Cayeye tire size chart for cycle computers

Discussion in 'Cycling Equipment' started by Rik O'Shea, Jan 12, 2006.

  1. Rik O'Shea

    Rik O'Shea Guest

    I just inspected the Cateye tire size chart (well it was a wet night on
    Monday) - this is the chart that relates tyre size to tire
    circumference. I noticed what I think is a discrepency.

    Tire size => Circumerference (mm)
    700 x 28C => 2136
    700 x 30C => 2170
    700 x 32C => 2155

    How is it that when you go to the wider 32C tire that the circumference
    reduces ? I think the "error" maybe in the 30C value as there is quite
    a big jump from 28C to 30C (34 mm) which is not consistent with the
    other increments in the chart.

    Here is the section for 27" tires:

    Tire size => Circumerference (mm)
    27 x 1 => 2145
    27 x 1-1/8 => 2155
    27 x 1-1/4 => 2161
    27 x 1-3/8 => 2169

    Notice that the mm increments are smaller and Cateye seem to directly
    equate a 700 x 32C tire with a 27 x 1-1/8.

    I noticed that on Cateye's on-line manuals they now express
    circumference in cm (rounding to 3 digits) rather than mm but this
    discrepency is still there.

    PS before people go off on tangents about "actual" circumference is
    based on rider weight, inflation pressure, blah, blah - yes correct but
    the concern is the "reference" circumference in the the chart.
     
    Tags:


  2. Dan Burkhart

    Dan Burkhart New Member

    Joined:
    Nov 27, 2003
    Messages:
    333
    Likes Received:
    0

    I for one, never bother with manufacturers charts. Too much variation among various tire makers actual circumference. Mark, roll, and measure is the only reliable method IMHO!
    Dan
     
  3. JeffWills

    JeffWills Guest

    Rik O'Shea wrote:
    > I just inspected the Cateye tire size chart (well it was a wet night on
    > Monday) - this is the chart that relates tyre size to tire
    > circumference. I noticed what I think is a discrepency.
    >
    > Tire size => Circumerference (mm)
    > 700 x 28C => 2136
    > 700 x 30C => 2170
    > 700 x 32C => 2155
    >
    > How is it that when you go to the wider 32C tire that the circumference
    > reduces ? I think the "error" maybe in the 30C value as there is quite
    > a big jump from 28C to 30C (34 mm) which is not consistent with the
    > other increments in the chart.
    >


    The only "700 x 30C" tire that I'm familiar with is the IRC Tandem EX,
    which has a tall, ellipitical cross-section. This could give it a
    larger circumfrence than a rounder 700 x 32C tire. It may be that
    they're reporting "real" measurements.

    Or it could be a typo.

    Jeff
     
  4. anonymous writes:

    > I just inspected the Cateye tire size chart (well it was a wet night
    > on Monday) - this is the chart that relates tyre size to tire
    > circumference. I noticed what I think is a discrepancy.


    > Tire size => Circumference (mm)
    > 700 x 28C => 2136
    > 700 x 30C => 2170
    > 700 x 32C => 2155


    > How is it that when you go to the wider 32C tire that the
    > circumference reduces ? I think the "error" maybe in the 30C value
    > as there is quite a big jump from 28C to 30C (34 mm) which is not
    > consistent with the other increments in the chart.


    > Here is the section for 27" tires:


    > Tire size => Circumference (mm)
    > 27 x 1 => 2145
    > 27 x 1-1/8 => 2155
    > 27 x 1-1/4 => 2161
    > 27 x 1-3/8 => 2169


    > Notice that the mm increments are smaller and Cateye seem to
    > directly equate a 700 x 32C tire with a 27 x 1-1/8.


    > I noticed that on Cateye's on-line manuals they now express
    > circumference in cm (rounding to 3 digits) rather than mm but this
    > discrepancy is still there.


    > PS before people go off on tangents about "actual" circumference is
    > based on rider weight, inflation pressure, blah, blah - yes correct
    > but the concern is the "reference" circumference in the the chart.


