Re: An Indirect Measure of Tire Rolling Resistance

[Joe Riel]

[email protected] writes:

> On Sun, 19 Nov 2006 23:06:37 GMT, Joe Riel <[email protected]> wrote:
>
> [snip]
>
>>A surprising result of this test is that Cr appeared to have a maximum
>>with a pressure of around 50psi (I expected it to increase with
>>pressure). The total range of the computed Cr was small, from 0.830
>>to 0.846 as the pressure ranged from 40 to 120psi. At 30psi Cr
>>reduced slightly and the noise of the bounce changed, indicating that
>>the rim might be hitting the tire.
>
> [snip]
>
> Dear Joe,
>
> Here's a possible explanation.
>
> In the links below, "lively" is shorthand for bouncier, higher
> coefficient of restitution.

My thought was that the increased impulsive force of the tire at high
pressure may be inducing larger losses in the setup. One of the
reasons for using two wheels was to ensure that the impact with
the ground would be purely normal (perpendicular), which should
minimize the effect of this impulse on the apparatus.

--
Joe Riel

 

[[email protected]]

On Mon, 20 Nov 2006 15:07:02 GMT, Joe Riel <[email protected]> wrote:

[snip]

>I'd like a better way to quickly and repeatedly measure the bounce
>height. The eyeball technique has more error than I'd like.

[snip]

Dear Joe,

That's what stopped me after I fixed an old fork in pivoting pipe last
night and dropped a tire. I couldn't figure out how to see how high
the tire bounced.

Could you aim a digital camera at an upright ruler and use the movie
function?

My low-range camera will do 20 frames-per-second. The manual slyly
mentions that the next model up offers 60 fps.

I didn't even think of movies when I bought the camera and was
surprised when I stumbled over the feature, just as I was surprised
years ago when I realized that my computer CD drive would play a music
CD.

Cheers,

Carl Fogel

 

[Joe Riel]

[email protected] writes:

> On Mon, 20 Nov 2006 15:07:02 GMT, Joe Riel <[email protected]> wrote:
>
> [snip]
>
>>I'd like a better way to quickly and repeatedly measure the bounce
>>height. The eyeball technique has more error than I'd like.
>
> [snip]
>
> Dear Joe,
>
> That's what stopped me after I fixed an old fork in pivoting pipe last
> night and dropped a tire. I couldn't figure out how to see how high
> the tire bounced.
>
> Could you aim a digital camera at an upright ruler and use the movie
> function?
>
> My low-range camera will do 20 frames-per-second. The manual slyly
> mentions that the next model up offers 60 fps.
>
> I didn't even think of movies when I bought the camera and was
> surprised when I stumbled over the feature, just as I was surprised
> years ago when I realized that my computer CD drive would play a music
> CD.

My wife has a digital camera, I'll see if it has such a feature,
however, that seems like complicated (uploading, etc). On the other
hand, it would improve the documentation and have other lab uses.
Interesting, my wife mentioned---while amusing herself at my
undertaking---that, when she was taking a physiology course, they used
video cameras to record and determine the range of motion. Parallax
error would be an issue; I suppose I could calibrate it with a little
work.

My technique was to set a ladder next to the apparatus and clamp a
board that just passed over the top the wheel, near the bounce height
(which fortunately didn't change much with pressure). To the end of
the board I clamped an adjustable height gauge that I kept aligned to
a second site (via a level) attached to the ladder. I then
progressively adjusted these so that the top of the wheel would just
be inline with both sites at the height of its bounce. However, given
the dynamic movement, there is still a subjective component. Better
would be an adjustable stop that is moved into position after the
wheel drops. Ideally it would have an indicator that records the peak
(might take a few drops to get it to the highest point). Probably
there is a simple means to do this that I haven't thought of.

