any word on the ibike unit?

[acoggan]

Allright...what % of the potential users and what % of the time they use the dang thing do you think your "braking and pedalling at the same time" example would apply. That sounds like a pretty small nit to be picking.

I wasn't really trying to "pick a nit"...I just threw that thought in there because it occurred to me that there are, indeed, times when you might both brake and pedal at the same time.

I'm still not following how in your braking from ~50 mph to ~5 mph example the power calculation will get skewed. If the calculation turns out that for the given conditions power needs to be removed from the system, it's just going to display and record zeros and assume that braking is occurring. What am I missing?

Nothing, perhaps...I'm thinking in terms of total energy, not power.

In any case, my question isn't meant to be rhetorical in nature...I really am curious as to how they handle negative values for power.

 

[Tom Anhalt]

In any case, my question isn't meant to be rhetorical in nature...I really am curious as to how they handle negative values for power.

Well...just join the topica iBike list and ask the question there since the developers monitor and respond there...sounds simple enough

 

[RapDaddyo]

There you go with the "eyeball test" again...

That's my problem with looking at an iBike ride file side by side with an SRM or PT ride file. It doesn't cut to the question. Let's say there is a significant lag in the iBike's response to power changes. If one is managing power with the SRM or PT, the lag won't have the same effect and the power timeline will be pretty stable after the lag. If one is managing power with the iBike, he will be constantly over-correcting and the power timeline would be "wandering" all over the place. That's why the test protocol matters.

 

[Alex Simmons]

What happens on a banked velodrome in a bunch?

 

[tigermilk]

That's my problem with looking at an iBike ride file side by side with an SRM or PT ride file. It doesn't cut to the question. Let's say there is a significant lag in the iBike's response to power changes. If one is managing power with the SRM or PT, the lag won't have the same effect and the power timeline will be pretty stable after the lag. If one is managing power with the iBike, he will be constantly over-correcting and the power timeline would be "wandering" all over the place. That's why the test protocol matters.

I think that is going a little overboard. The PT has a slight lag, and if you increase the averaging to cut down on noise the situation is worse (though admittedly it's the only way I ride with the PT as I prefer not to have noisy data). My ergomo has a much longer lag. Neither is a detriment to training or managing the interval. I certainly don't "wander" with my powermeters - I don't ride my intervals as a slave to the real-time display but instead use it to stay within the prescribed zone. Gusts, bumps in the road, etc will all keep your power from being constant, and I know it's futile to use my brain, lungs, and legs as a sampler, signal processor, and actuator with a fast response. I use the real-time power data to trend the long-term (i.e., several seconds worth) trends rather than the transients.

There was nothing in the ibike data I saw that would have concerned me about response. Since I don't have one I can only assume the display is updated in a timely manner, but even if there is a 1-2 second delay, my experience with the ergomo has trained me to accept that at account for it in my interval management. I'm not going to get bent out of shape for a few seconds of "suspect" data over a 1-60 minute interval. I'm already anal as it is when it comes to power; I don't want to get even moreso.

 

[RapDaddyo]

I think that is going a little overboard.

Overboard? What do we use a PM for? To manage intensity of efforts and the most important efforts are our high-intensity efforts. I am suggesting that a rider attempt the same interval on the same course, managing power with the PT or SRM one time and the iBike unit the second time. We use PMs differently to manage power, with different accuracy. A direct comparison of how well the rider achieved his objective with a PT or SRM and with the iBike unit would be meaningful to me. This isn't a long or complicated set of tests. It's a typical workout, simply using first one device then the other to manage an effort. How is that overboard?

 

[tigermilk]

Overboard? What do we use a PM for? To manage intensity of efforts and the most important efforts are our high-intensity efforts. I am suggesting that a rider attempt the same interval on the same course, managing power with the PT or SRM one time and the iBike unit the second time. We use PMs differently to manage power, with different accuracy. A direct comparison of how well the rider achieved his objective with a PT or SRM and with the iBike unit would be meaningful to me. This isn't a long or complicated set of tests. It's a typical workout, simply using first one device then the other to manage an effort. How is that overboard?