    So if you know all this, why do you care what these charts say when in
    reality, for people who want to have accurate readings measure the
    roll-out distance and enter it in the instrument?

    If you missed it, it's done as follows. With the tire inflated to its
    preferred pressure and the rider on the bicycle in the preferred
    position, place the front wheel with the valve stem directly over a
    line on the floor (floor tile or the like) and roll forward until the
    stem is again in its vertical position. Measure the distance in the
    units used by the Cyclometer and enter that number.

    Jobst Brandt
     
  5. Diablo Scott

    Diablo Scott Guest

    [email protected] wrote:

    >anonymous writes:
    >
    >
    >
    >>I just inspected the Cateye tire size chart (well it was a wet night
    >>on Monday) - this is the chart that relates tyre size to tire
    >>circumference. I noticed what I think is a discrepancy.
    >>
    >>

    >
    >
    >
    >>Tire size => Circumference (mm)
    >>700 x 28C => 2136
    >>700 x 30C => 2170
    >>700 x 32C => 2155
    >>
    >>

    >
    >
    >
    >>How is it that when you go to the wider 32C tire that the
    >>circumference reduces ? I think the "error" maybe in the 30C value
    >>as there is quite a big jump from 28C to 30C (34 mm) which is not
    >>consistent with the other increments in the chart.
    >>
    >>

    >
    >
    >
    >>Here is the section for 27" tires:
    >>
    >>

    >
    >
    >
    >>Tire size => Circumference (mm)
    >>27 x 1 => 2145
    >>27 x 1-1/8 => 2155
    >>27 x 1-1/4 => 2161
    >>27 x 1-3/8 => 2169
    >>
    >>

    >
    >
    >
    >>Notice that the mm increments are smaller and Cateye seem to
    >>directly equate a 700 x 32C tire with a 27 x 1-1/8.
    >>
    >>

    >
    >
    >
    >>I noticed that on Cateye's on-line manuals they now express
    >>circumference in cm (rounding to 3 digits) rather than mm but this
    >>discrepancy is still there.
    >>
    >>

    >
    >
    >
    >>PS before people go off on tangents about "actual" circumference is
    >>based on rider weight, inflation pressure, blah, blah - yes correct
    >>but the concern is the "reference" circumference in the the chart.
    >>
    >>

    >
    >So if you know all this, why do you care what these charts say when in
    >reality, for people who want to have accurate readings measure the
    >roll-out distance and enter it in the instrument?
    >
    >If you missed it, it's done as follows. With the tire inflated to its
    >preferred pressure and the rider on the bicycle in the preferred
    >position, place the front wheel with the valve stem directly over a
    >line on the floor (floor tile or the like) and roll forward until the
    >stem is again in its vertical position. Measure the distance in the
    >units used by the Cyclometer and enter that number.
    >
    >Jobst Brandt
    >
    >

    I find it difficult to get an accurate measurement using a rollout. The
    minimum difference on my computer is 0.5cm. I know from riding surveyed
    routes and measured miles that the correct setting is 211.0 (catalogue
    data says 210.5) but I've never been able to actually produce that value
    from a rollout. I've used your valve stem method and I've wetted my
    tire and measured the space between the wet spots. A 1cm measurement
    error is about 1/2 % output error. The perfectionist in me wants to
    make the rollout method work and understand what I'm doing wrong, but
    the pragmatist already knows the right answer.

    I suggest riding a measured course for best accuracy - or you can
    measure your own course with GPS or even a map making program. Use the
    catalog value, ride the course, make a correction. This has the added
    advantage of accounting for real world conditions of not riding
    perfectly straight lines; after all, we're not really after perfect
    accuracy so much as we're after perfect agreement with correct signage
    and route maps, right?
     
