TrainerRoad

A light bike doesn't replace good fitness.

Moderator: Moderator Team

NatePearson
in the industry
Posts: 10
Joined: Thu Dec 15, 2011 10:17 pm

by NatePearson

Tapeworm wrote:So how do you apply an offset if you have no power meter?


We're looking at some roll down tests for that but you don't need to. That's the point. If your setup is the same you'll always have the same offset. With that you can track your progress as you get stronger. You can't however use your numbers to compare to others.

Fatmantis
Posts: 2
Joined: Mon Dec 19, 2011 9:20 pm

by Fatmantis

Nate, that's a persuasive graph. I'm sure there are many considerations involved, but that graph should be somewhere on your site (if it isn't).

Now, I was persuaded even without the graph, so I signed up without even seeing it. The site is a great training tool. I have used RPE and HRM in combination in the past, but the Virtual Power function makes workout calibration so much simpler. My HR drifts up quite a bit for workouts over an hour, so with Trainerroad I feel more confident that I am hitting my workout goals. I hope to use the site extensively over the coming months, so keep the improvements coming!

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



NatePearson
in the industry
Posts: 10
Joined: Thu Dec 15, 2011 10:17 pm

by NatePearson

Fatmantis wrote:Nate, that's a persuasive graph. I'm sure there are many considerations involved, but that graph should be somewhere on your site (if it isn't).

Now, I was persuaded even without the graph, so I signed up without even seeing it. The site is a great training tool. I have used RPE and HRM in combination in the past, but the Virtual Power function makes workout calibration so much simpler. My HR drifts up quite a bit for workouts over an hour, so with Trainerroad I feel more confident that I am hitting my workout goals. I hope to use the site extensively over the coming months, so keep the improvements coming!


Thanks!

I wanted to publish a bunch of data at once on the site...but I should start blogging/publishing this.

We're working on building a motor so we can make repeatable test and take out some of the variables. I'll also need to do this for just about every trainer we support. I suspect all trainers are not created equal. My goal is to give you guys really accurate data so you can make a decision which Trainer is best to use.

I should really add this to http://www.trainerroad.com/virtual-power page.

User avatar
Tapeworm
Posts: 2585
Joined: Thu May 03, 2007 10:39 am

by Tapeworm

That's so awesome that two people with four posts between them have come, not only to this forum, but this actual thread and pass comment about how fantastic this product is. What are the odds?

Oh and there's a money back guarantee too. Solid :thumbup:
"Physiology is all just propaganda and lies... all waiting to be disproven by the next study."
"I'm not a real doctor; But I am a real worm; I am an actual worm." - TMBG

wheelzqc
Posts: 266
Joined: Thu Jun 09, 2011 6:51 pm

by wheelzqc

I don't post at all, but I come everyday to read the Training forum... You do put alot of interesting information for a rookie.

P.S. I hope TrainerRoad will use the KK Road Machine first for the calibration :)

NatePearson
in the industry
Posts: 10
Joined: Thu Dec 15, 2011 10:17 pm

by NatePearson

Tapeworm wrote:That's so awesome that two people with four posts between them have come, not only to this forum, but this actual thread and pass comment about how fantastic this product is. What are the odds?

Oh and there's a money back guarantee too. Solid :thumbup:


They're pretty high.

We have a good amount of users and the same thing has happened on a few other forums:
http://www.beginnertriathlete.com/discu ... 60&start=1
http://forum.slowtwitch.com/gforum.cgi? ... ASC;mh=50;
http://www.tritalk.co.uk/forums/viewtop ... 93#1308193
http://www.bikeforums.net/showthread.ph ... ad-Reviews

I have Google alerts setup to let me know when someone mentions us. I also watch hits coming into our website and where they are coming from.

Tapeworm, I'm pretty sure you really really don't like us. And whatever I say or do won't change that :). I feel like I have to respond to your charges of fraud though, so that's why I'm posting this.

Fatmantis
Posts: 2
Joined: Mon Dec 19, 2011 9:20 pm

by Fatmantis

what are the odds of finding this thread when looking for info about trainerroad? using the search terms "trainerroad" and "forum"?

