If you wanted to make it "slicker" - you could machine a groove into the flywheel for the CT. Then just connect a shorter belt to the rear of the two rollers, and bolt the CT to the aft portion of the roller frame.

I have a resistance unit on my TACX rollers that work sort of like that.

---

___________________________________
MS: Exercise Science
Your speed matters a lot, sometimes you need to be very fast, where sometimes you need to breakdown your speed.

Do you think it would be difficult to get the right size belt? In my limited research on rollers, it sounds like most don't offer sufficient resistance on their own. However, I wonder if the inherent resistance of any set of rollers may be so high that adding a resistance unit would make the rollers much too hard. As a way to overcome this, you could put a small cog on the resistance unit and a large cog on the rollers. Machining the grooves directly into the flywheel would limit the gearing options while still in the early stages of development. It would be nice to be able to switch the resistance unit between a set of rollers and the traditional stand.

I will say that I HATE the resistance unit on my rollers right now. I'm not sure if it's that unit in particular, or just any kind of load being put on rollers. It makes it very difficult to ride smoothly when you're fighting a lot of resistance.

But you're right, I think a chain system might make it a bit easier to switch.

---

___________________________________
MS: Exercise Science
Your speed matters a lot, sometimes you need to be very fast, where sometimes you need to breakdown your speed.

The CT works in large part because it is calibrated to a specific press on force. Without bracing the frame and the roller drum you get variable force and bad data. I think the aims are incompatible. A better solution is to use an on-bike power meter with rollers but there is no erg mode in this solution.

I hadn't thought about the press on force for CT. Does the Tacx Bushido/Genius use the same approach? Using the CT, I wonder if something that offers a variable resistance would help out. Maybe something link a rubber belt that go around the CT drum and the roller drum. I will see if I can find such a band (such as a replacement roller band); I don't think an automotive fan belt would work as it would likely be too stiff. Of course, if I f' this up my wife will banish me to 10 hours of spin class per week until spring finally arrives.

I was thinking about this a bit more and it seems that if the CT unit is securely attached to the roller frame then the press-on force would be simulated by the bike on the roller drums. That is, the CT press-on force just reads the built-in resistance of the tire force against the roller/flywheel and the amount of resistance felt throughout the roll-down calibration. If I have a chain-driven CT attached to the drum of the rollers, then when I do the CT calibration, it should interpret the virtual press on force as the combined resistance of the wheel against the drum and the drum resistance. Does this make sense?

I probably can't tell you with certainty, but there are two snags I can think of here:

1) I recall reading on the RacerMate forums about how the normal press on force of gravity is around 0.70. With a roller setup you are reliant on gravity. People report slipping issues with moderate gradients with less than 2.5 press on force. On rollers you won't get the tire slipping but I think unless you work with fairly flat terrain the wattage readings will be skewed.

2) The flywheel, such as it is, maintains a known amount of rotational inertia. By today's standards it is a very small flywheel but if you use the roller drum instead you will have a different rotational inertia. I don't know how different, but definitely different. This will certainly affect accuracy. I've seen some projects in passing that involve using a larger flywheel to create a more realistic road feel on the CT and they succeed in this regard but they distort the power values.

It is an interesting idea. I wish I had more than warning signs to share. You might try asking on the racermate forums as well. They have an engineer (jhgrud, I think) who can really tell if this is remotely feasible.

I will go do some searching on the RM forums. Haven't done anything there since I finally worked out my last problem with the CompuTrainer--just bad memories of trying to sort out the endless headaches that result from random CT problems.