--
Joe Riel

 

[Stephen Greenwood]

Joe Riel wrote:

> My wife has a digital camera, I'll see if it has such a feature,
> however, that seems like complicated (uploading, etc). On the other
> hand, it would improve the documentation and have other lab uses.
> Interesting, my wife mentioned---while amusing herself at my
> undertaking---that, when she was taking a physiology course, they used
> video cameras to record and determine the range of motion. Parallax
> error would be an issue; I suppose I could calibrate it with a little
> work.

I think you could reduce parallax error by aiming the camera at the
pivot point and measuring the angle, rather than directly measuring the
height. Since a camera lens is circular, an angular change in the
subject won't produce parallax, and you can infer the bounce height
from the angle.

I have heard that some photo editing programs have on-screen
protractors that you might use to make the measurement. Making some
kind of fiducial mark on the rotating arm may aid the process.

Stephen Greenwood

 

[Joe Riel]

"Stephen Greenwood" <[email protected]> writes:

> Joe Riel wrote:
>
>> My wife has a digital camera, I'll see if it has such a feature,
>> however, that seems like complicated (uploading, etc). On the other
>> hand, it would improve the documentation and have other lab uses.
>> Interesting, my wife mentioned---while amusing herself at my
>> undertaking---that, when she was taking a physiology course, they used
>> video cameras to record and determine the range of motion. Parallax
>> error would be an issue; I suppose I could calibrate it with a little
>> work.
>
> I think you could reduce parallax error by aiming the camera at the
> pivot point and measuring the angle, rather than directly measuring the
> height. Since a camera lens is circular, an angular change in the
> subject won't produce parallax, and you can infer the bounce height
> from the angle.

Not a bad idea. To reduce the "common-mode" error a deviation can be
measured from the nominal bounce angle, which can be drawn on the
background and the corresponding height (h0) measured accurately.
That is

h(dtheta) ~ h0 + R*cos(theta0)*dtheta

where dtheta is the angle from the nominal bounce. More clever might
be to superimpose the data images against the setup at the nominal
bounce angle, and extract dtheta from the difference, however, that
would take more photographical processing than I might like. I
suppose I could build a sensor to measure the angle directly, however,
my current lab equipment is limited.

--
Joe Riel

 

[[email protected]]

On Mon, 20 Nov 2006 17:43:04 GMT, Joe Riel <[email protected]> wrote:

>[email protected] writes:
>
>> On Mon, 20 Nov 2006 15:07:02 GMT, Joe Riel <[email protected]> wrote:
>>
>> [snip]
>>
>>>I'd like a better way to quickly and repeatedly measure the bounce
>>>height. The eyeball technique has more error than I'd like.
>>
>> [snip]
>>
>> Dear Joe,
>>
>> That's what stopped me after I fixed an old fork in pivoting pipe last
>> night and dropped a tire. I couldn't figure out how to see how high
>> the tire bounced.
>>
>> Could you aim a digital camera at an upright ruler and use the movie
>> function?
>>
>> My low-range camera will do 20 frames-per-second. The manual slyly
>> mentions that the next model up offers 60 fps.
>>
>> I didn't even think of movies when I bought the camera and was
>> surprised when I stumbled over the feature, just as I was surprised
>> years ago when I realized that my computer CD drive would play a music
>> CD.
>
>My wife has a digital camera, I'll see if it has such a feature,
>however, that seems like complicated (uploading, etc). On the other
>hand, it would improve the documentation and have other lab uses.
>Interesting, my wife mentioned---while amusing herself at my
>undertaking---that, when she was taking a physiology course, they used
>video cameras to record and determine the range of motion. Parallax
>error would be an issue; I suppose I could calibrate it with a little
>work.
>
>My technique was to set a ladder next to the apparatus and clamp a
>board that just passed over the top the wheel, near the bounce height
>(which fortunately didn't change much with pressure). To the end of
>the board I clamped an adjustable height gauge that I kept aligned to
>a second site (via a level) attached to the ladder. I then
>progressively adjusted these so that the top of the wheel would just
>be inline with both sites at the height of its bounce. However, given
>the dynamic movement, there is still a subjective component. Better
>would be an adjustable stop that is moved into position after the
>wheel drops. Ideally it would have an indicator that records the peak
>(might take a few drops to get it to the highest point). Probably
>there is a simple means to do this that I haven't thought of.