And I have no doubt that any of the PMs out there will allow you to do that. Perhaps you should be specific about your concerns. Is it L5, L6, L7? I'd wager that L5 isn't an issue - the intervals are long enough and efforts smooth enough to properly modulate your efforts. L7? Those are all out anyway and over before you can really digest what's on the display. That leaves L6. I tend to use PE for those with power as a guide to keep me honest (no problem overshooting, it's the maintaining L6 that is harder for me).

If you really want to know if the rider achived his objective, it sounds like you're using the wrong PM anyway. We all seem to agree that NP is an effective tool, and only one PM has that functionality on the fly. Everything else is AP and we can only estimate what NP may have been for the work interval.

If you are worried about 2 PMs giving you different values such that one day you ride with a PT and the next an ibike and you worry if you are meeting your goals, well, that's a real problem. But anyone who has more than one brand of PM already deals with that. I've got a couple of PTs, a couple of older ergomos, and a CT. One of the ergomos is way out of calibration, and the CT reads 10-11% low compared to the PT. It's the responsibility of the rider to know his equipment, how well calibrated it is, and adjust the "human filter" when riding those different pieces of hardware. For example, I know that when I hop on one TT bike with my oldest ergomo that I need to see numbers that are 25% or so higher than when I ride with a PT.

Accuracy is one thing, and it certainly is desirable to have. However, the more important quantity is consistency. As long as you have that, you can train effectively no matter if the response of power meter #2 is linear or nonlinear compared to power meter #1.

 

[acoggan]

Well...just join the topica iBike list and ask the question there since the developers monitor and respond there...sounds simple enough

Sounds to me like they've got enough customer support issues to deal with already. Besides, I have a sneaking suspicion that one of the reasons that list was established was to avoid critical comments from people like me.

 

[RapDaddyo]

And I have no doubt that any of the PMs out there will allow you to do that.

Really? The issue is the iBike unit. You know that you can manage power with the iBike unit? Where is that data?

Perhaps you should be specific about your concerns. Is it L5, L6, L7?

All of the above if the grade and wind are typically variable. I would expect the iBike unit to be more effective the longer the effort and the more consistent the grade and wind. But, the real world of training is variable grades and wind conditions. It's a simple question. If I set out to ride an interval with my PT, I can manage the effort to within a few watts of my target. How well can I manage the same effort with the iBike? That's how I would use it. Why do you have a problem testing the unit for its intended purpose -- managing intensity of effort in the real world?

 

[Tom Anhalt]

Sounds to me like they've got enough customer support issues to deal with already. Besides, I have a sneaking suspicion that one of the reasons that list was established was to avoid critical comments from people like me.

If that's the case, why is it a public list, i.e. anyone can join?

Aww, c'mon...don't be a wus...

 

[Rocket^]

My ergomo has a much longer lag.

Do you have the old Ergomo? I have the new Ergomo Pro and while I don't have the documentation with me, I believe you can set it as low as one pedal revolution for data. I have mine set to the lowest setting and I don't see any appreciable lag. As a matter of fact, I have been thinking about setting it to one of the higher settings to smoothe things out. Regardless, the interval function definately allows for great control during intervals. It displays the current wattage as well as the average wattage.

 

[tigermilk]

Really? The issue is the iBike unit. You know that you can manage power with the iBike unit? Where is that data?

All of the above if the grade and wind are typically variable. I would expect the iBike unit to be more effective the longer the effort and the more consistent the grade and wind. But, the real world of training is variable grades and wind conditions. It's a simple question. If I set out to ride an interval with my PT, I can manage the effort to within a few watts of my target. How well can I manage the same effort with the iBike? That's how I would use it. Why do you have a problem testing the unit for its intended purpose -- managing intensity of effort in the real world?

No problem, but the limited data I have seen has demonstrated to me (at least) that the ibike has the ability to manage intensity given an appropriate environment. We've seen that the performance is atrocious on rough road and not worth a dime for training in that environment. I personally do my intervals on a course with about 1 mile of rough chip-seal. If I had an ibike I'd have to either avoid that stretch, make it a rest period, or ignore the data. Most of those options are not ideal.

