Where I got the 4hz: http://www.slowtwitch.com/Tech/Power_Meter_101_3643.html, it might be wrong though, but that was my experience too from trying to read speed & cadence sensors. Back then, very little of the ant+ protocol was available though, and so I was reverse engineering and not messing with the initialization protocol. It might actually be possible to change it, I’m still not clear and the ant+ stuff is still behind the alliance paywall, and I’m too lazy to dig out old ugly code and look at the initialization sequence, or worse, get out hardware and test it. Ughh.
ant: “The channel period represents the basic message rate of data packets sent by the master. By default, a broadcast data
packet will be sent (master) and received (slave), on every timeslot at this rate. The channel message rate can range from
0.5Hz to above 200Hz, with the upper limit dependant on the specific implementation. …] The default message rate is 4Hz, which is chosen to provide robust performance as described below. It is recommended
that the message rate be left at the default to provide more readily discoverable networks with good power and latency
characteristics.” (p18-19)
[pdf] http://www.thisisant.com/resources/ant-message-protocol-and-usage/
ant+ device profiles fix certain parameters, I can’t find out whether these include the rate: “ANT+ has device profiles that specify data formats, channel parameters and the network key. Examples of ANT+ Device” (p19)
Quarq power meters sample torque in a 60Hz macro-sample, each of which is composed of many sub-samples bringing the total sampling frequency to over 5000Hz. All of these samples are averaged for each discreet pedal revolution and reported to the head unit over the industry standard ANT+ protocol. i.e. If you are pedaling at 90rpm, power data will be received by the head unit every two-thirds of a second. Current generation ANT+ head units display and store power data once per second.
Apparently powertap only responds to every 5 message, at a 4hz frequency, so every 1.25 seconds on ant+:
“I’ve been meaning to post up software that I wrote that can capture ANT+ data quicker at the 0.25s intervals. This is how I found out that Powertaps only update at 1.25s intervals. They only update when they update the “wheel” page, which is every 5th message. www.keithhack.blogspot.ca”
From an electrical standpoint, the battery life is proportional to the data transmit rate. If 4Hz is 200 hours, ~1Mhz would only last a couple of minutes.
A lot of sampling is done on the strain gauges, but they are averaged like Jon said. Doing this consumes barely any power compared to an ANT+ message.
From an electrical standpoint, the battery life is proportional to the data transmit rate. If 4Hz is 200 hours, ~1Mhz would only last a couple of minutes.
A lot of sampling is done on the strain gauges, but they are averaged like Jon said. Doing this consumes barely any power compared to an ANT+ message.
Cheers.[/quote]
Mike, sure but you could record a bunch of time based data in the meter itself with a modicum of processor and a small memory module (which is basically there anyway) and then dump the entire time history in a data packet triggered on the cadence sensor cycle.
a cadence of 90 RPM is a mere 1.5 Hz, but the length of the data packet is entirely variable. So here the only assumption is one of constant angular velocity for one crack rotation in order to parse out the power vs angular position.
The polling rate is thus ultimately a largely secondary concern to the sampling rate of the sensors themselves, and the fidelity of what sampled data is transmitted to the head unit.
To go back to what Jon was saying, Quarq does 60Hz, but each poll has about 100 samples (sampling rate of about 55 Khz). And they time off the cadence sensor for synch. Which is basically the scheme I’m talking about.
The upstroke alone will not provide the observed results guys. I mentioned this in my preliminary discussions of the LPP pedaling process. It is the combination of the force of the upstroke and the inertial mass provided by the descending leg that delivers the observed results. In addition, no active force is applied on the downstroke. That is why there no appreciable work done by the Quads. The dead giveaway of LPP is that the toes are always effortlessly pointing down and there is very little variance in the foot position throughout the pedaling cycle. As I worked on the technique, I found out that the moment any active force is applied on the downstroke, the heel is activated and starts falling down…
The CT spin scan analyser is a great way to go! I have read what it can do and it will provide some excellent efficiency data relative to LPP…Thanks so much…Really…AWESOME INPUT!
I’m totally and pointedly disregarding your foul language and comments. Don’t waste one second your time directing anything to me. It is not registering, it is not being computed, it is not being accepted with any kind of serious intent, no credibility, nothing…Really… I don’t care one bit…
Armed with quite a bit of information, reprobation, motivation, support, good advice, not so good one too, great cartoons form David and my firm promise to MikeB to never give up on LPP, I bid all of you, including my thorn in the back Jason L a fond farewell.
It is through challenge and altercation that we test our beliefs. This experience has only strengthen mine. I will continue using, teaching others to use, promoting, studying and advancing Lift Propulsive Pedaling. Conversations 7, 9,11,13,16 and 17 describe my perspective of it.
I will also do my best to put a considerable amount of science behind it so that its performance characteristics can be more properly explained. I shall find the means to do so since Kickstarter funding sources are more interested in pasta salads than pedaling! I will definitively run LPP through the paces with the CT spin scan analysis that Justin suggested. My plan is to ultimately provide technical data, actual race results from athletes using it and material developed for on a Lift Propulsive Pedaling website. That way, those who are interested in following its progress can access the information.
Thanks so much for the interactions and please take care of yourselves on the road!
Jon did actually bring up some interesting points about spin scan. Most power meters have limited number of feedback per second which makes the calculation of angular torque more of an estimation. However, what jason brought up is also true. In laboratory testing of angular torque for cycling, they do have modified power meters and programs that can give torque readouts for all 360 degrees. So you would need to invest a little more money than just for your standard power meter for accurate data to be taken seriously. Furthermore, if you really want to prove that the quadriceps are truly not activated during the downstroke motion, you should also be using muscle EMGs during testing to know if there is actually no active contraction.