  6. Rik O'Shea wrote:

    > I just inspected the Cateye tire size chart (well it was a wet night on
    > Monday) - this is the chart that relates tyre size to tire
    > circumference. I noticed what I think is a discrepency.
    >
    > Tire size => Circumerference (mm)
    > 700 x 28C => 2136
    > 700 x 30C => 2170
    > 700 x 32C => 2155
    >
    > How is it that when you go to the wider 32C tire that the circumference
    > reduces ? I think the "error" maybe in the 30C value as there is quite
    > a big jump from 28C to 30C (34 mm) which is not consistent with the
    > other increments in the chart.
    >

    Most likely a typo, but it's possible that they measured a nominal 30 mm
    tire that had an unusually thick tread.

    See: http://sheldonbrown.com/cyclecomputer-calibration

    Sheldon "Numbers" Brown
    +--------------------------------------------------------+
    | As far as the laws of mathematics refer to reality, |
    | they are not certain; and as far as they are certain, |
    | they do not refer to reality. --Albert Einstein |
    +--------------------------------------------------------+
    Harris Cyclery, West Newton, Massachusetts
    Phone 617-244-9772 FAX 617-244-1041
    http://harriscyclery.com
    Hard-to-find parts shipped Worldwide
    http://captainbike.com http://sheldonbrown.com
     
  7. Diablo Scott <[email protected]> writes:

    >>> PS before people go off on tangents about "actual" circumference
    >>> is based on rider weight, inflation pressure, blah, blah - yes
    >>> correct but the concern is the "reference" circumference in the
    >>> the chart.


    >> So if you know all this, why do you care what these charts say when
    >> in reality, for people who want to have accurate readings measure
    >> the roll-out distance and enter it in the instrument?


    >> If you missed it, it's done as follows. With the tire inflated to
    >> its preferred pressure and the rider on the bicycle in the
    >> preferred position, place the front wheel with the valve stem
    >> directly over a line on the floor (floor tile or the like) and roll
    >> forward until the stem is again in its vertical position. Measure
    >> the distance in the units used by the Cyclometer and enter that
    >> number.


    > I find it difficult to get an accurate measurement using a rollout.
    > The minimum difference on my computer is 0.5cm. I know from riding
    > surveyed routes and measured miles that the correct setting is 211.0
    > (catalogue data says 210.5) but I've never been able to actually
    > produce that value from a rollout. I've used your valve stem method
    > and I've wetted my tire and measured the space between the wet
    > spots. A 1cm measurement error is about 1/2 % output error. The
    > perfectionist in me wants to make the rollout method work and
    > understand what I'm doing wrong, but the pragmatist already knows
    > the right answer.


    I can measure that distance to +-1.5mm which is better than an
    estimate from a table that knows neither the inflation pressure nor
    the load. Where do you have difficulty? My Cyclometer accepts mm
    and my wheel is 2096mm.

    > I suggest riding a measured course for best accuracy - or you can
    > measure your own course with GPS or even a map making program. Use
    > the catalog value, ride the course, make a correction. This has the
    > added advantage of accounting for real world conditions of not
    > riding perfectly straight lines; after all, we're not really after
    > perfect accuracy so much as we're after perfect agreement with
    > correct signage and route maps, right?


    That is definitely untrue because that depends on how you take
    corners, inside or out. The only one that might work is a straight
    line course. For that we have California HWY1 with aircraft speed
    check markers that are placed at one mile intervals to great accuracy
    from Pigeon Point to Santa Cruz. I haven't checked that but it's
    an idea I should follow up.

    Jobst Brandt
     
  8. Rik O'Shea

    Rik O'Shea Guest

    Yes, I've always used the tire rollout method when calibrating my cycle
    computer but today out of curiosity I took a straw poll among some of
    the commuter freds in work who use a cycle computer and guess what, 4/4
    used the chart that came with their cycle computer !
     
