Page 1 of 251

New powermeter rumours

Posted: Thu Feb 26, 2009 5:51 pm
by Francois_Viviers
In cycling magazine today, aparantly new powermeter. This time a pedal based powermeter, one on both sides, so measures bith legs independantly. Also supposed to be more affordable. Does annyone have any more info on this?

Please do share

Re: New powermeter rumours

Posted: Thu Feb 26, 2009 6:05 pm
by djconnel
I've also heard about this before. It's a bit of a challenge, as you want force · velocity (vector dot product), not |force|×|velocity|.

So what you need is some way to determine the speed and direction of pedal motion relative to the direction force is being applied to the pedal. And that force has two significant components, even relative to the pedal surface, as you can apply a motion normal to the surface, but with cleats locked in, an additional component parallel to the surface ("scraping mud off the bottom of your shoe", as Coach Troy encourages, after LeMond).

Re: New powermeter rumours

Posted: Thu Feb 26, 2009 6:05 pm
by Weenie

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

www.starbike.com



Posted: Thu Feb 26, 2009 8:13 pm
by timsterc
Could you tell me the name of the Cycling Magazine and the Month ?

Thanks.

Re: New powermeter rumours

Posted: Thu Feb 26, 2009 8:30 pm
by rruff
djconnel wrote:So what you need is some way to determine the speed and direction of pedal motion relative to the direction force is being applied to the pedal.


You'd need the component that is perpendicular to the crank... but how hard would this be? Seems like you could orient the strain gauges in that direction and it would take care of itself.

Re: New powermeter rumours

Posted: Thu Feb 26, 2009 8:39 pm
by djconnel
rruff wrote:You'd need the component that is perpendicular to the crank... but how hard would this be? Seems like you could orient the strain gauges in that direction and it would take care of itself.


Oh -- they'd need to be in the spindle, not the pedal (duh). Then you need to make sure the spindle is aligned properly in the crank, somehow. Errors would be second order (proportional to 1 - cos θ) in the rotation error.

Then you just need to know the rate of rotation. Using a cadence sensor and assuming constant rate of rotation is probably usually a decent approximation. You could do better by separately detecting each crank arm, but that would add complexity and mass.

I have no idea how to engineer a strain gauge in a pedal spindle, but then I know very little about how to design with strain gauges...

Posted: Thu Feb 26, 2009 8:41 pm
by J-Nice
I would think this would be the next evolution in power meters, and a logical step forward in the technology-not only to make the actual gadget affordable, but also to measure the force being exerted by each leg independently.

Re: New powermeter rumours

Posted: Thu Feb 26, 2009 8:41 pm
by jamiewilson3
rruff wrote:
djconnel wrote:So what you need is some way to determine the speed and direction of pedal motion relative to the direction force is being applied to the pedal.


You'd need the component that is perpendicular to the crank... but how hard would this be? Seems like you could orient the strain gauges in that direction and it would take care of itself.


You will need a strain gage in both directions to calculate the normal force. I used a set of these as part of a research study (not road worth - CNC machined blocks the size of a cigarette pack).

One problem with this is that pedals are a fairly high wear item as compared to a crank spider, since it contains small axles and bearings. The wear surfaces could be modular though.

Posted: Thu Feb 26, 2009 8:56 pm
by djconnel
You've also got to communicate the information off the pedal/spindle. On a crank like the Quarq, you have plenty of room for battery & electronics. With pedal stance (Q) and cornering clearance concerns, as well as individual preference on pedal design (for example preferring double-sided entry, different amounts of float, etc), it seems the crank is a much better location for this sort of thing. Hard to see a practical design.

Posted: Thu Feb 26, 2009 9:12 pm
by Francois_Viviers
not much info in article, this weeks cycling weekly in UK. Basically states in about 3 lines new powermeter, not crank hub or bottom bracket based. Pedal based, in theory cheaper and would give individual leg outputs

Posted: Thu Feb 26, 2009 9:42 pm
by rruff
djconnel wrote:Hard to see a practical design.


I tend to agree... if it is any good it won't be simple or cheap. You need to two complete meters (each pedal) plus they need to communicate... probably wirelessly... to the cpu. Hard to fit all that into a tiny space... but looking at a Shimano pedal for instance, they could put "stuff" in the hollow spot ahead of the spindle... maybe that is enough.

I thought of using a system were the chainrings and left crank are attached to an outer tube (optical torque measuring tube) while the right crank is connected to the left one via a separate (inner) spindle. Sort of like Ergomo but without the one leg limitations. I think with the large diameter BB tubes being used now, this would be viable.

Posted: Thu Feb 26, 2009 10:06 pm
by GrahamB
rruff wrote: You need to two complete meters (each pedal) plus they need to communicate... probably wirelessly... to the cpu. Hard to fit all that into a tiny space...


If it was actually on the spindle, hard wiring it to an electronics pack elsewhere on the crank might be doable... but for the other reasons, it still seems like a dumb idea. Aren't the elastic elements in most strain gauges aluminium? That's not going to work for pedal spindles...

Posted: Thu Feb 26, 2009 10:51 pm
by racyrich
My old teammate has spent 3 years writing software for cyclekinetics. They have designed a system of strain gauges and accelerometers in the sole of bespoke shoes. This would measure power independently obviously and also allow analysis of pedalling efficiency. They had some pretty heavyweight interest - eg a presentation to Bjarne Riis, plus coaching input from Heiko Salzwedel.

But...... the last I heard it was shelved as they ran out of money for hardware. My friend wrote the software for free for a small share of the company.
They still have a token website, bizarrely in Latin!

P.S. My mate Steve is proprietor of sdeals.com, if any of you have dealings with him.

Posted: Thu Feb 26, 2009 11:07 pm
by ericm
racyrich wrote:They still have a token website, bizarrely in Latin!


That's the standard typesetter (and now, web designer) placeholder text. http://www.lipsum.com/

Why not in the crank arms?

Posted: Fri Feb 27, 2009 12:24 am
by andrello
Why not in the crank arms? Does somebody hold a patent on that concept? That would be the best and most accurate place to take the measurements, yet nobody produces one. Why?

Why not in the crank arms?

Posted: Fri Feb 27, 2009 12:24 am
by Weenie

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

www.starbike.com



Posted: Fri Feb 27, 2009 12:49 am
by whodesigns
A friend of mine has developed a system in the crank arms. The German trackies have also used them over the years, Check out www.fes-sport.de