Dear Joe,

If you're not familiar with them, the uploading from a digital camera
to a computer is surprisingly quick and easy, at least on mine.

Stick mini USB cable into port. Push button to turn camera on.

The download program senses the camera, shows what pictures and movies
are on the camera, waits for me to click on the images that I want,
and downloads them in a few seconds.

If the camera LCD is big enough, you might even be able to see the
bounce without downloading.

Like most tools, the camera repays fooling around it because the more
familiar you are with how it works, the more likely you are to use it.

For example, taking a quick picture or two beats hell out of drawing
crude diagrams of wires or complicated parts that you're about to
disassemble.

But first you have to use the camera enough to be confident that
you'll push the right buttons. Taking two digital pictures is
practically free, and can avoid unexpected mistakes.

Cheers,

Carl Fogel

 

[Joe Riel]

[email protected] writes:

> If you're not familiar with them, the uploading from a digital camera
> to a computer is surprisingly quick and easy, at least on mine.

The pain is that I have a linux box, so I need to first upload the
stuff to my wife's laptop then transfer it my machine, which takes a
bit of hacking.

--
Joe Riel

 

[[email protected]]

On 20 Nov 2006 10:45:02 -0800, "Stephen Greenwood"
<[email protected]> wrote:

>Joe Riel wrote:
>
>> My wife has a digital camera, I'll see if it has such a feature,
>> however, that seems like complicated (uploading, etc). On the other
>> hand, it would improve the documentation and have other lab uses.
>> Interesting, my wife mentioned---while amusing herself at my
>> undertaking---that, when she was taking a physiology course, they used
>> video cameras to record and determine the range of motion. Parallax
>> error would be an issue; I suppose I could calibrate it with a little
>> work.
>
>I think you could reduce parallax error by aiming the camera at the
>pivot point and measuring the angle, rather than directly measuring the
>height. Since a camera lens is circular, an angular change in the
>subject won't produce parallax, and you can infer the bounce height
>from the angle.
>
>I have heard that some photo editing programs have on-screen
>protractors that you might use to make the measurement. Making some
>kind of fiducial mark on the rotating arm may aid the process.
>
>Stephen Greenwood

Dear Stephen,

Here's an onscreen ruler and protractor:

http://www.markus-bader.de/MB-Ruler/

To use the mouse pointer to measure squiggly lines on maps, try
Mouse-O-Meter:

http://www.softlookup.com/display.asp?id=9758

A nice graphics file viewer with screenshot capture and various
graphics file format conversions:

http://www.irfanview.com/

Alas, they're not Linux, so they won't work for Joe.

Cheers,

Carl Fogel

 

[Marcus Coles]

Joe Riel wrote:
> [email protected] writes:
>
>> If you're not familiar with them, the uploading from a digital camera
>> to a computer is surprisingly quick and easy, at least on mine.
>
> The pain is that I have a linux box, so I need to first upload the
> stuff to my wife's laptop then transfer it my machine, which takes a
> bit of hacking.
>

Linux can handle lots of digital cameras these days, check the available
packages.
Many newer ones can be treated simply as a USB drive or memory stick and
then files viewed though one of the many photo viewers or graphic
manipulation programs.

If your Linux distro is fairly recent you might be pleasantly surprised.

As for Carl's suggestion of taking pictures of complicated projects,
that gave me one of those "duh" moments and is going to save me a lot of
confusion, masking tape labels and scrap paper documentation.
Nothing is worse than returning to a self inflicted basket case after
taking some time off. Maybe a picture is worth a thousand expletives.


Marcus






Marcus

 

[Joe Riel]

Marcus Coles <[email protected]> writes:

> Linux can handle lots of digital cameras these days, check the
> available packages.
> Many newer ones can be treated simply as a USB drive or memory stick
> and then files viewed though one of the many photo viewers or graphic
> manipulation programs.
>
> If your Linux distro is fairly recent you might be pleasantly surprised.