Regarding small changes in wind and grade, take a look at the files I posted links to. Those, per the rider's input, were on variable grade and wind. To play devil's advocate, are you so sure that your current PM of choice is accurately capturing those micro-gusts and micro-grades? Is that one sample every 1.26s that the PT is recording REALLY indicative of the stress at the moment? I'd say no, but that doesn't bother me. The random nature will average out in the end, and the beauty of NP is that such events really do get smoothed out.

Another devil's advocate question is, are you really managing your interval to within a few watts with your PM? You say you can get within a few watts. Considering the error associated with the PMs out there is 1.5-10+%, depending on your PM of choice you may think you're within 1-2 W where in reality daily fluctuations (strain gage drift/out of cal/temperature for PT/SRM, small change in mass for Polar, degradation of BB for Ergomo) with those errors put you outside your 1-2 W tolerance.

 

[tigermilk]

Do you have the old Ergomo? I have the new Ergomo Pro and while I don't have the documentation with me, I believe you can set it as low as one pedal revolution for data. I have mine set to the lowest setting and I don't see any appreciable lag. As a matter of fact, I have been thinking about setting it to one of the higher settings to smoothe things out. Regardless, the interval function definately allows for great control during intervals. It displays the current wattage as well as the average wattage.

Yes it updates every revolution, but it uses 5 seconds of averages with each update. Start a sprint and notice the lag between when you put the power down and when the ergomo increases in wattage. Likewise, stop pedaling and see how long it takes to get to zero.

 

[RapDaddyo]

No problem, but the limited data I have seen has demonstrated to me (at least) that the ibike has the ability to manage intensity given an appropriate environment. We've seen that the performance is atrocious on rough road and not worth a dime for training in that environment. I personally do my intervals on a course with about 1 mile of rough chip-seal. If I had an ibike I'd have to either avoid that stretch, make it a rest period, or ignore the data. Most of those options are not ideal.

If the iBike unit wasn't used to manage an effort (with a known and reliable PM such as SRM and PT to capture the actual effort), how do you know that the iBike has the ability to manage intensity given an appropriate environment. This is where lag comes into play.

To play devil's advocate, are you so sure that your current PM of choice is accurately capturing those micro-gusts and micro-grades? Is that one sample every 1.26s that the PT is recording REALLY indicative of the stress at the moment? I'd say no, but that doesn't bother me.

The 1.26s recording interval is the average of ~60 observations per second. It is not a 1.26s snapshot of effort.

The random nature will average out in the end, and the beauty of NP is that such events really do get smoothed out.

I'm talking about managing the effort real-time, not just looking at the ride post-facto. I set out to do certain specific efforts, planned in advance. I know that I can manage those efforts very precisely with the PT. At least I know that I can manage the efforts precisely as recorded by the PT. If you're questioning the accuracy of the PT, that's a question that I think has been asked and answered.

Another devil's advocate question is, are you really managing your interval to within a few watts with your PM? You say you can get within a few watts.

Yes. As I said above, I can attain a result within a few watts as recorded by my PT. If I can do that with the iBike unit as recorded by a PT or SRM, then I would be much more inclined to recommend the unit to my friends who don't want to spend the bucks for a PT (much less an SRM). I'm more concerned with varying grade and wind conditions than with the rough road issue. I can easily avoid rough roads. If I were to limit my rides to consistent grade and wind conditions, I would be eliminating ~90% of my routes.

Anyway, you don't seem to like the idea of my concept of how to test the iBike unit. How would you test it?

 

[tigermilk]

The 1.26s recording interval is the average of ~60 observations per second. It is not a 1.26s snapshot of effort.

Anyway, you don't seem to like the idea of my concept of how to test the iBike unit. How would you test it?

From http://www.midweekclub.com/powerFAQ.htm Other than the Polar units, power-measuring devices display current power as an average over some short time period, which leads to a problem known as the “precession effect.” That is, unless you are pedaling at a rate where one or several revolutions are exactly completed in each averaging interval, an extra quarter-revolution can occur, and that partial turn of the crank may be either a power stroke or a dead-center (and perhaps the opposite for the next sampling period), which will produce a less consistent reading, especially for short intervals; the maximum power value captured in the PowerTap’s display memory, for example, is significantly affected, since it is the highest average value achieved over just 1.26 seconds. Thus, averaging over one (or just a few) crank revolutions would reduce variability in the current power display, track power more nearly as a rider senses it, and result in more accurate maximum values for instantaneous power. Recorded power values could, and perhaps should, still be based on time.

