Why not the Polar chain-pickup system? That's what I'm thinking about. One pickup for the timing chain to get captain's power, another for the chainstay to get combined power off the stoker's HRM. Subtract for stoker power. The Polar recording HRMs move from bike to bike, so all you need is multiple pickups. Riders I know using the Polar power setup say they work +- about 5% compared with SRM.
I agree: Vector appears the most promising solution, especially if they release a Frog version (Speedplay road pedals being more oriented towards competition than rest stops). Frogs share the same spindles as X-series and Zeros, but it's a matter of Metrigear devoting the resources to calibrating frogs. In due time , I am sure.
The negative torque discussion is quite interesting. I'd never considered that before: only the stoker drive-side spider never sees negative torque. Vector will have no issues handling negative torque. And with the Quarq Qollector, you should be able to collect captain & stoker synchronized data, alleviating the challenge of aligning independently recorded ride files on two separate head units.
BTW you no longer need to rely on my flickr shots: it's now all on the Metrigear web site.
Interbike 2009 - New Quarq Power Meter Crankset Options
INTERBIKE 2009 - Quarq has been busy putting together powermeter rings for lots of crankset brands, with models for the following manufacturers coming out within the next two months:
* Rotor Agilis 3G
* Rotor 3D
* SRAM S975
* Cannondale Hollowgram SI / SL
* Lightning Carbon
* TA Specialties Carmina
All of these are pictured above or after the break, along with a prototype Collector data retrieval “black box” for cyclists that don’t want or use a big cycling computer on their bike (think track cyclists) but still want to collect power data.
...and Lightning is coming out with a tandem crankset.
Last edited by Ritterview; 09-26-09 at 04:29 PM.
Okay..... let me make sure I understand the problem with each position:
- Captain crankset: supports captain power only; supports negative torque, so must be a track version of a Quarq or SRM to operate correctly. Negative torque may occur when captain applies negative torque but the stoker applies a positive torque of greater magnitude such that the net drivetrain experiences positive torque and therefore the freehub does not disengage.
- Stoker crankset, timing chain: same as captain crankset, minus drivetrain losses in timing drivetrain.
- Stoker crankset, drive side: supports captain + stoker power - timing chain drivetrain losses, so can be used for total power. However, a triple crankset seems to be the preferred solution here, and I don't believe there's any sets which work with triple cranksets.
- Rear hub (PowerTap): reports total power - total drivetrain losses
The negative torque issue is a big one. For example, the Polar estimates power by estimating chain tension from the primary mode of chain vibrations. But this doesn't work if the monitored segment of chain (the top part) is slack. It also hasn't been calibrated to belt drives.
SRM makes a track version ($$$$$) and Quarq has one in development. So then perhaps you have a PowerTap for total power (- drivetrain losses), SRM track for captain power. But to display the difference would require a very clever head unit, and none are available yet (Qranium might do it; I'm not sure). Post-processing Qollector data would work, but that's not visible on the bike.
Metrigear Vector is the most promising solution, but may end up delayed from its '00Q1 release projection, especially if mountain bike pedals are to be used (Speedplay Frogs are not part of the Q1 projection). Brim Brothers (cleat system) is another possibility, as is the pedal system revealed at Eurobike. But I don't know if it works with mountain bike shoes, or for which pedal systems it is designed. There's another system which is designed to measure crank arm bending, and that would also be similar, but I really really doubt it's available for tandems.
Very interesting discussion. I don't have a negative torque issue on the Davinci of course. Our stoker crankset is to say the least a bit unusual for crankset solutions might be out. A pedal-based solution could be very interesting if SPDs were supported. iBike might be entertaining, but in the absence of wind I could estimate power output myself and I'm very suspicious of the ability of a $700 device to accurately measure relative wind direction and velocity and calculate its effect. In any event wouldn't take a tandem into account (or does it). A pure headwind isn't too bad, but 30 degrees one way or the other and we feel like two 1/2 bikes.
Volagi - Triple"ized" and Tubeless
daVinci Joint Venture
You could run a full right-side-drive system with the existing technology, I think.
Or put your Cinqo/SRM on the left side of the stoker crank - no reverse torque. But the crank would be turning backwards, so I don't think a stock Cinqo would work. I don't know if the SRM knows which direction it is turning.
