Power without hub or BB

[Donald Munro]

http://www.ibikesports.com/techinfo.html

Has anyone tried one of these with another power meter on at the same time
to test accuracy ?

[ilan]

Donald Munro wrote:
> http://www.ibikesports.com/techinfo.html
>
> Has anyone tried one of these with another power meter on at the same time
> to test accuracy ?

I've been looking at power meters, and found this description for
battery replacement
for the Powertap SL, which convinced me not to get it:

27
HUB BATTERIES
Normal battery life in the hub is 250 to 300 hours of actual ride time.
Use #357 Silver Oxide type
batteries and always replace batteries in pairs. Alkaline type
batteries give shorter life and poorer performance.
Battery replacement
-Unscrew the plastic battery cover. Use a spanner wrench or bench vise
opened wide to get it started if stuck.
There are O-ring seals that can cause the cover to resist the start of
motion. The threading is normal right hand.
-Remove the inner O-ring for ease of installation.
-Remove the battery pack. A small screw driver can be used back and
forth between the to ends to start to lift
it out.
-Pop the battery out by bending the plastic retaining tap back and
pushing the batter up from the bottom.
-Check that the electrical connector tab from the inside of the hub is
perpendicular to the bottom of the battery
pocket. If the tab gets bent, gently press it back to perpendicular
with a non-sharp tool.
-Slide the battery pack back into position along the center core. As
the battery is engaged a small increase of
resistance to installation should be felt as the electrical connections
are made.
-Reinstall the O-ring.
-Grease the inside of the cap where the inner and outer O-rings
make contact. Use a high quality bicycle grease.
-Thread the cap all the way back on.
-Replacement O-rings and battery
packs are available from Saris
Cycling Group.

-ilan

[Robert Chung]

Donald Munro wrote:
> http://www.ibikesports.com/techinfo.html
>
> Has anyone tried one of these with another power meter on at the same
> time to test accuracy ?

Several people have, but usually they compare only average power.

We looked at a couple of ibike-PT files a few months ago on the Wattage
list (for example, http://tinyurl.com/ye9atp ). Since then, they've
upgraded the ROM in the head unit to minimize the drop-out issue but I
haven't analyzed any additional data to know whether any other problems
have been addressed. At the moment, I'm guessing (though I emphasize this
is entirely a guess) that the ibike is quite reasonably good for riding
around, though it may not be nearly as good for structured training -- and
of course it's completely useless for indoor training, whether structured
or not. If that's what you do, that's a deal killer. If you don't care
about that sort of thing, it seems like a bargain.

It appears to be a power meter to ride with, not a power meter to ride to.

[Donald Munro]

Donald Munro wrote:
>> http://www.ibikesports.com/techinfo.html
>> Has anyone tried one of these with another power meter on at the same
>> time to test accuracy ?

Robert Chung wrote:
> Several people have, but usually they compare only average power.
> It appears to be a power meter to ride with, not a power meter to ride to.

Thanks, I'll stick to my PT for the moment. Some of the reviews I read
suggest an interesting application for it. Apparently its quite sensitive
to changes in rider position so it might be useful for optimizing
aerodynamics even if it isn't 100% accurate; that is if they can make it
work with TT bars.

[don vescio]

I've been using one for a little over two months on a bike with an
Ergomo. Depending upon how much care is taken during setup and
calibration (which are pretty easy), I've found that the iBike is
reasonable close to my Ergomo, +/- 10%. For me, it gives a good
ballpark reading during workouts and useful data when downloaded into
CyclingPeaks. It doesn't do well on rough or chipsealed roads, and I
haven't found it to be useful for sprint workouts. Also, occasionally
odd high readings are displayed when coasting down a hill at high
speeds. For $400, it does much better than I would have expected.

I like it for its simplicity and convenience and the fact that its
readings seem consistent from workout to workout (and not to mention
that the device seems pretty indestructable). I think that there
certainly is value to the device, especially for those who don't have
the funds for a PT, Ergomo, or SRM. I'm at the point where I'd be
comfortable using it as a primary device.

