rehab it's likely when I do start training properly, I'll need to do quite a bit more indoors than I used to (which was bugger all). So if I'm going to have a trainer - I want one that I'll want to get on, rather than dread.

So here are the rough specs:

• free wheel
  
• 7 speed hub gearing (a Shimano of some sort - heavy but bullet proof)
  
• 28-35kg flywheel with high quality industrial bearings (have to remove some mass from the 40kg flywheel we have available as it seems to make it too easy). Need to decide on the weight (moment of inertia I suppose).
  
• foils attached to flywheel for air resistance (possibly adjustable for variable resistance - but only adjustable when not moving)
  
• double reduction gearing (chainring to cog - hub gearing - to chainring to flywheel cog)
  
• a stainless steel or alu frame that's fully adjustable in all directions (think of those specialised cycle fitter frames with square tubing) and solid as all hell but with a real bike feel
  
• SRM crank with their adjustable length cranks would be ideal (ergomo is out for various reasons - but mainly due to power measured by left leg - half of which I don't have anymore!) - the prototype has an ergomo
  
• Have considered putting a powertap in centre of the flywheel but concerned about longevity vs industrial grade bearings.
  
• able to be moved (infrequently) by picking up at one end and rolling along. It'll be heavy!
  
• covers for safety (esp. flywheel) and protection from sweat
  
• fine tune resistance using a felt padded brake on flywheel
  
• Did consider using an electromagnetic brake of some kind to provide the fine tune resistance. Car alternators work but the resistance is not smooth (on/off). Any thoughts on alternatives?
  
• Large flywheel/gearing will enable full on acceleration work as well as standard aerobic and anaerobic development workouts.
  

So what else could go on there? Alternative ideas?

A programmable ergo brake would be great but I have no idea if such a thing exists that meets my criteria:

Needs to be solid, reliable and readily available (and not that expensive). The SRM will cost significantly more than the rest of this item together!

Sort of like the one below - a different one built by another good mate but same concept. This one has a 33kg flywheel and double reduction gearing.

[img]fixedgearfever.compict1697b 133.jpgOpen ↗[/img]

Alex:
Consider the Cateye Cyclosimulator for an electromagnetic brake/resistance unit -- the Cateye has a very nice resistance curve and you could likely find an inexpensive old one to part out (cut-up).
Best,
Bill Black

Alex,

To widen this up a bit, what's wrong with any of the commercially-available units out there on the market, e.g. the Saris model with a PT in it?

Steve

Cool project! I'll post if I come up with more ideas. Those fan blades look deadly! :eek:

Alex Simmons said:

28-35kg flywheel with high quality industrial bearings (have to remove some mass from the 40kg flywheel we have available as it seems to make it too easy). Need to decide on the weight (moment of inertia I suppose).

The inertia of the flywheel is not just a function of the mass, but also the rotating speed. If you already have a 40kg flywheel available, it could be simpler to adjust the gearing (different sized cog) than to make a new flywheel or tweak the mass. Plus, if you wanted to change something later it would just be a matter of swapping the cog.

Alex Simmons said:

....Did consider using an electromagnetic brake of some kind to provide the fine tune resistance. Car alternators work but the resistance is not smooth (on/off). Any thoughts on alternatives?...

If that flywheel is metal you might consider eddy current braking. You can make it active and controllable like the CT or passive with fixed electromagnets like the e-motion rollers use. Either way the idea is to place magnets very close to a rotating, electrically conductive disk in this case your flywheel. The magnets induce small currents that are basically shorted out and converted to heat which acts as the braking force.

I've got a friend that sometimes lurks on these forums playing with both active and passive magnets on just such a design. The e-motion rollers just have a heavy duty permanent magnet on an arm that can be moved closer to or further from the conductive rotor and that seems to work well.

Anyway, there are a lot of ways to introduce resistance but many are simpler and smoother than a generator/alternator design.

Good luck,
Dave

Thanks guys - keep the comments, ideas coming! The ability to control the resistance at a finer level than cadence and gear changes is particularly what I'm interested in - think of the ability to do incremental ramps tests, like a MAP test.

The cateye idea sounds interesting and any more on the magnet eddy current braking idea would be cool to hear about too. Or where I could go to find out more.

My buddy tried it with the 40kg flywheel in tact and found it too easy once the flywheel was up and running, so I suppose reducing the weight seemed to help more than reducing the gear. But I'll ask. We did talk about changing the gearing.

The one in the photo - my other buddy did all the calcs on the rotational inertia and gearing so he could practice track TT starts as well as normal trainer type riding. He's a world masters TT & sprint champion and is a clever dude.