~the mystery of google~

thesimones
Posts: 20
Joined: Wed Mar 02, 2011 2:50 pm

by thesimones

TapeWorm,

I'll post up in support of TrainerRoad. The application and the workouts are excellent and it seems to me that this concept has some legs. Kudos to Nate so far and it seems to me that he is working on some new ideas within TrainerRoad to make it even more interesting. Nothing is as good as being out on the road but these focused sessions are pretty legit and once you plug in your FTP, all the workouts are custom fit for you. They have workouts that are 20 minutes...like the FTP test. All the way to 4 hours. Might not be for everyone but for those that are using a Trainer, it does help put some structure, interest and motivation into the workouts.

User avatar
Tapeworm
Posts: 2585
Joined: Thu May 03, 2007 10:39 am

by Tapeworm

Hope it's worth the monthly subscription fee.

Enjoy.
"Physiology is all just propaganda and lies... all waiting to be disproven by the next study."
"I'm not a real doctor; But I am a real worm; I am an actual worm." - TMBG

thesimones
Posts: 20
Joined: Wed Mar 02, 2011 2:50 pm

by thesimones

So far it has been. Hard pressed to find anything at $10/month these days and for what you get, I think it is worth the money. The site also has complete workout packages for those that would like to follow a six to eight week routine. The administrator is also very receptive to what the subscribers want and ideas that they think might make it a better site/product. I suggested for them to develop a forum and so far they are now planning on actually doing this.

The other thing that is nice is that all the workouts you do are immediately stored on their server and you can also download them in a .tcx if you want to import them into your WKO or Golden Cheetah.

NatePearson
in the industry
Posts: 10
Joined: Thu Dec 15, 2011 10:17 pm

by NatePearson

thesimones wrote:So far it has been. Hard pressed to find anything at $10/month these days and for what you get, I think it is worth the money. The site also has complete workout packages for those that would like to follow a six to eight week routine. The administrator is also very receptive to what the subscribers want and ideas that they think might make it a better site/product. I suggested for them to develop a forum and so far they are now planning on actually doing this.

The other thing that is nice is that all the workouts you do are immediately stored on their server and you can also download them in a .tcx if you want to import them into your WKO or Golden Cheetah.


Thanks man :).

An update on the forum. We plan on releasing it with the workout creator and online workouts. I'm looking at IP Board right now as my preferred choice but am open to other suggestions. I used to be an admin on Tri-Talk and they used IP Board so I'm familiar with it. They all seem pretty good though.

mrfish
Posts: 1749
Joined: Wed Mar 28, 2007 12:49 pm
Location: Near Horgen, Switzerland

by mrfish

Whether it's useful?
I'm a bit mystified about what having a function which converts your speed on the trainer into power is really worth. In terms of measuring training effect it's not really any better than measuring speed. If I can do 40kph for 20 minutes on my trainer, I have improved if I can do 40.1kph. The only advantage I can see is that power allows you to design a training program around FTP, which simplifies parameterisation of training design. But you know what - you could do the exact same thing with speed - this is how running training works after all. Garmin watches for running show your pace for 1 mile, and the training program says run X distance at Y pace or time .

Whether it's accurate, or has potential to be accurate
Then we come to the correlation between the Powertap reading and calculated power. A-level physics suggests you would use a cubic function along the lines A*speed ^ 3 + B * speed ^2 + C * speed + D = Calculated Power. (This is also how an ibike estimates power, but in slightly different form to account for your weight, height gain, barometric pressure and wind speed).

Now the interesting thing is that your chart shows a ~30W offset between calculated power and measured power. The difference is because your setup (tyre, pressure, roller contact, machine wear etc.) is different to the original setup used to calculate constants A, B, C, D in the power equation. Because the difference doesn't seem to vary with speed, you need to tweak constant D only. For simple linear offsets you could develop a procedure to do this without a power meter by timing a roll-down between 40kph and 30kph.

But then I thought some more and the picture is less clear - if I ride a 200g climbing rim with 32 round spokes a Tufo tub then it will come to a dead halt much more quickly than if I ride my old 1.5kg Aluminium Mavic Cosmic wheel with 16 bladed spokes. So now you need to account for changes in moment of interia and rotational wind resistance of the wheel itself to accurately set up the formula. Again you could pre-measure some typical wheels and allow the user to input some variables into the coefficient calculator, but again more hassle and measurement error potential.