As for the tire slippage issues, I wonder if this will be a problem as the CT + roller resistance will be higher than the CT-only resistance. I can still envision some slippage at the extremes though. For example, let's say the CT by itself offers ~75w of resistance (seems reasonable as when the erg video shows zero resistance, my crank based PM will often read 75-100+ watts if I am still pedaling). For the CT to bring it up to 200w (steady pace on a flat road), the unit must add ~125w. When I want to simulate hill climbing (say 300w at ~15 mph), then the CT unit would have to add 225w; this could lead to slippage if the press on force is too low. With a CT+roller unit, I would expect the minimum resistance would be at least 150w if I am going as a speed of ~15mph (obviously roller resistance is higher when going faster as rollers offer dynamic resistance). Let's assume that at 20 mph, the resistance is 75 from CT and 100 from the rollers. So to simulate a 200w & 20 mph effort, the CT unit would only have add 25w of resistance. For a simulated 300w hill climb at 15 mph, the rollers would offer ~75w of resistance and the CT would add 225w (with ~75 being already built in). The tire would only slip if the tire-roller interface is too low as I can't increase the press-on force (unless I keep adding the winter pounds). I think I will just have to rig this up and see what happens. I am now a bit concerned that the inherent resistance of the rollers would be too high when combined with the CT.

Trainer Road?

Yea yea you have to do your own resistance by gearing, pay them 10 a month, and maybe buy a sufferfest video or 2. But it is quite a bit less than screwing up a CT.

ah hah...
from http://www.dcrainmaker.com/...therly-incoming.html
(who else?)
" A new Elite Trainer (well, technically rollers) unit came in, the Arion Digital Roller. (http://www.elite-it.com/arion-digital/)
<...>
We decided to go with the rollers first. These are unique though in that they’re actually ANT+ enabled. The unit actually transmits your power and speed, and allows digital resistance control. Sorta like the mother of all rollers. I’ll look at their other trainer units as we approach next fall."

I will look forward to the review. I hope that it is or can be compatible with software like the erg video series. I was planning to get the inside ride rollers but as this is about the same price, I will have to think about this a bit more.

I grafted my CompuTrainer onto my TruTrainer rollers a couple of winters ago. The TruTrainer rollers have a built in flywheel that spins within the rear most roller (at about four times the speed of the roller IIRC). The rollers provide a very lifelike feeling with moderately high rolling resistance tires but they top out at about 375 watts (at around 60 kph).

The TruTrainer uses a belt to drive the flywheel so I had a pulley made that would provide the same wheel to roller (CompuTrainer) ratio. I removed the CompuTrainer's flywheel and replaced it with the pulley. I then made a mounting bracket for the CompuTrainer that located the pulley in between the two rear rollers of the TruTrainer. The setup worked well, from a mechanical engineering stand point.

The issues I had were with accuracy of the resistance curve. The lowest calibration number I could get out of the CompuTrainer was 8.5 which is way outside of the CompuTrainer's range of 2.0 to 5.0. I could easily do Erg workouts, however, I had to adjust the *.erg file's wattage down by about 150 watts to get the system to match up to my SRM. When it came to using the virtual training portion of the CompuTrainer software I couldn't get it to work at all in a manner that was safe to ride on the rollers.

I have been spending a great deal of time on a CycleOps PowerBeam Pro this winter and have been thinking about grafting it onto my rollers quite a bit. One thing I didn't take into consideration a couple of years ago was speed. I find my speed drops on some of the longer VirtualTraining climbs to less than 15 kph. I have spent more than 1,000 hours on rollers and it is very difficult to ride them at such a slow speed.

---

I have a personality disorder, I don't drink coffee...

That $10/ month is the best money I've ever spent on bike speed.

---

The point is, ladies and gentleman, that speed, for lack of a better word, is good. Speed is right, Speed works. Speed clarifies, cuts through, and captures the essence of the evolutionary spirit.

I would love to see a picture of this set up. Do you think you could get a lower calibration number if you changed the gearing (i.e. pully size) of the CT? Also, is the 150w difference consistent throughout the range of efforts (i.e. is the difference the same if you are doing a 200w effort vs a 300w effort)?

At the risk of highjacking my own thread, I need to look at TR more to decide if it makes indoor training more interesting. It looked to me that it just gave you various workouts and monitored cadence/HR/power (and controlled resistance if hooked up to the CT). If it did this and combined it with video footage (like the Erg Videos do), then it would be more interesting; I have 5-6 erg videos and even they get a little boring after a while. My interest in the CT + rollers is to have the benefits of Erg video/CT along with the "real road" feel of a good set of rollers.