  9. Diablo Scott

    Diablo Scott Guest

    [email protected] wrote:

    >Diablo Scott <[email protected]> writes:
    >
    >
    >
    >>I find it difficult to get an accurate measurement using a rollout.
    >>The minimum difference on my computer is 0.5cm. I know from riding
    >>surveyed routes and measured miles that the correct setting is 211.0
    >>(catalogue data says 210.5) but I've never been able to actually
    >>produce that value from a rollout. I've used your valve stem method
    >>and I've wetted my tire and measured the space between the wet
    >>spots. A 1cm measurement error is about 1/2 % output error. The
    >>perfectionist in me wants to make the rollout method work and
    >>understand what I'm doing wrong, but the pragmatist already knows
    >>the right answer.
    >>
    >>

    >
    >I can measure that distance to +-1.5mm which is better than an
    >estimate from a table that knows neither the inflation pressure nor
    >the load. Where do you have difficulty? My Cyclometer accepts mm
    >and my wheel is 2096mm.
    >
    >
    >

    Sitting on the bike in riding position and rolling out a straight line
    seems to be the difficult part. A 1.5mm difference between your wheel's
    true path and the point-to-point distance is pretty impressive, if you
    really did it. The fact that you can measure between the two points to
    such an accuracy is not the same thing of course.

    >>I suggest riding a measured course for best accuracy - or you can
    >>measure your own course with GPS or even a map making program. Use
    >>the catalog value, ride the course, make a correction. This has the
    >>added advantage of accounting for real world conditions of not
    >>riding perfectly straight lines; after all, we're not really after
    >>perfect accuracy so much as we're after perfect agreement with
    >>correct signage and route maps, right?
    >>
    >>

    > <>
    > That is definitely untrue because that depends on how you take
    > corners, inside or out. The only one that might work is a straightline
    > course.


    Yes but these are real world conditions and we generally have the same
    number of left and right turns. Maybe I've introduced an unintentional
    fudge factor to my calibration that accounts for the occaisional sprint
    or stretch or pothole dodging maneuver, but I don't care if my distance
    function perfectly tracks my rolling contact patch; I want it to
    perfectly match the route sheet for the century I'm riding or the map I
    downloaded.

    > <>For that we have California HWY1 with aircraft speed
    > check markers that are placed at one mile intervals to great accuracy
    > from Pigeon Point to Santa Cruz. I haven't checked that but it's
    > an idea I should follow up.
    >
    > Jobst Brandt


    Yes, you might be surprised. And a straightline course would be a
    better control than one with lots of turns. Ride normally, not like
    you're trying to stay on a wire.
     
  10. Bob Flumere

    Bob Flumere Guest

    On Thu, 12 Jan 2006 16:11:13 -0800, Diablo Scott
    <[email protected]> wrote:

    >[email protected] wrote:
    >
    >>Diablo Scott <[email protected]> writes:
    >>
    >>
    >>
    >>>I find it difficult to get an accurate measurement using a rollout.
    >>>The minimum difference on my computer is 0.5cm. I know from riding
    >>>surveyed routes and measured miles that the correct setting is 211.0
    >>>(catalogue data says 210.5) but I've never been able to actually
    >>>produce that value from a rollout. I've used your valve stem method
    >>>and I've wetted my tire and measured the space between the wet
    >>>spots. A 1cm measurement error is about 1/2 % output error. The
    >>>perfectionist in me wants to make the rollout method work and
    >>>understand what I'm doing wrong, but the pragmatist already knows
    >>>the right answer.


    >>

    >Sitting on the bike in riding position and rolling out a straight line
    >seems to be the difficult part. A 1.5mm difference between your wheel's
    >true path and the point-to-point distance is pretty impressive, if you
    >really did it. The fact that you can measure between the two points to
    >such an accuracy is not the same thing of course.
    >
    >>>I suggest riding a measured course for best accuracy - or you can
    >>>measure your own course with GPS or even a map making program. Use
    >>>the catalog value, ride the course, make a correction. This has the
    >>>added advantage of accounting for real world conditions of not
    >>>riding perfectly straight lines; after all, we're not really after
    >>>perfect accuracy so much as we're after perfect agreement with
    >>>correct signage and route maps, right?
    >>>
    >>>

    >> <>

    >
    >Yes, you might be surprised. And a straightline course would be a
    >better control than one with lots of turns. Ride normally, not like
    >you're trying to stay on a wire.