And then the other thing is that if some of the non-linear parameters change (e.g. air resistance from those 16 bladed spokes versus a 36 spoke training wheel) you need more data points to solve the equation, so you need to add points to the roll-down curve, e.g. time for each 10kph interval down from 50kph. Now the problem becomes one of measurement error and ill conditioning - to get the constants right, my lab experience suggests you either need to average a lot of roll-down tests, and also have bike computers with almost zero lag and 'pre-averaging' of data. Some practical tests would be needed to see what the bounds of error really are.

So yes, I do think this method has potential to accurately allow users to measure power from almost any turbo trainer.

Whether it's worthwhile
I keep coming round to the same point - the user has four options:
1) Keep their setup constant and just measure speed or time to complete a given distance. Zero hassle
2) Keep their setup constant and do some moderately complicated and uncertain calibrations to come up with a power number. Some hassle but no better measurement-wise than 1.
3) Be free to change their setup and do a lot of calibrations to come up with an number. Likely there will be some error, but hopefully tests will mean you can build a procedure that captures the main variables.
4) Buy a power meter.

Since we're looking for small differences in performance, it does seem easier to just get a power meter. And then you only need to worry about whether your powermeter is calibrated, not anything else. And given a basic powertap hub is $500, and is considerably more useful than a website (not to mention re-sellable if you decide you don't like it), I know which I would go for.

Personally I think there is more mileage in the idea to offer an neat ANT+ onscreen display and structured workout plans than pseudo powermeter measurement.

commfire
Posts: 128
Joined: Thu Oct 22, 2009 11:14 pm

by commfire

mrfish wrote:Whether it's useful?


Personally I think there is more mileage in the idea to offer an neat ANT+ onscreen display and structured workout plans than pseudo powermeter measurement.


This is exactly how I use trainer road. Using a garmin Ant+ stick and a wireless SRM on my rollers. There are dozens of workouts to choose from based on TSS scores and even integrates with Sufferfest. For $10 a month it has proven to be well worth the money during the winter months. Definitely better than the hundreds of $30 dollar spin video that has been offered until now. Tapeworm, have you even used the product yet? I think you would probably change your opinion if you had
and before you go accusing me of sandbagging this forum I have zero association with this company.

NatePearson
in the industry
Posts: 10
Joined: Thu Dec 15, 2011 10:17 pm

by NatePearson

You bring up some good points.

Just about all the power curves aren't linear, so it's not a 1:1 relationship with speed to power. That's why it's better than just looking at speed in order to measure your improvements.

You're also right about spokes not having a linear relationship with power. I wonder how many watts that would be. Remember, the best power meters are +/- 1.5% accuracy. So if you're power meter reads 300 watts, you could be at 295.5 to 304.5. And that's only if they are calibrated correctly.

You're also right about the inertia it takes to speed up and slow down different size wheels. We don't calculate this or even try. How much are you losing? I don't think much. If you're doing 3x15 minutes your wheel is already spinning so you're not losing that extra power to speed up the wheel. With the nature of indoor riding where you are pretty stead state this works out well. If we were outside, with constant accelerations/deceleration, this wouldn't work well.

I think you're assessment is dead on in terms of usefulness. It's all a cost vs accuracy question for people. In terms of small measurements, I think VirtualPower is accurate enough to measure changes of 10-15 watts in your FTP. We've had a lot of beginner riders make gains of about 15 watts after doing a 6 week plan. That's a pretty big jump, but these are people who start with an FTP around 180-210.

After you figure out if you want to use a PowerMeter or use VirtualPower, then I agree again. It's all about doing the intervals and doing the work in a structured program. If we get you to ride more often and are less bored, then I think that makes it worth it for a lot of people.

Oh, and when you're FTP gets up around 300, you should just get a power meter because you are obviously a bad ass and into the sport :).

mrfish
Posts: 1749
Joined: Wed Mar 28, 2007 12:49 pm
Location: Near Horgen, Switzerland

by mrfish

Nate,

I think we are on the same wavelength. The power calculation thing is really a starter drug to get users hooked on the hard metrics :D Or until starter power meters come down to $200 (it will happen - SRM watch out).

Regarding wheel inertia, I agree it's unimportant for general riding. I just wanted to highlight that roll-down tests are not as simple as they look.

Best of luck on the new business.

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



Post Reply