Note that opinions will vary significantly on this device ;)

Don

[[email protected]]

don vescio wrote:
> I've been using one for a little over two months on a bike with an
> Ergomo. Depending upon how much care is taken during setup and
> calibration (which are pretty easy), I've found that the iBike is
> reasonable close to my Ergomo, +/- 10%. For me, it gives a good
> ballpark reading during workouts and useful data when downloaded into
> CyclingPeaks. It doesn't do well on rough or chipsealed roads, and I
> haven't found it to be useful for sprint workouts. Also, occasionally
> odd high readings are displayed when coasting down a hill at high
> speeds. For $400, it does much better than I would have expected.
>
> I like it for its simplicity and convenience and the fact that its
> readings seem consistent from workout to workout (and not to mention
> that the device seems pretty indestructable). I think that there
> certainly is value to the device, especially for those who don't have
> the funds for a PT, Ergomo, or SRM. I'm at the point where I'd be
> comfortable using it as a primary device.
>
> Note that opinions will vary significantly on this device ;)
>
> Don

Can you please elaborate on the rough roads issue?

Joseph

[ilan]

http://www.nyvelocity.com/content.php?id=889

-ilan

Donald Munro wrote:
> http://www.ibikesports.com/techinfo.html
>
> Has anyone tried one of these with another power meter on at the same time
> to test accuracy ?

[ilan]

It is mentioned in this review (note that it refers to another rbr):
http://www.roadbikerider.com/producttests.htm
I liked Fred Matheny's books, but in this case he makes a number of
simple
errors, in particular, his description of the Ergomo system is
completely incorrect.

-ilan


wrote:
>
> Can you please elaborate on the rough roads issue?
>
> Joseph

[Phil Holman]

> wrote in message
oups.com...
>
> don vescio wrote:
>> I've been using one for a little over two months on a bike with an
>> Ergomo. Depending upon how much care is taken during setup and
>> calibration (which are pretty easy), I've found that the iBike is
>> reasonable close to my Ergomo, +/- 10%. For me, it gives a good
>> ballpark reading during workouts and useful data when downloaded into
>> CyclingPeaks. It doesn't do well on rough or chipsealed roads, and I
>> haven't found it to be useful for sprint workouts. Also,
>> occasionally
>> odd high readings are displayed when coasting down a hill at high
>> speeds. For $400, it does much better than I would have expected.
>>
>> I like it for its simplicity and convenience and the fact that its
>> readings seem consistent from workout to workout (and not to mention
>> that the device seems pretty indestructable). I think that there
>> certainly is value to the device, especially for those who don't have
>> the funds for a PT, Ergomo, or SRM. I'm at the point where I'd be
>> comfortable using it as a primary device.
>>
>> Note that opinions will vary significantly on this device ;)
>>
>> Don
>
> Can you please elaborate on the rough roads issue?

*I'm guessing* it doesn't measure rolling resistance. It may have a
nominal value built in depending on speed but on a rough road where RR
is high, it doesn't account for a significant increase.

Phil H

[Daniel Connelly]

Phil Holman wrote:
> > wrote in message

>> Can you please elaborate on the rough roads issue?
>
> *I'm guessing* it doesn't measure rolling resistance. It may have a
> nominal value built in depending on speed but on a rough road where RR
> is high, it doesn't account for a significant increase.
>
> Phil H
>
>

It measures the following:

* acceleration in the direction of travel
* road grade (derived from the same accelerometer): note this differs
from many computers, which compute grade from measured speed and the
measured rate of altitude gain. This works even if the bike is still.
* measured speed
* measured wind speed relative to the bike

It then calculates power, using coefficients derived from calibrations
(coast down measurements), and from reported system weight. So if
conditions differ from the coast-down, or if the weight is different
than is reported, the result will be in error.

Road grade is calibrated by measuring grade, flipping the bike around,
and measuring it again. The difference between the average and zero is
the offset which must be applied.

Dan

[Robert Chung]

wrote:
>
> Can you please elaborate on the rough roads issue?

When going over rough roads the power readings would either peak high or
else (under the older version of the firmware) drop out. I have no direct
experience with this, but reportedly the current version of the firmware
handles drop outs by simply holding the previous power reading constant
until conditions return to normal. Whether this is a good solution or not
depends on your point of view.