As to other trainers, well the only ones I know about with power and a large flywheel are the SRM, PT 300 and velodyne. The SRM trainer is probably the closest. Velodyne is 5 times the price. PT 300 looks interesting but I don't like the handlebar set up, the less than fine tune adjustment of power levels and it's a fixed gear. I'd like gearing to allow a greater variety of workouts.

Something about designing your own on this that's kind of fun.

Alex Simmons said:

Thanks guys - keep the comments, ideas coming! The ability to control the resistance at a finer level than cadence and gear changes is particularly what I'm interested in - think of the ability to do incremental ramps tests, like a MAP test.

As to other trainers, well the only ones I know about with power and a large flywheel are the SRM, PT 300 and velodyne. The SRM trainer is probably the closest. Velodyne is 5 times the price. PT 300 looks interesting but I don't like the handlebar set up, the less than fine tune adjustment of power levels and it's a fixed gear. I'd like gearing to allow a greater variety of workouts.

Something about designing your own on this that's kind of fun.

I think you're confusing the Velodyne and Velotron. I considered the former for awhile but there were too many issues and I got a Velotron instead.

300PT $1800
Velodyne $2400?
SRM trainer $7250 + $540 for adjustable cranks
Velotron basic $5700
Velotron Dynafit $8000
+ Racermate Varicranks $550 (I found an unused set for $250 on eBay)

If you have a coaching license from your national federation, you may be able to get a significant discount on Velotrons (it's either 10 or 20%). That would make the Dynafit equal in price to the SRM trainer which, in the words of an SRM dealer, is just a glorified exercise bike.

I found ergometer mode and the ability to create workouts and just load the file extremely valuable. There's no fiddling with shifters (SRM) or some dial (300PT) to adjust resistance. If you want to do tests, it's much easier to have the computer control resistance so all you have to do is pedal. I would think having to watch a clock and adjust resistance manually to be tedious and distracting.

It's simply the best cycling purchase I've ever made. Far better than my SRMs. We've had another unusually cold winter in Colorado and I've been riding the Velotron for 2:40 (4x30 at 0.85 IF) everyday yielding an increase in FTP of 25-30W so far after having stagnated for the past 2-3 years.

Here are some pictures from my incomplete blog:

Squint said:

I think you're confusing the Velodyne and Velotron. I considered the former for awhile but there were too many issues and I got a Velotron instead.

300PT $1800
Velodyne $2400?
SRM trainer $7250 + $540 for adjustable cranks
Velotron basic $5700
Velotron Dynafit $8000
+ Racermate Varicranks $550 (I found an unused set for $250 on eBay)

If you have a coaching license from your national federation, you may be able to get a significant discount on Velotrons (it's either 10 or 20%). That would make the Dynafit equal in price to the SRM trainer which, in the words of an SRM dealer, is just a glorified exercise bike.

I found ergometer mode and the ability to create workouts and just load the file extremely valuable. There's no fiddling with shifters (SRM) or some dial (300PT) to adjust resistance. If you want to do tests, it's much easier to have the computer control resistance so all you have to do is pedal. I would think having to watch a clock and adjust resistance manually to be tedious and distracting.

It's simply the best cycling purchase I've ever made. Far better than my SRMs. We've had another unusually cold winter in Colorado and I've been riding the Velotron for 2:40 (4x30 at 0.85 IF) everyday yielding an increase in FTP of 25-30W so far after having stagnated for the past 2-3 years.

I think you're right about me confusing the Velotron with Velodyne. The Velotron is the one that I had in mind.

Man those prices are still silly compared to what this thing will cost - even with a discount through my Cycling Australia coaching accreditation. That SRM trainer price is unbelievable! I got a quote from the local supplier of the Velotron and it was five figures :eek:

Super sweet trainer BTW but price is just too much. I'm looking at this costing $1,000 - $1,200 at most + an SRM, which I'm sure I can pick up at a reasonable cost (and my buddy will lend me his variable length SRM cranks in the interim). I may need something shorter than the 165s for a while (currently my left leg is on a 100mm crank). The SRMs go down to 150 or 155 mm I think. Considering the only thing missing is the programmable ergo brake, I reckon that's not bad and the other features provide great flexibility. It's also simple and bomb proof.

If I can find an example of the cateye ergo brake Bill mentioned, we may be able to see how that can be hooked up.

Thanks for the info 🙂

Hi Alex, the bottom line is whether you enjoy riding it. If you have a go on a couple of good spin bikes, ie the ones they conduct classes on, it may fuel with with some more ideas for your own project.

mikesbytes said:

Hi Alex, the bottom line is whether you enjoy riding it.

Bingo - hence my opening statement about building something I'll want to ride rather than dread.🙂

Alex Simmons said:

If I can find an example of the cateye ergo brake Bill mentioned, we may be able to see how that can be hooked up.