    If you would like to improve the accuracy of a "Roll Out", do a
    measured 100 feet or so..
    I have a long side walk in front of my house and I have measured out
    100' with a tape and marked it on the pavement. I use it all the time
    for speedo calibration..
    Seems very accurate, and yes, the cal factors vary quite a bit from
    those in the "Emanuel". <G>
    Just check your tire pressure, ride slowly and estimate the change at
    the end. It is very repeatable.

    Bob F.
     
  11. Rik O'Shea writes:

    > Yes, I've always used the tire rollout method when calibrating my
    > cycle computer but today out of curiosity I took a straw poll among
    > some of the commuter friends at work who use a cycle computer and
    > guess what, 4/4 used the chart that came with their cycle computer!


    Yes? How large is your sample, and what conclusion do you expect
    others to draw from that?

    Jobst Brandt
     
  12. Diablo Scott writes:

    >>> I find it difficult to get an accurate measurement using a
    >>> rollout. The minimum difference on my computer is 0.5cm. I know
    >>> from riding surveyed routes and measured miles that the correct
    >>> setting is 211.0 (catalogue data says 210.5) but I've never been
    >>> able to actually produce that value from a rollout. I've used
    >>> your valve stem method and I've wetted my tire and measured the
    >>> space between the wet spots. A 1cm measurement error is about 1/2
    >>> % output error. The perfectionist in me wants to make the rollout
    >>> method work and understand what I'm doing wrong, but the
    >>> pragmatist already knows the right answer.


    >> I can measure that distance to +-1.5mm which is better than an
    >> estimate from a table that knows neither the inflation pressure nor
    >> the load. Where do you have difficulty? My Cyclometer accepts mm
    >> and my wheel is 2096mm.


    > Sitting on the bike in riding position and rolling out a straight
    > line seems to be the difficult part. A 1.5mm difference between
    > your wheel's true path and the point-to-point distance is pretty
    > impressive, if you really did it. The fact that you can measure
    > between the two points to such an accuracy is not the same thing of
    > course.


    I think you can summon the skill to roll straight ahead for one wheel
    revolution, however, indoors you can use a hand on the wall if you
    like. This is not a good excuse for deferring to a probably
    inappropriate chart.

    >>> I suggest riding a measured course for best accuracy - or you can
    >>> measure your own course with GPS or even a map making program.
    >>> Use the catalog value, ride the course, make a correction. This
    >>> has the added advantage of accounting for real world conditions of
    >>> not riding perfectly straight lines; after all, we're not really
    >>> after perfect accuracy so much as we're after perfect agreement
    >>> with correct signage and route maps, right?


    >> That is definitely untrue because that depends on how you take
    >> corners, inside or out. The only one that might work is a
    >> straight line course.


    > Yes but these are real world conditions and we generally have the
    > same number of left and right turns.


    Are you suggesting that in the "real world" distances traveled are
    different from precise measurement? I don't understand your point.

    > Maybe I've introduced an unintentional fudge factor to my
    > calibration that accounts for the occaisional sprint or stretch or
    > pothole dodging maneuver, but I don't care if my distance function
    > perfectly tracks my rolling contact patch; I want it to perfectly
    > match the route sheet for the century I'm riding or the map I
    > downloaded.


    I see. you don't care if the rider who measured the course was off by
    an amount, you want some arbitrary correlation counter. Then you had
    best get your calibration number from the course measurer rather than
    entering an accurate one.

    >> For that we have California HWY1 with aircraft speed check markers
    >> that are placed at one mile intervals to great accuracy from
    >> Pigeon Point to Santa Cruz. I haven't checked that but it's an
    >> idea I should follow up.


    > Yes, you might be surprised. And a straight line course would be a
    > better control than one with lots of turns. Ride normally, not like
    > you're trying to stay on a wire.