The same sort of thing happens when the bike makes a sharp turn, as around
a corner.

It appears to react quickly to positive accelerations but lag a bit when
decelerating (i.e., it ramps up fast but ramps down slowly). I'm guessing
it would do best on steady efforts, especially long steady climbs, and
would probably do worst on something like sharp-cornered crits, sprints,
interval workouts, or on the pave at Paris-Roubaix.

Reportedly, if you calibrated it for the brake hoods and then suddenly get
down into the drops, the iBike will record that you just increased your
output by dozens of watts.

On the other hand, it's half the price of a PT (and much less than half
the price of an SRM or Ergomo), much lighter, and there are no wires
unless you want to track cadence.

[SLAVE of THE STATE]

Daniel Connelly wrote:
> Phil Holman wrote:
> > > wrote in message
>
> >> Can you please elaborate on the rough roads issue?
> >
> > *I'm guessing* it doesn't measure rolling resistance. It may have a
> > nominal value built in depending on speed but on a rough road where RR
> > is high, it doesn't account for a significant increase.
> >
> > Phil H
> >
> >
>
> It measures the following:
>
> * acceleration in the direction of travel
> * road grade (derived from the same accelerometer): note this differs
> from many computers, which compute grade from measured speed and the
> measured rate of altitude gain. This works even if the bike is still.
> * measured speed
> * measured wind speed relative to the bike
>
> It then calculates power, using coefficients derived from calibrations
> (coast down measurements), and from reported system weight. So if
> conditions differ from the coast-down, or if the weight is different
> than is reported, the result will be in error.
>
> Road grade is calibrated by measuring grade, flipping the bike around,
> and measuring it again. The difference between the average and zero is
> the offset which must be applied.


I had been wondering about how to practically deal with the static
acceleration problem ("auto finding" the reference in a cheap,
reliable, and physically robust manner) for some time. So you got me
looking:

http://www.analog.com/en/subCat/0,2879,764%255F800%255F0%255F%255F0%255F,00.html

Ultra cool. I hadn't been keeping track of MEMS development. Nice.
Maybe even I will buy a power meter.

[Mike Jacoubowsky]

>> Can you please elaborate on the rough roads issue?

My guess, and it's only a guess, is that rough roads cause an issue with the
built-in inclinometer, which, if off even slightly, is going to do quite a
number on the readings. This could also be the issue when going around tight
corners.

--Mike-- Chain Reaction Bicycles
www.ChainReactionBicycles.com


"ilan" > wrote in message
ups.com...
> It is mentioned in this review (note that it refers to another rbr):
> http://www.roadbikerider.com/producttests.htm
> I liked Fred Matheny's books, but in this case he makes a number of
> simple
> errors, in particular, his description of the Ergomo system is
> completely incorrect.
>
> -ilan
>
>
> wrote:
>>
>> Can you please elaborate on the rough roads issue?
>>
>> Joseph
>

[[email protected]]

Robert Chung wrote:
> wrote:
> >
> > Can you please elaborate on the rough roads issue?
>
> When going over rough roads the power readings would either peak high or
> else (under the older version of the firmware) drop out. I have no direct
> experience with this, but reportedly the current version of the firmware
> handles drop outs by simply holding the previous power reading constant
> until conditions return to normal. Whether this is a good solution or not
> depends on your point of view.
>
> The same sort of thing happens when the bike makes a sharp turn, as around
> a corner.
>
> It appears to react quickly to positive accelerations but lag a bit when
> decelerating (i.e., it ramps up fast but ramps down slowly). I'm guessing
> it would do best on steady efforts, especially long steady climbs, and
> would probably do worst on something like sharp-cornered crits, sprints,
> interval workouts, or on the pave at Paris-Roubaix.
>
> Reportedly, if you calibrated it for the brake hoods and then suddenly get
> down into the drops, the iBike will record that you just increased your
> output by dozens of watts.
>
> On the other hand, it's half the price of a PT (and much less than half
> the price of an SRM or Ergomo), much lighter, and there are no wires
> unless you want to track cadence.

