p2templin

+1 on this. Lately we're happy to be in the 'Steady' pace group (12-14mph) for our local club, so we in no way are racers. However, I attended a seminar on training with power, and learned great new guidance on how to manage my efforts for maximum endurance. Last September, we did a 60+50 mile weekend event with truly insufficient training. Based on what I learned, I really worked to keep an upper limit on my power at 70% of threshold (I predicted we'd be on the bike for 6 hours on day 1, 5 hours on day 2 given the dynamics of event riding). I've never felt so strong going into day 2 and even 75% of the way through day 2. (Mechanical issues with the stoker BB and a relentless headwind put us down into the 7-8 mph range on the final straightaways, and I wasn't at my best for the final 10 miles.)

waynesulak

DC Rainmaker is a pretty unbiased reviewer and he states when leaning the bike on the road hit pedal before pod and it appears the only way to hit the pod is on a curb. He does mention that off road all bets are off and some design work will need to be done to make it workable.

I understand that the angle of the pod does not have to be 90 degrees but that it does effect calibration so when moving pedals between bikes it might be an issue.

Dean V

Looking very nice. Make sure you post some power data from your next rides. We want to see who is doing all the work
We have done a couple of TT's with a powertap wheel. I know what power I put out so from deduction have a fair idea of my stokers contribution.

Chris_W

If you want to do this on the cheap, you could always buy just one pair of Vector pedals and split them between the front and rear of the tandem (plus a 2nd pair of Look compatible pedals obviously).

You could put both Vector pedals on the front for a while to find out how equal the captain's left-right power distribution is, then do the same with them both on the rear, then put one on the front and one on the rear and do some mental maths to correct for any left/right power discrepancies of each rider. Not the most user-friendly setup, but it saves $1500 plus a few grams that should keep the weight weenies happy.

waynesulak

Garmin has tried to head this approach off with two friends on singles. The two pedals talk to each other using a non-ANT link with one primary pedal and one slave. It would be interesting to see if the stokers pedals are close enough to work and defeat Garmin's defense strategy.

djconnel

This is an interesting question. Left power would be captain left, right stoker right, or vice-versa, and total would be the sum of these (roughly half of total). It might (not sure) be necessary to have pedals in phase.

I doubt the transmission has sufficient range, however.

Another option is what DC Rainmaker does, which is to use WASP to synchronize two independent ANT+ Sport power streams, so you could record the result of two Vector sets in the same file. This wouldn't allow live update, however, and visualization might require some hackery.

I think the pedal-based systems are the only system right now which works for both C & S separately. That's Vector and the more expensive Polar/Look.

Ritterview

Well, here's our debut ride (the stoker's data remains classified). The left captain's crank pedal stopped reading after 13 miles. The 16.9 mm depth of the Lightning crank arm proving problematic. It is being worked over more at Mike's Bikes even as we speak, so stay tuned!

Dean V

Great that you have got it working (sort of). Your power numbers look quite low relative to the average speed and the terrain. Unless your stoker was cranking out a fair bit more than you I would be suspicious of the accuracy.
For myself to average 28 kph on an averagely hilly ride would need an average power of around 180w. Weighted power 200w+. This is for a solo bike. For same result on tandem our combined power would be around 1.7x those numbers. I am 72kg and wife is 62kg.
From experience I would also say that if you are both in good aero positions it will take at least 400w (combined) to do 25mph on the flat. That is with zero gradient and wind which can be hard to detect so you really need to go both ways to get an indication.

Ritterview

For most of the ride, it was only registering the power from the right pedal. I discuss that in the Comments on the Strava upload.

p2templin

For traditional drivetrains, yes. I've been on record for two years that crank-based systems work on daVinci tandems.

Ritterview

Just did our first ride with working Vectors.

Contrary to widespread apprehension, the stoker likes having power. She can see now and have a record of her contribution, which before was just speculation.

We set a PR on Highway 9, of 50:12. Amazingly, if our numbers are plugged into Kreuzotter, with 2% wattage reduction for drivetrain losses, the estimate is 49:08, within 2% accuracy of our actual time.

djconnel

Nice!

I wonder about drive train efficiency, however. The rear drivetrain should be fairly efficient, as the chain is running under high tension, so the fixed losses associated with the low-tension part of the chain run are fractionally less important. But with a tandem you additionally have the captain losses from the front chain, which in your case is the majority of the power. This is a Gates belt, correct? I would expect at least 3% net drivetrain loss.

In any case, your earlier concerns about low readings seem to have been resolved.

It is interesting you PR'ed. No coincidence, I'd guess: feedback = faster.

Ritterview

I imagine you are right about the 3%, I had a citation for the 2% figure, so I went with that, but I think that is more of a half-bike figure.

I'll have to check, with 3% efficiency loss, Kruezotter might be spot on!

With this dual Vector setup, the questions we could answer with a Powertap rear wheel (disc, 135 mm OLD). How much power is lost in a tandem drivetrain before it gets to the rear hub? Then switch out a belt for a sync chain, and finally know whether the Gates sync belt is less efficient. How about a stiff versus flexy frame, does that rob power?