    Are you suggesting that in a straight ahead level mile I cannot
    control the line enough to get an accurate reading? I suspect you
    haven't tried riding on such a route, especially with a good tailwind
    that often prevails on the Coast Highway in that direction.

    Jobst Brandt
     
  13. Diablo Scott

    Diablo Scott Guest

    [email protected] wrote:

    >
    >
    > Are you suggesting that in a straight ahead level mile I cannot
    > control the line enough to get an accurate reading? I suspect you
    > haven't tried riding on such a route, especially with a good tailwind
    > that often prevails on the Coast Highway in that direction.
    >
    > Jobst Brandt


    I'm suggesting that in real riding situations, the total distance
    travelled as measured along the actual path of the tire's contact patch
    (which is what you'd have with a "perfectly" calibrated computer),
    doesn't necessarily correspond to the actual sum of straight line
    distances between boundry points (which I contend is what you {I} really
    want to know). No comment at all on your ability to ride a line over a
    flat straight mile with a tail wind, other than perhaps it's better than
    my own.

    We're arguing small errors here, but I did several rollouts and entered
    the average (small deltas) and I was surprised at how far off I was when
    I rode a surveyed and stationed road course (neither straight nor
    tailwindy). Maybe you will be surprised too. Looks like crappy weather
    on Saturday - do it on Sunday.

    --
    Check out my bike blog!
    http://diabloscott.blogspot.com
     
  14. Andrew Lee

    Andrew Lee Guest

    Diablo Scott wrote:
    > [email protected] wrote:
    >>Diablo Scott <[email protected]> writes:
    >>I can measure that distance to +-1.5mm which is better than an
    >>estimate from a table that knows neither the inflation pressure nor
    >>the load. Where do you have difficulty? My Cyclometer accepts mm
    >>and my wheel is 2096mm.
    >>
    >>

    > Sitting on the bike in riding position and rolling out a straight line
    > seems to be the difficult part. A 1.5mm difference between your wheel's
    > true path and the point-to-point distance is pretty impressive, if you
    > really did it. The fact that you can measure between the two points to
    > such an accuracy is not the same thing of course.


    I have also always been able to repeat rollout measurements to within about
    1 mm or 2. I put a small piece of tape across the tread on the tire and
    mark one edge of it to use as a reference mark. That makes it easy to line
    up the tire with the starting line, and to pin point the end of the full
    revolution.

    Andrew Lee
     
  15. Paul Kopit

    Paul Kopit Guest

    On 12 Jan 2006 15:47:02 -0800, "Rik O'Shea" <[email protected]>
    wrote:

    >Yes, I've always used the tire rollout method when calibrating my cycle
    >computer but today out of curiosity I took a straw poll among some of
    >the commuter freds in work who use a cycle computer and guess what, 4/4
    >used the chart that came with their cycle computer !


    When measuring rollout, I usually measure the distance of 2
    revolutions w/o riding the bike. For a 700x23 tire, I usually select
    2090 mm and then ride to a spot that is 11.88 mi from my house. That
    spot was verified with a GPS several times. I then readjust the
    'puter.

    If you are super compulsive, put 2 magnets on when doing the test ride
    and divide by 2.

    I really have no circumstances when I need such accuracy and
    frequently ride w/o a computer.
     
  16. Diablo Scott

    Diablo Scott Guest

    Paul Kopit wrote:

    >
    >When measuring rollout, I usually measure the distance of 2
    >revolutions w/o riding the bike. For a 700x23 tire, I usually select
    >2090 mm and then ride to a spot that is 11.88 mi from my house. That
    >spot was verified with a GPS several times. I then readjust the
    >'puter.
    >
    >If you are super compulsive, put 2 magnets on when doing the test ride
    >and divide by 2.
    >
    >I really have no circumstances when I need such accuracy and
    >frequently ride w/o a computer.
    >
    >

    You seem to contradict yourself several times in that post. But your
    key point is that you readjust your computer based on the GPS measured
    route; signifying that you want your odometer to agree with a mapped
    distance rather than having it be perfectly calibrated to your wheel
    circumference. That's exactly my point as well and I've got a similar
    measured course.