I don't even use the offensive term "Fred." -- Sheldon "All Cyclists Are My Friends" Brown (1944-2008)
If I were Metrigear, I would focus on Shimano and Look pedals next. I am not sure why they picked Speedplay first other than Speedplay probably made an investment in the company. I am interested in a Metrigear solution for Shimano so that I can put power on my track bike without spending 3700 Euros.
the SYZR. So if Metrigear was inclined to work on a Speedplay mountain pedal, it would more likely be on the upcoming SYZR rather than the Frog.
As far as the Vector system on a tandem, I think you could get away with one set, and two headunits (or the cool Qollector for just post ride analysis). Left pedal for stoker, and right pedal for captain. Of course if you have some significant difference in power between L&R then the reading based on one side will not be as accurate.
we have characterized the power at each position during stationary tandem cycling and we have some data on the open road as well.
briefly, we used an srm at each postion and ran a right side only setup with the inner ring serving as the timing chain. we also had a powertap on the rear wheel.
check out some video at: www.pedalingforparkinsons.org
we have did a fair amount of testing to ensure accurate and consistent readings at both positions. at some point i may write it up and submit to a journal..though not too sure if there would be interest.
Clark Foy of Metrigear has been testing the Vector at the Low-Key Hillclimbs. He posted some of his data on their blog. As you can see, it's still prototype at this stage: some algorithm tuning is still occurring.
Algorithms are non-trivial. The naive approach would be to assume power = F · v. But if I push forward on both pedals and pull back on the handlebars while the bike is coasting, F · v > 0 while the bike is not being propelled, and muscles are doing no work. It's all Newton's Third Law. So the pedal needs to isolate the rotational velocity. This is just my wild speculation, but it seems to me from the data since they showed force data and not velocity data, the velocity data is the one for which algorithms are being tuned.
The data are very cool, though.
He was back at the next week's Low-Key Hillclimb, as well, this time with two pedals working together: Clark's on the outside in this photo going into the steep section.
It looks like the Metrigear will also record cadence as well according to the article.
I wonder how this will fit with the cadence/speed sensor that already comes with the Garmin 705. You will have two pickups for cadence displayed on the 705 unit, unless you use just the Metrigear cadence and the GarminGPS for speed
Has anyone run power measurement of both cranks and also the rear wheel to see how it all adds up. I am just wondering what the inefficiencies of a tandem are and what may be done to reduce them.
I noticed it when I did a race with a stoker of similar ability to myself. On solo bikes we can hill climb at very similar speeds to each other. When we raced on the tandem we were noticably slower climbing than when solo even though the power to weight ratio is the same. So I can only assume that power is being lost somewhere.
I assume you and your stoker have similar climbing cadences. Did you experiment with OOP? Are your respective HR in the same zone were you to be climbing solo? It seems possible that either differences in cadence or in where in the pedal stroke power is being applied could cause some inefficiency. How big a difference in climbing speed did you experience?
Volagi - Triple"ized" and Tubeless
daVinci Joint Venture
Last edited by Hermes; 11-16-09 at 12:58 PM.
Some climbing data....
- Brian and Janet on Mt Diablo in the Low-Key Hillclimbs: 110.3% of the male single median speed.
- Janet on Alba in the Low-Key Hillclimbs: 81.3% of male single median speed.
- Brian on Old La Honda at Low-Key Hillclimbs: 131.8% of median single male speed.
Okay: I'll assume the ratio of masses of Brian to Janet is 4:3. Then I should weight their speeds (roughly proportional to power:mass) proportionately. Doing this, using their individual speeds normalized to the median male single speeds, I get 110.2% for the predicted tandem speed. As a tandem they were 110.3%.
Not bad, all things considered.
I was interested to see how out of phase their cranks were: 90°:
Last edited by djconnel; 11-23-09 at 08:46 AM.
For anyone still tracking this thread, we've successfully implemented two independent power meters on our daVinci tandem. The DV drivetrain "made it easy" - no math needed, just some metalwork on behalf of our dealer to mate the Quarq units to the DV cranks. No need to flip the pedals (we ride Look, so it'd be messy to reverse them), and Quarq was able to program them to read power in the opposite direction. Unfortunately it's not a solution for standard tandem drivetrains.