Any idea how rough the surface must be to cause these drop-outs? I
don't do crits, and I am not interested in acceleration, so aside from
the rough road issue this might be just what I need. After reading the
review linked in another post, it seems perfect for my needs. Unless of
course my roads are too rough for it to work. $400 seems a good price
even considering it's limitations, but $400 would be a waste if if it
hardly ever worked due to this rough roads issue.

Joseph

[DirtRoadie]

Robert Chung wrote:
> It appears to react quickly to positive accelerations but lag a bit when
> decelerating (i.e., it ramps up fast but ramps down slowly). I'm guessing
> it would do best on steady efforts, especially long steady climbs,

Speaking from experience with the iBike, that is correct. It appears to
be nearly flawless on steady climbs, especially where the measured
"apparent wind" is low or zero. Accordingly the power calculation
comes from a fairly straightforward (and verifiable) formula involving
only speed, weight, and grade (with only minor adjustments for friction
and wind speed).

> On the other hand, it's half the price of a PT (and much less than half
> the price of an SRM or Ergomo), much lighter, and there are no wires
> unless you want to track cadence.

There IS a wire for a front wheel speed sensor, just like any other
wired cyclometer.
But it does not require special wheels, cranks, bottom brackets, etc.,
and can be readily installed and used on nearly any bike (Probably
would not work well on any bike with suspension, since the fore/aft
"tilt" can vary).

Aside from its power calculating function, it's very much a "recording"
cyclometer allowing the following to be recorded at 1 or 5 second
intervals:
Speed
Apparent wind speed
Elevation
Grade
Power
Cadence (with optional adapter)
Heart Rate (with optional adapter, not yet offered for sale)

For those who like data, it's interesting to be able to see a graph
showing, for example, actual speed vs. wind speed (Yes, that personal
best last week was the result of a tailwind).

It should be noted that the software that comes with the unit has only
"bare bones" functionality with lots of room for improvement,

DR

[DirtRoadie]

Robert Chung wrote:
> It appears to react quickly to positive accelerations but lag a bit when
> decelerating (i.e., it ramps up fast but ramps down slowly). I'm guessing
> it would do best on steady efforts, especially long steady climbs,

Speaking from experience with the iBike, that is correct. It appears to
be nearly flawless on steady climbs, especially where the measured
"apparent wind" is low or zero. Accordingly the power calculation
comes from a fairly straightforward (and verifiable) formula involving
only speed, weight, and grade (with only minor adjustments for friction
and wind speed).

> On the other hand, it's half the price of a PT (and much less than half
> the price of an SRM or Ergomo), much lighter, and there are no wires
> unless you want to track cadence.

There IS a wire for a front wheel speed sensor, just like any other
wired cyclometer.
But it does not require special wheels, cranks, bottom brackets, etc.,
and can be readily installed and used on nearly any bike (Probably
would not work well on any bike with suspension, since the fore/aft
"tilt" can vary).

Aside from its power calculating function, it's very much a "recording"
cyclometer allowing the following to be recorded at 1 or 5 second
intervals:
Speed
Apparent wind speed
Elevation
Grade
Power
Cadence (with optional adapter)
Heart Rate (with optional adapter, not yet offered for sale)

For those who like data, it's interesting to be able to see a graph
showing, for example, actual speed vs. wind speed (Yes, that personal
best last week was the result of a tailwind).

It should be noted that the software that comes with the unit has only
"bare bones" functionality with lots of room for improvement,

DR

[DirtRoadie]

Phil Holman wrote:

> *I'm guessing* it doesn't measure rolling resistance. It may have a
> nominal value built in depending on speed but on a rough road where RR
> is high, it doesn't account for a significant increase.

For all practical purposes, it does "measure" rolling resistance as
part of its "coastdown" calibration procedure, whch generates values
for "aero" and "frictional" resistance. The calibration could be done
on a smooth surface or a rougher one and that would be taken into
account in the power calculations.
However once the rolling resistance value has been determined, that
value is used for all its power calculations. It cannot adjust "on the
fly" for differences in road surfaces which may affect the rolling
resistance.