    What percent error do you find between the rollout calibration and the
    corrected number from GPS?

    There are three sources for error. 1 - Mechanical error caused by
    problems with the sensor/magnet or the computer itself. 2 - Calibration
    error from entering an incorrect wheel circ number. and 3 - Deviation
    error from not riding in a straight line.

    The third is the most interesting to me. In Jobst's proposed test he'll
    ride a measured mile on his bike with the computer that has perfect
    calibration. I asked him not to concentrate too much on maintaining a
    perfectly straight line but to ride "normally". In normal riding we
    swerve around potholes and debris, we take different lines around
    corners, we vary our position relative to the edge of the road, we get
    out of the saddle and stretch or sprint - there are a whole host of
    things we do that make the path of our front wheels different (longer)
    than the straight line distance or whatever path GPS uses in its
    measurement. Since Jobst is certain of his calibration data, at the end
    of his test he'll know exactly how much he's deviated from a straight
    line by how much his computer differs from 1.00 miles.

    The question is which value we really want our computers to read. By
    calibrating to a measured course of 5 miles or so, I have incorporated a
    fudge factor that accounts for my riding style and conditions over that
    course. By calibrating based on his roll-out, Jobst is recording the
    true distance his front wheel has traveled rather than the true distance
    between those two points. I'm not sure why he wants to know the former
    rather than the latter except maybe he just doesn't like fudge.
     
  17. Diablo Scott writes:

    > The third is the most interesting to me. In Jobst's proposed test
    > he'll ride a measured mile on his bike with the computer that has
    > perfect calibration. I asked him not to concentrate too much on
    > maintaining a perfectly straight line but to ride "normally". In
    > normal riding we swerve around potholes and debris, we take
    > different lines around corners, we vary our position relative to the
    > edge of the road, we get out of the saddle and stretch or sprint -
    > there are a whole host of things we do that make the path of our
    > front wheels different (longer) than the straight line distance or
    > whatever path GPS uses in its measurement. Since Jobst is certain
    > of his calibration data, at the end of his test he'll know exactly
    > how much he's deviated from a straight line by how much his computer
    > differs from 1.00 miles.


    I don't understand how you think this is preferable. If your
    Cyclometer is based on vagaries of riding ability, what then is a mile
    or kilometer? How is it measured and why do you want to redefine
    distance?

    If you want to record bicycling effort, you should include the gear
    ratio so that the distance the feet travel is recorded. That would be
    like zig-zagging mileage on a straight course.

    > The question is which value we really want our computers to read.
    > By calibrating to a measured course of 5 miles or so, I have
    > incorporated a fudge factor that accounts for my riding style and
    > conditions over that course. By calibrating based on his roll-out,
    > Jobst is recording the true distance his front wheel has traveled
    > rather than the true distance between those two points. I'm not
    > sure why he wants to know the former rather than the latter except
    > maybe he just doesn't like fudge.


    Tis is similar to the altimeters that have no hysteresis threshold
    that prevents surface irregularities from entering cumulative
    altitude. They add every bump in the road that falls within the
    resolution of the instrument, as well as atmospheric fluctuations.
    These devices record climbs on essentially level courses between to
    towns at the same elevation with no point being more than 20 feet
    higher or lower than the endpoints... including railway underpasses.

    This increases trip totals and your suggested calibration decreases
    distance traveled. It certainly is not accuracy.