Bad1

Concerning drive systems here a link that answer that question: https://www.bikeradar.com/road/news/a...-faster-36074/

Ritterview

Yeah, that is the answer for a half-bike drive, with varying pulley sizes. It may well be a different answer for a tandem's identcal size sync pulleys. Then again, if the difference between belt and chain is around 1 watt, the Vector x 2 + Powertap set up might not be sufficiently sensitive to provide an answer.

waynesulak

The accuracy of consumer power meters is a problem with many small changes we might wish to measure. Most advertise +/- 2% which IF they meet that spec it leaves a 4% window. That is pretty large for any incremental change to bike setup.

Team Fab

I would like to know if it is a consistant +2% or -2% or does the +- vary dependant on enviromental conditions. If a unit measures a consistent +2% changes in the drivetrain may be detectable but if a power meter varies due to temperature or elevation then small changes could not be detected.

djconnel

Friction Facts tested a Gates. The report is free., although it may not be reproduced without permission. But I think it's safe to say that the Gates dissipates around 3-4 watts depending on power. Download the report for more details.

On the use of the power meter for drivetrain efficiency tests: I agree that it would work if the conditions are similar, especially ambient temperature, but as pointed out not for an absolute efficiency measurement, but as a comparison between one system and another at the same cadence and same transmitted power.

p2templin

Why not, instead, answer the question of how much you can improve your power output week-to-week, or across a season? Isn't that what power meters are for? Someone suggested I try the free Golden Cheetah software package, and for those with power it's the bomb. The graphs are really neat, and the 3D visualizations of the route are spectacular. I found it very informative to export my rides out of Garmin Training Center individually, then load them into Golden Cheetah in chronological order oldest to newest. After I'd load each one, I'd pick the Critical Power tab and let it build the graph: you can see the red ?threshold? curve bump up if the most recent ride represented a performance improvement. I've been amazed at which rides were positive and which ones, uh, weren't.

Ritterview

If we were to delve into the efficiency of Gates belt vs. a sync chain, it would be in addition to, rather than in lieu of, analyzing our power improvement (or lack thereof) over time.

Other tandem teams would be a lot more interested in the relative efficiency of a Gates drive than they would be if I am making 240 or 245 watts climbing Highway 9.

merlinextraligh

Just installed the Vector for the Captain's pedals. First half of the ride the power balance was way off. Recalibrated a couple of times and that appears to have fixed it.


Any thoughts on whether you're getting accurate independent power for both riders? Did an expirent with my feet off, and stoker pedaling probaly around 100-150 watts. That would read 10-20 watts on the Captain's vector.

My hope is that when we're both pedaling the Captain's crank is pulling the stoker's, rather than vice versa, and that the Captain's power reading is accurate. I'm going to do some field tests on my single and on the tandem to compare.

https://www.strava.com/activities/83343811

Turbotandem

The loss of power read at the captains pedals, from their to the back of the tandem, also would account for the fact that the captains power bends the frame, and on a left side drive transfers through the rear spindle, etc. etc. There's been a long technical discussion, I think on Wheelsmith's article about crank ratings, about where these losses may go. That is, when the frame or crank, or other parts "rebounds" from the flex, does that recoil power actually end up at the back wheel after all. Anyway, I would suspect the power that reaches the rear wheel based on a reading at the captains pedal would be more than 3%.

I've read the Friction Facts article comparing watts lost between tandem timing belts and chains about a month ago. It seemed to me a bit biased. They used Gates very old and since recanted position on 85psi tension in the belt. A more accepted 65psi (or less) I think cuts the watts lost in half. Along with other assumptions in their analysis, I believe they came up with a high figure for watts lost in a belt when compared to a chain. But even so, the lost watts they report are, IMHO, vanishingly small. Super interesting that they undertook the study, even if it seems commissioned by someone selling chains.

waynesulak

Quotes from the publication:

Turbotandem

RE: Friction Facts report:

The report, while using the aged 80psi tension, is oddly used the very newest 60t sprockets which were made for a
unique single tandem design. The standard is 69t. As the sprocket gets bigger, the tension actually goes down under a 250w or even 1000w rider force. So the belt can run at lower tensions and not under load the bottom length of belt. Maybe they used a 60t sprocket just so they could compare it to a small diameter chain ring timing chain.

It should also be noted that most of us spray a dry film silicone on the belt every week or two which noticeably reduces resistance when spinning by hand.

If the result is 2 watts penalty for the belt as reported (and I'd say the belt actually comes out ahead after setting it up as described above) that miniscule number is making a problem out of a solution. But even IF a loss, I'll take that for the other benefits of the belt. 2 watts is maybe equal to the power it takes to propel an extra 50 grams on the bike??? Well, the belt is lighter than a chain by more than that. Plus it is more responsive, clean, low maint, quiet, etc. This is also why high end motor cycles use a belt rather than a chain, and those belts see really huge wattage and higher tensions applied to them.

rahill

Actually, 2 watts is worth something on the order of a pound* on a moderate grade. However, it's hard to argue that the cleanliness, etc. of a belt isn't worthwhile.

* play with the numbers yourself using a calculator or just the simple numbers -- neglecting wind resistance and other speed effects (which are less important at typical climbing speeds), % reduction in total weight (bike + rider(s)) is proportional to % reduction in required power.