Here's a description of the concept at an unrelated site:
http://oxide.eng.uci.edu/Personnel/max/Final%20Report-Velocomputer.pdf

The "rough road" dropout issue of the iBike appears to be more a matter
of road vibration affecting the accelerometer measurements thereby
affecting calculation of power.

DR

[[email protected]]

So how does the Ibike deal with changes in wind speed. It would be
useful if the Ibike had a recalibration reset button to deal with
changes in road surface or wind speed.
DirtRoadie wrote:
> Robert Chung wrote:
> > It appears to react quickly to positive accelerations but lag a bit when
> > decelerating (i.e., it ramps up fast but ramps down slowly). I'm guessing
> > it would do best on steady efforts, especially long steady climbs,
>
> Speaking from experience with the iBike, that is correct. It appears to
> be nearly flawless on steady climbs, especially where the measured
> "apparent wind" is low or zero. Accordingly the power calculation
> comes from a fairly straightforward (and verifiable) formula involving
> only speed, weight, and grade (with only minor adjustments for friction
> and wind speed).
>
> > On the other hand, it's half the price of a PT (and much less than half
> > the price of an SRM or Ergomo), much lighter, and there are no wires
> > unless you want to track cadence.
>
> There IS a wire for a front wheel speed sensor, just like any other
> wired cyclometer.
> But it does not require special wheels, cranks, bottom brackets, etc.,
> and can be readily installed and used on nearly any bike (Probably
> would not work well on any bike with suspension, since the fore/aft
> "tilt" can vary).
>
> Aside from its power calculating function, it's very much a "recording"
> cyclometer allowing the following to be recorded at 1 or 5 second
> intervals:
> Speed
> Apparent wind speed
> Elevation
> Grade
> Power
> Cadence (with optional adapter)
> Heart Rate (with optional adapter, not yet offered for sale)
>
> For those who like data, it's interesting to be able to see a graph
> showing, for example, actual speed vs. wind speed (Yes, that personal
> best last week was the result of a tailwind).
>
> It should be noted that the software that comes with the unit has only
> "bare bones" functionality with lots of room for improvement,
>
> DR

[DirtRoadie]

wrote:
> So how does the Ibike deal with changes in wind speed. It would be
> useful if the Ibike had a recalibration reset button to deal with
> changes in road surface or wind speed.

I'm not sure what you are asking. The iBike measures "apparent" wind
speed (wind pressure) using a pressure sensor on the front of the
device, and measures actual bike speed using a sensor on the front
wheel just like any other cyclometer. In calm air bike speed and wind
speed are the same. With a headwind, the wind speed is higher than
actual speed. And so on. From these values and the "aero drag"
coefficient (determined with a "coastdown" test) the iBike calculates
the drag forces and the power necessary to overcome them.
So changes in wind speed ARE measured and taken into account in the
power calculation.

Frictional drag is another matter. The only way to recalibrate for a
difference in road surface is to do another "coastdown" on the surface
for which proper calibration is sought. It should also be noted though,
that frictional drag is a small part of the entire drag, so variation
in road surface is not going to account for a huge change in power
readings. For most riding, the most signifciant forces impeding
forward motion come from either aero drag or gravity.

DR

[[email protected]]

DirtRoadie wrote:
> wrote:
> > So how does the Ibike deal with changes in wind speed. It would be
> > useful if the Ibike had a recalibration reset button to deal with
> > changes in road surface or wind speed.
>
> I'm not sure what you are asking. The iBike measures "apparent" wind
> speed (wind pressure) using a pressure sensor on the front of the
> device, and measures actual bike speed using a sensor on the front
> wheel just like any other cyclometer. In calm air bike speed and wind
> speed are the same. With a headwind, the wind speed is higher than
> actual speed. And so on. From these values and the "aero drag"
> coefficient (determined with a "coastdown" test) the iBike calculates
> the drag forces and the power necessary to overcome them.
> So changes in wind speed ARE measured and taken into account in the
> power calculation.
>
> Frictional drag is another matter. The only way to recalibrate for a
> difference in road surface is to do another "coastdown" on the surface
> for which proper calibration is sought. It should also be noted though,
> that frictional drag is a small part of the entire drag, so variation
> in road surface is not going to account for a huge change in power
> readings. For most riding, the most signifciant forces impeding
> forward motion come from either aero drag or gravity.
>

I suppose one could do a few coast downs on different surfaces and in
different positions so you could mentally adjust on the fly. Ie "Rouch
road = add 20W, drops= subtract 25W" or whatever. This would kind of
make the recorded info not very interesting, but it could make the live
info for pacing or training more useful.