Thanks for the info 🙂

Alex:
I emailed you (off list) a PDF copy of the Cateye info that I tracked down. I have found that by shortening the gradient control cable (which advances the electromagnetic resistance unit further along the steel rotor in the Load generator), the 3% gradient setting comes very close to matching with my SRM Pro in the 300W - 400W range.

Good luck and have fun.

Best,
Bill Black

Squint said:

Thanks for posting the Velotron pics. Question:

Looks like a copper ring on the outer of the flywheel has slits around the ring dividing the copper ring into about a dozen sections. That's what goes through the ergo brake at the bottom, which I assume is a controllable eletromagnet.

Are the copper sections insulated from each other and the rest of the flywheel?

I've been reading up on eddy current braking and that may not be all that hard to do. Either with a controllable electromagnet or a strong magnet which can be moved closer to / further away from the flywheel.

Alex Simmons said:

...Are the copper sections insulated from each other and the rest of the flywheel?...l.

I can't say how they built the Velotron, but you don't need to insulate the individual rotor sections or segment them at all. The eddy currents are localized to the region where magnetic flux lines are cut by the rotating disk. You can use eddy current braking on a single conductive disk or an arrangement like the one shown, good conductors help(hence the copper or brass rotor plates) but they don't need to be electrically isolated from one another.

-Dave

daveryanwyoming said:

I can't say how they built the Velotron, but you don't need to insulate the individual rotor sections or segment them at all. The eddy currents are localized to the region where magnetic flux lines are cut by the rotating disk. You can use eddy current braking on a single conductive disk or an arrangement like the one shown, good conductors help(hence the copper or brass rotor plates) but they don't need to be electrically isolated from one another.

-Dave

Thanks Dave - I wondered about that as I read somewhere about putting slits in to make the effect stronger. Certainly the localisation of the eddy effect was clear from the stuff I've read so far.

Now I need to work out what sort of and strength of magnet would do the trick on a 35-40kg flywheel with a ~2' diameter. An electromagnet with a controllable current via transformer presumably would do the trick. Any clues on what it'd require? Any electrical engineers out there?

Alex Simmons said:

Are the copper sections insulated from each other and the rest of the flywheel?

I'm pretty sure they're not insulated. The bolts go into the flywheel and would probably conduct. Also, there is enough play around the bolts that adjacent copper fins could (and probably do) contact each other.

Squint said:

I'm pretty sure they're not insulated. The bolts go into the flywheel and would probably conduct. Also, there is enough play around the bolts that adjacent copper fins could (and probably do) contact each other.

OK - thanks. Then it was just easier to make them that way and bolt on rather than for some special e-m purpose.

Alex Simmons said:

OK - thanks. Then it was just easier to make them that way and bolt on rather than for some special e-m purpose.

could be an expansion issue as well I guess though I don't know how much a flywheel of *that* size is going to heat up spinning at say 40 kph.

Re your project, I'm certainly willing to offer time to help Alex. I'm an EE by degree but not by work experience so I've been out of the true hardware design end for too long now. However, I'm pretty good at reviewing and criticizing 😄 if you need someone to do a design review etc...

Are you actually hoping to construct a full-blown ergometer with power regulation (load x speed) or more of an extremely heavy-duty mag-trainer with widely adjustable resistance?

rmur17 said:

could be an expansion issue as well I guess though I don't know how much a flywheel of *that* size is going to heat up spinning at say 40 kph.

A friend measured his at 10F above ambient.

daveryanwyoming said:

I've got a friend that sometimes lurks on these forums playing with both active and passive magnets on just such a design.

That would be me.

I'm an EE, but I'm afraid I must have slept through most of my electromagnetics courses. I made a few attempts at store-bought and home-wound electromagnets and didn't have much luck. So, although it's not rocket science, unless you've got a lot of time for trial and error, or a working model you could reverse engineer, or know somebody with a bit of experience in the matter, it might prove to be tricky to get an electromagnet powerful enough to provide the resistance you'll require. With the electromagnets I wound, it was looking like the voltage and current required was going to be in the potentially lethal range, so I changed my plans.
My current efforts involve using some strong neodymium magnets mounted on a swing-arm that I can swing closer of further from the flywheel. My ultimate goal is to add a 'constant power' feature to my 300PT. The magnetic braking wouldn't necessarily provide all of the resistance, but it would increase or decrease as needed to compensate for a decrease or increase in cadence.
Here's a link to a commercial spin bike that only uses the permanent-magnet/proximity idea. Click on the photo and it will pop-up a slide show, one of the slides is a close up of the magnetic brake assembly.
valorathleticsinc.comproduct.phpOpen ↗
Those magnets would need to be pretty strong, so I would think the lateral force would cause premature failure of the bearings. A better design would be to have the magnets on both sides of the flywheel in order to balance out that effect.
I'll post more once I have some successful results,
-- Bryan