TR - Today, does tie into a number of videos. It supports Sufferfest, Spinnervals, Epic (I think that is the name), and a number of others. At first it was just the target workout, and you can still do that. Now they setup the workouts to target the power based on the videos instructions. When the Sufferfest gun goes off the power is going to jump.

The only "downside" is that when the Video tries to lie to you and tell you one more interval you can see on the upcoming chart that there are really 2 more.

With your CT it will control the resistance as well. Without the CT, with rollers for example, it will just give you a target wattage and you have to use the gearing / cadence to get there.

There is no setup fee, and it costs 10 bucks... and if you hate it they will refund a month. Only extra cost would be buying a new video.

Sorry if I am being a bit slow on grasping this but can I combine TR + sufferfest video + comptrainer to have the resistance dictated by TR? That's what it sounds like but I want to be clear. On a related note, does the Sufferfest video control the CT (like Erg Videos) or are they like Spinervals (where you change the resistance manually through gearing/cadence)?

Thanks again for the info.

The video itself doesn't control anything. But trainer road set up a workout with timing to match the video.

I don't have a CT, but I have been with TR since beta and TR does control the resistance on the CT. It should be able to do that with sufferfest, or spinervals.

I just wants to say that I love the trajectory of the discussion on this thread.

What I would like to have the CT guys do is actually deliver a product along these lines to the market. Just attach the CT load resistance unit to a set of rollers through a belt or chain and then have them adjust their firmware to account for the "virtual press on force" that the combo of the tire, rollers and belt/chain deliver to the load generation unit. This would be a winner.

Also, I have noted that the CT courses totally suck unless they are less than 6% uphill grade or more 2-3% downhill grade. Once you get outside that, the CT does not approximate outdoor riding due to low wheel speeds. I made a whole wack of courses that keeps the gradients in this range and they would be perfect for a "roller based CT" where you don't want the wheel speeds to get crazy high (for the sake or roller safety) or crazy slow (staying upright and applying force the pedals).

Having said the above, I think the market for this is just a bunch of hard core guys like us. I can't see racermate investing in this tangential product line, even though roller diehards like us would love it.

Dev

I don't think the issue is actually a market for the product (I've learned there's a market for anything as long as triathletes and cyclists are involved). I think the issue is CT bringing any new and innovative to the market.

---

-
My tiny little slice of the internets: dcrainmaker.com

Though I planned to attempt this last year, the thread about combining the KK Rock & Roll frame to CT inspired me to dust off the old idea. CT doesn't have to get motivated to do anything to make it happen. Just a bit of hardware to combine 2 existing products & some software updates (for Erg Video or Trainer Road).

I don't know I think the market would be bigger then you think. I could see hardcore roadies using this also instead of riding outside. Also people in larger cities where it is just a PIA to get to a place to ride. A spin/trx (think circus ole hangie things or what ever you call it) just opened up near me with around 30 spin bikes. Be awesome to have 30 rollers set up like that with power.

---

__________________________________________________
Official Polar Ambassador
http://www.google.com/...P7RiWyEVwpunlsc2JtQQ

Hah, that's funny, but so true. Their hardware is rock solid. Their user interface and software really needs work. At one point the Computrainer fell into the category of "empowering innovation". Most companies that back into that, turn it into the "sustaining innovation" category as they add incremental features to the break through technology to sustain it over time and keep the ASP up and hang on to market share. Somehow the CT guys have done almost no sustaining innovation and have managed to keep their price up!!! WTF???

The wattage difference was more of an offset. A resistance setting of 50 watts on the CompuTrainer resulted in a 200 watt reading on my SRM and a resistance of 150 watts on the CompuTrainer resulted in a 300 watt reading on my SRM.

I did not try different pulley sizes as they cost $100.00 each to have made and making the pulley larger would have made it more difficult to achieve the required 25mph speed for calibration purposes. I also figured a different sized pulley would skew the speed reading capabilities of the CompuTrainer which would make the whole exercise moot.

---

I have a personality disorder, I don't drink coffee...