Joseph

[Dan Connelly]

DirtRoadie wrote:

> It should also be noted though,
> that frictional drag is a small part of the entire drag, so variation
> in road surface is not going to account for a huge change in power
> readings.

A typical rolling resistance fraction is 12% of total power. This could
easily double on rough roads. 12% variations in power are small by some
standards but quite significant for the purposes of considering power data.

For example, suppose I am doing a climb on a 5% grade in which the Crr
changes from 0.5% to 1% due to rougher roads where the road crosses a
county line (different maintenance schedules). If wind resistance is
roughly 10% of power during the climb, this corresponds to a 9% change
in power required at a given speed. Analyzing data where useful power
appears to sag by 9% might suggest that I started too fast, and should
have saved more at the beginning, when in reality, my pace was optimal.

Dan

[Michael Press]

In article
om>,
"ilan" > wrote:

> http://www.nyvelocity.com/content.php?id=889

"Secondly, even though aero drag increases as a square
of the speed, the power necessary to overcome that drag
increases as a cube. In other words, drag force will
quadruple if your speed doubles, but you'll need eight
times as many watts to overcome it. The reason is that
you're traveling twice as fast, so you're pushing
through twice as much air. Given this exponential
increase, ..."

Bzzzzzzzzzzzt.

--
Michael Press

[DirtRoadie]

Dan Connelly wrote:
> A typical rolling resistance fraction is 12% of total power. This could
> easily double on rough roads. 12% variations in power are small by some
> standards but quite significant for the purposes of considering power data.

> For example, suppose I am doing a climb on a 5% grade in which the Crr
> changes from 0.5% to 1% due to rougher roads where the road crosses a
> county line (different maintenance schedules). If wind resistance is
> roughly 10% of power during the climb, this corresponds to a 9% change
> in power required at a given speed. Analyzing data where useful power
> appears to sag by 9% might suggest that I started too fast, and should
> have saved more at the beginning, when in reality, my pace was optimal.

Go here and plug in your numbers :
http://analyticcycling.com/ForcesPower_Page.html

For simplicity, use the default values except for those you have
hypothesized
(5% grade .005/.01 Crr)

The required power output comes out to 402.1 watts and 431.5, about a
7% difference.
Indeed, using extreme Crr's of 0 /.01 the power figures are 372.7 and
431.5 respectively, an increase of only 15.7 % between no rolling
resistance and VERY rough pavement (nothing like the 2 x 12% you
suggest).

I'm not sure what your point was but, yes, if the iBike is correctly
calibrated for one stretch of pavement it will not be EXACTLY correct
for a rougher/smoother stretch.
So if you think you NEED that kind of precision either calibrate for
the pavement you want to be accurate on, learn how to take into account
the quirks of the tool being used, or spend a bunch more money on a
different device.

DR

[Andy Coggan]

Michael Press wrote:
> In article
> om>,
> "ilan" > wrote:
>
> > http://www.nyvelocity.com/content.php?id=889
>
> "Secondly, even though aero drag increases as a square
> of the speed, the power necessary to overcome that drag
> increases as a cube. In other words, drag force will
> quadruple if your speed doubles, but you'll need eight
> times as many watts to overcome it. The reason is that
> you're traveling twice as fast, so you're pushing
> through twice as much air. Given this exponential
> increase, ..."
>
> Bzzzzzzzzzzzt.

I take it that obnoxious noise is meant to signify your displeasure at
description of what is really a power function as an "...exponential
increase..."?

Andy Coggan

[ilan]

Hey, it's cycling guru terminology as in Joe Friel's "grows
exponentially like the square,"
which I've often quoted here.

-ilan

Andy Coggan wrote:
>
>
> I take it that obnoxious noise is meant to signify your displeasure at
> description of what is really a power function as an "...exponential
> increase..."?
>
> Andy Coggan