    Jobst Brandt
     
  18. Sandy

    Sandy Guest

    Dans le message de news:[email protected],
    [email protected] <[email protected]> a
    réfléchi, et puis a déclaré :
    > Diablo Scott writes:
    >
    >> The third is the most interesting to me. In Jobst's proposed test
    >> he'll ride a measured mile on his bike with the computer that has
    >> perfect calibration. I asked him not to concentrate too much on
    >> maintaining a perfectly straight line but to ride "normally". In
    >> normal riding we swerve around potholes and debris, we take
    >> different lines around corners, we vary our position relative to the
    >> edge of the road, we get out of the saddle and stretch or sprint -
    >> there are a whole host of things we do that make the path of our
    >> front wheels different (longer) than the straight line distance or
    >> whatever path GPS uses in its measurement. Since Jobst is certain
    >> of his calibration data, at the end of his test he'll know exactly
    >> how much he's deviated from a straight line by how much his computer
    >> differs from 1.00 miles.

    >
    > I don't understand how you think this is preferable. If your
    > Cyclometer is based on vagaries of riding ability, what then is a mile
    > or kilometer? How is it measured and why do you want to redefine
    > distance?
    >
    > If you want to record bicycling effort, you should include the gear
    > ratio so that the distance the feet travel is recorded. That would be
    > like zig-zagging mileage on a straight course.
    >
    >> The question is which value we really want our computers to read.
    >> By calibrating to a measured course of 5 miles or so, I have
    >> incorporated a fudge factor that accounts for my riding style and
    >> conditions over that course. By calibrating based on his roll-out,
    >> Jobst is recording the true distance his front wheel has traveled
    >> rather than the true distance between those two points. I'm not
    >> sure why he wants to know the former rather than the latter except
    >> maybe he just doesn't like fudge.

    >
    > Tis is similar to the altimeters that have no hysteresis threshold
    > that prevents surface irregularities from entering cumulative
    > altitude. They add every bump in the road that falls within the
    > resolution of the instrument, as well as atmospheric fluctuations.
    > These devices record climbs on essentially level courses between to
    > towns at the same elevation with no point being more than 20 feet
    > higher or lower than the endpoints... including railway underpasses.
    >
    > This increases trip totals and your suggested calibration decreases
    > distance traveled. It certainly is not accuracy.
    >
    > Jobst Brandt


    If you're off by 1%, then over a year's 10 000 miles, you over or understate
    your total by 100 miles. Who's going to be impressed ? Who needs that
    information ? In most cases, when you have to make a turn, let's say 13
    miles from intersection X, there will be only a single road, and it's
    probably marked, and reasonably close to the measured amount. Who cares
    about a millimetre at that point ? Is there something other than a
    compulsive nature of measuring minutiae that I missed ?
    --
    Sandy
    --
    C'est le contraire du vélo, la bicyclette.
    Une silhouette profilée mauve fluo dévale
    à soixante-dix à l'heure : c'est du vélo.
    Deux lycéennes côte à côte traversent
    un pont à Bruges : c'est de la bicyclette.
    -Delerm, P.
     
  19. Sandy wrote on Friday 13 January 2006 22:08:

    > If you're off by 1%, then over a year's 10 000 miles, you over or
    > understate
    > your total by 100 miles. Who's going to be impressed ? Who needs
    > that
    > information ? In most cases, when you have to make a turn, let's say
    > 13 miles from intersection X, there will be only a single road, and
    > it's
    > probably marked, and reasonably close to the measured amount. Who
    > cares
    > about a millimetre at that point ?


    Well, I like to know how far I've ridden. So I'm with Jobst here - I use
    the measured rollout when setting up my speedo.
    --
    Regards
    Alex
    The From address above is a spam-trap.
    The Reply-To address is valid
     
  20. Sandy wrote on Friday 13 January 2006 23:09:

    > I still don't get it. If you know to the fraction of a mile,
    > absolutely precisely, what did that add to your year of cycling ?
    >

    I reaslly don't care *that* much, but, given what purports to be a
    precision instrunment, it would offend my sense of the rightnes of
    things to mis-calibrate it.

    However, as I said, I like to know how far I've ridden, and how long it
    took me, as a measure of my climb back from nearly dead to health and
    fitness.
    --
    Regards
    Alex
    The From address above is a spam-trap.
    The Reply-To address is valid
     
Loading...
Loading...