For current power, the PowerTap Standard displays only the power calculated every 1.26 seconds, and when set to record every 2.52 seconds, discards values calculated at 1.26 seconds, i.e., it records every other value without averaging. The Pro model, on the other hand, can display average over the last 1.26, 2.52, 5.04, 10.08, or 30.24 seconds for the current power value, but like the Standard, it records the instantaneous value at the selected recording interval, so for instance, when at the 10.08 second recording interval, every 8th value is stored, and the other 7 are discarded. Some have noted that displayed memory is often a couple Watts higher than what is downloaded. In fact, the “raw,” recorded data represents is the most accurate and unaltered information, coming directly from the hub. The reason the display is slightly off is that it uses lower-precision arithmetic, rounds improperly, or computes running averages using a method that is prone to accumulated errors or truncation. These corners are cut because memory and CPU computing power are at a premium.

The PT ain't perfect either.

How would I test? Just go out, ride a variable effort ride with a variety of wind conditions, grades, and road surfaces. Do some intervals. Compare post-ride. Unless you've got some incredible bio-feedback which allows you put out constant power, trying to monitor 2 power meters at once will make you go crazy. I've done that (er, attempted) with an ergomo and PT and because of the phase shifting between the two, it's impossible. Post-ride analysis is the only effective way to compare ibike (or any PM for that matter) to another PM.

Play around with the data I posted links to. Look between t=683-1061 for the first file, for instance. The ibike data starts out with a higher AP/NP but eventually gets closer. For that 6+ minute section, both AP and NP was within 2%. Considering the "Coggan levels" are a much wider band, are you really missing out on training effect with that kind of error? Heck, depending on where I do my workouts I'll get a lot more difference than that due to having to make a turn at an intersection.

 

[RapDaddyo]

How would I test? Just go out, ride a variable effort ride with a variety of wind conditions, grades, and road surfaces. Do some intervals. Compare post-ride. Unless you've got some incredible bio-feedback which allows you put out constant power, trying to monitor 2 power meters at once will make you go crazy. I've done that (er, attempted) with an ergomo and PT and because of the phase shifting between the two, it's impossible. Post-ride analysis is the only effective way to compare ibike (or any PM for that matter) to another PM.

Your test description is incomplete. When you say, "Do some intervals. Compare post-ride." What do you mean? How would you manage your intensity during the intervals? With which PM?

 

[acoggan]

From http://www.midweekclub.com/powerFAQ.htm Other than the Polar units, power-measuring devices display current power as an average over some short time period, which leads to a problem known as the “precession effect.”

I don't know who wrote that part of the FAQ, but it's incorrect, as the way the SRM handles the data makes it "immune" to any precession effect.

 

[acoggan]

I know that when I hop on one TT bike with my oldest ergomo that I need to see numbers that are 25% or so higher than when I ride with a PT.

Yikes! And to think that I once returned a PowerTap because, based on a static calibration, it was off by just 3%...

 

[rmur17]

Yikes! And to think that I once returned a PowerTap because, based on a static calibration, it was off by just 3%...

And I subjected my poor CT to the indignity of 'pot tweaking' to correct 5% under-reporting compared to my PT It survived the operation just fine however ...

rmur

 

[tigermilk]

Your test description is incomplete. When you say, "Do some intervals. Compare post-ride." What do you mean? How would you manage your intensity during the intervals? With which PM?[/size]

You're completely missing my point. I'd much rather have a large volume of data with varying degrees of intensity and durations to compare post-test. You are trying to characterize the performance of a device, and as such the larger the sample base, the better you can judge that device. It really is as simple as go out and ride and different intensities and durations, compare post ride. If you want to get more info, do one interval with PM#1 as the guide and repeat with PM#2 as the guide. Ultimately, however, it's all about collecting a substantial data set from which to assess performance.

I mean really, does it matter if your interval is done at isopower, a half sine wave, sawtooth, or a travelling salemen shape? Nope. Perhaps it's the engineer in me talking, but really it doesn't have to be as difficult as you make it out to be.