Monday, July 15, 2013

New boards from AP circuits, SRM takes pot shots at Garmin

DSC_9427
From left to right: V4.1, V4, V3, V3 circuit boards.
I like the purple ones best but they just don’t have the turn around time of APC.
In other news SRM seems to be taking pot shots at Garmin / Look Keo Power.
image
“Rider contact points, like the pedal, were initially looked at for prime PowerMeter placement. After extensive testing, SRM determined the pedal option would not give the same durable performance and accurate torque measurements as the crank spider”
This isn’t Garmin’s arrangement. Garmin is using a differential bending beam arrangement with piezo strain sensors. The differential bending beam would reduce much of the error in SRM’s design which is highly sensitive to pedal offset and angle torqueing as my experiments showed for the shear gauge on the crank arm.
image
However, measuring at the crank arm seemed like they couldn’t find fault. In fact it’s written as if this is their comparison case, but didn’t like that it only measured one leg each.
“In order to accurately measure power, SRM tested torque measurements in the crank arm, against the measurements in the spider. The crank spider is the first point where pedaling forces from the left and right leg meet, before that force is transmitted to the drive train”
Seems like they don’t have anything negative to say on this type of instrumentation. That’s at least a positive for me. As it should be as this is how a professional load cell would be setup essentially.

6 comments:

  1. Wow, now thats a fast turn around time

    ReplyDelete
    Replies
    1. Yup. They are super fast. My turn around time to get a board working, about 28 hours (8 sleeping, 9.5 at legit job, so not that bad)

      Delete
  2. Keith, been watching your little project with interest. Reveals that making a powermeter isn't just a matter of whacking a few strain gauges on a bit of the drive train. There is a lot to packaging something that is accurate, consistent, reliable and affordable. SRM's dissing of alternative measurement approaches is expected, but I'm not sure it holds true today. I have a pedal based (keo power), spider based (SRM) and crank based (Rotor) power meter. I can tell you that the Keo Power works very well once set up correctly with data almost indistinguishable from the SRM if run simultaneously. Look have done their homework yet their efforts go unnoticed while everyone speculates about the as yet non-existent Garmin vector... I can't say the rotor power meter works too well. There are problems with their cadence measurement approach (peak to peak torque) and rough roads which just doesn't jive...

    ReplyDelete
    Replies
    1. Thanks for your reply. It's quite true, this involves a lot of skill and engineering and for me that means learning new fields.

      Throughout the last year I feel like SRM's professionalism in marketing is waning. Everyone automatically assumes they are the gold standard. Stages has active thermal compensation, SRM does not for instance (Though I'm not a big fan of single sided measurement). One of my interests in L/R (even if not that useful) is that I was hit by a truck years ago and mangled my left knee. I want to know if I am protecting it and over stressing my right to compensate.

      Looks problems stem from elsewhere I figure, and price is the big one. Total package cost in Canada with GPS computer is over 3k. I can get an SRM or 2 Quarqs for that. I believe DCrainmaker covered the other one -- ANT+? Every now and then my Linkedin Profile is visited by the Director of interoperability of Polar. All I think is "What interoperability?". Fix price to Quarq Level, add ANT+ (Maybe BTLE but not ever before ANT+), ditch being Polar Exclusive. My opinions, but I don't think that many will argue against that.

      Rotor... I've heard rumours and that there are issues. Not much, but I think they limited how many got out in the wild. They are actually taking a lean startup approach which I'm impressed with. If they keep on that road they might shore up the issue. If I was to get further I can't deny that I might have similar issues.

      I'm using fairly idiot proof magnetic rotation sensors. It's a compromise, but accelerometers require a lot of signal processing, and gyro's are to heavy on current load making for rechargeable only and 40 - 50 hour run times at best. I wanted to head those roads but failure analysis reveals too much time and money.

      Delete
  3. Thanks for your thoughtful reply Keith. Yes Look have limited their market with their Polar partnership and pricing. The polar head units are very basic, though I'm advised that will change shortly with bluetooth transmitters for the keo and new bluetooth head units (maybe around Eurobike). While the polar units are basic they display and record the essential data very well. I have had numerous problems with my garmin edge headunits with dodgy firmware and flaky heart rate monitoring. If you look around the Keo Power pricing is competitive with Quarq. Sets with a head unit are regularly sold on e-bay by authorized sellers for $1500-$2000. I'll know shortly how responsive Rotor is to early feedback. They have been denying the cadence problem with customers. Either the next firmware release will address it (better algorithms) or they will need to revert to hardware modifications to augment the peak to peak approach with accelerometers (which seems to be approach adopted by P2Max). I agree individual leg measurement will be valuable for people recovering from injury. The new Rotor metrics available through their software interface are very interesting. It would be good if they released some information to end users to interpret the data better and provide advice on how to address apparent problems. Cheers

    ReplyDelete
    Replies
    1. You're right, Edge's have had issues. The new 510 has had a cadence issue and I have the lame "battery stuck at 100%" issue. These are good head units, but not perfect.

      I've only seen Look Keo pricing from a few places, and they weren't that low - I'll have to take a look at that and eat my words. BTLE -- mixed feelings on this. It's essentially cell phone technology, not sports technology. I've posted several times that in the Canadian market, putting a cell phone that costs 700 dollars on your bike with poor daytime view ability, being at the hands of an app developer, and a 3rd party manufacturer without any liability coverage for a mount is not my prefered avenue of data recording. Unless you get a IP65/67 certified phone (which 99% are not, Sony New Xperia Z is), they are easily susceptible to water damage which is immediately refused for warranty coverage. In the EU this is better I've heard: more phone choice, less being locked to providers, and easy availability of lower cost phones -- but none with BTLE, that's the 700+ territory still. This only serves to increase the cost. I can get 4 garmin edge 500's for the price of one iphone 5 + decent 3rd party waterproof mount with little benefit. Give this 2 years and it'll change. So this is a now versus future scenario.

      You obviously know more about Rotor. I've heard good things on their general customer service, but also other rumours. Here's to hoping.

      The Rotor metrics (TE / PS) are open to all ANT+ developers. My prototype will support it and I'll eventually get a 510 to see these metrics. If they are useful or not... that's up to the coaches and people like yourself.

      Delete