I'm pretty sure I need to upgrade to a 2.6 kernel (I have 2.4) to get
the USB 2.0 stuff to work. Maybe that isn't necessary with the
camera. But I need to do that, regardless, so I can make backups.
I'll probably get around to it a few days after my hard disk crashes
. Thanks for the suggestion, I'll see what's available.

--
Joe Riel

 

[[email protected]]

On Mon, 20 Nov 2006 21:12:59 GMT, Joe Riel <[email protected]> wrote:

>Marcus Coles <[email protected]> writes:
>
>> Linux can handle lots of digital cameras these days, check the
>> available packages.
>> Many newer ones can be treated simply as a USB drive or memory stick
>> and then files viewed though one of the many photo viewers or graphic
>> manipulation programs.
>>
>> If your Linux distro is fairly recent you might be pleasantly surprised.
>
>I'm pretty sure I need to upgrade to a 2.6 kernel (I have 2.4) to get
>the USB 2.0 stuff to work. Maybe that isn't necessary with the
>camera. But I need to do that, regardless, so I can make backups.
>I'll probably get around to it a few days after my hard disk crashes
>. Thanks for the suggestion, I'll see what's available.

Dear Joe,

My camera works fine on an ancient pre-2.0 USB device.

It's just downloading files. Maybe for remote camera control you need
more, but I doubt it.

You could plug the camera in and see if anything is recognized.

Cheers,

Carl Fogel

 

[Gary Young]

On Mon, 20 Nov 2006 13:43:41 -0700, carlfogel wrote:

> On 20 Nov 2006 10:45:02 -0800, "Stephen Greenwood"
> <[email protected]> wrote:
>
>>Joe Riel wrote:
>>
>>> My wife has a digital camera, I'll see if it has such a feature,
>>> however, that seems like complicated (uploading, etc). On the other
>>> hand, it would improve the documentation and have other lab uses.
>>> Interesting, my wife mentioned---while amusing herself at my
>>> undertaking---that, when she was taking a physiology course, they used
>>> video cameras to record and determine the range of motion. Parallax
>>> error would be an issue; I suppose I could calibrate it with a little
>>> work.
>>
>>I think you could reduce parallax error by aiming the camera at the
>>pivot point and measuring the angle, rather than directly measuring the
>>height. Since a camera lens is circular, an angular change in the
>>subject won't produce parallax, and you can infer the bounce height
>>from the angle.
>>
>>I have heard that some photo editing programs have on-screen
>>protractors that you might use to make the measurement. Making some
>>kind of fiducial mark on the rotating arm may aid the process.
>>
>>Stephen Greenwood
>
> Dear Stephen,
>
> Here's an onscreen ruler and protractor:
>
> http://www.markus-bader.de/MB-Ruler/
>
> To use the mouse pointer to measure squiggly lines on maps, try
> Mouse-O-Meter:
>
> http://www.softlookup.com/display.asp?id=9758
>
> A nice graphics file viewer with screenshot capture and various
> graphics file format conversions:
>
> http://www.irfanview.com/
>
> Alas, they're not Linux, so they won't work for Joe.
>
Kruler is a linux-compatible onscreen ruler. Also, you could save
screenshots as image files and view them in "The Gimp," which has an
onscreen ruler and protractor.

 

[[email protected]]

On Mon, 20 Nov 2006 23:11:15 -0600, Gary Young <[email protected]>
wrote:

>On Mon, 20 Nov 2006 13:43:41 -0700, carlfogel wrote:
>
>> On 20 Nov 2006 10:45:02 -0800, "Stephen Greenwood"
>> <[email protected]> wrote:
>>
>>>Joe Riel wrote:
>>>
>>>> My wife has a digital camera, I'll see if it has such a feature,
>>>> however, that seems like complicated (uploading, etc). On the other
>>>> hand, it would improve the documentation and have other lab uses.
>>>> Interesting, my wife mentioned---while amusing herself at my
>>>> undertaking---that, when she was taking a physiology course, they used
>>>> video cameras to record and determine the range of motion. Parallax
>>>> error would be an issue; I suppose I could calibrate it with a little
>>>> work.
>>>
>>>I think you could reduce parallax error by aiming the camera at the
>>>pivot point and measuring the angle, rather than directly measuring the
>>>height. Since a camera lens is circular, an angular change in the
>>>subject won't produce parallax, and you can infer the bounce height
>>>from the angle.
>>>
>>>I have heard that some photo editing programs have on-screen
>>>protractors that you might use to make the measurement. Making some
>>>kind of fiducial mark on the rotating arm may aid the process.
>>>
>>>Stephen Greenwood
>>
>> Dear Stephen,
>>
>> Here's an onscreen ruler and protractor:
>>
>> http://www.markus-bader.de/MB-Ruler/
>>
>> To use the mouse pointer to measure squiggly lines on maps, try
>> Mouse-O-Meter:
>>
>> http://www.softlookup.com/display.asp?id=9758
>>
>> A nice graphics file viewer with screenshot capture and various
>> graphics file format conversions:
>>
>> http://www.irfanview.com/
>>
>> Alas, they're not Linux, so they won't work for Joe.
>>
>Kruler is a linux-compatible onscreen ruler. Also, you could save
>screenshots as image files and view them in "The Gimp," which has an
>onscreen ruler and protractor.

Dear Gary,

Alas, KRuler seems to be a ruler only, with no protractor to measure
the angle that Stephen had in mind for Joe:

"KRuler is a very simple application, with only one aim in life. To
measure distances on your screen."

http://rpmfind.net/linux/rpm2html/search.php?query=kdegraphics-kruler

And it's a bit hard to tell which package Joe would need to download
and install.

My hopeful google for kprotractor came up empty. I saw a linux
protractor mentioned, but I can't find the program.

Cheers,

Carl Fogel

 

[Joe Riel]

[email protected] writes:

> measure distances on your screen."
>
> http://rpmfind.net/linux/rpm2html/search.php?query=kdegraphics-kruler
>
> And it's a bit hard to tell which package Joe would need to download
> and install.

I have a debian system and the package I need is relatively easy to
determine (kruler). However, I see that it depends on the cupsys
printing system, which I refuse to install 'cause it doesn't work with
my setup and I end up having to reinstall and configure lprng which
takes half a day (mostly remembering which configuration files need to
be tweaked to work with my printer).

Anyhow, I devised a simpler way to make the measurements without
dealing with the annoyance of video editing.

--
Joe Riel

 

[[email protected]]

On Tue, 21 Nov 2006 06:32:06 GMT, Joe Riel <[email protected]> wrote:

>[email protected] writes:
>
>> measure distances on your screen."
>>
>> http://rpmfind.net/linux/rpm2html/search.php?query=kdegraphics-kruler
>>
>> And it's a bit hard to tell which package Joe would need to download
>> and install.
>
>I have a debian system and the package I need is relatively easy to
>determine (kruler). However, I see that it depends on the cupsys
>printing system, which I refuse to install 'cause it doesn't work with
>my setup and I end up having to reinstall and configure lprng which
>takes half a day (mostly remembering which configuration files need to
>be tweaked to work with my printer).
>
>Anyhow, I devised a simpler way to make the measurements without
>dealing with the annoyance of video editing.

Dear Joe,

Ah, sweet technology! Our willing and helpful servant!

With apologies to Bierce and his definition of Newtonian . . .

GATESIAN, adj. Pertaining to a philosophy of the universe invented by
Gates, who discovered that a computer will crash, but was unable to
say why. His successors and disciples have advanced so far as to be
able to say when.

Cheers,

Carl Fogel

 

[Pat Lamb]

Joe Riel wrote:
> [email protected] writes:
>
>> If you're not familiar with them, the uploading from a digital camera
>> to a computer is surprisingly quick and easy, at least on mine.
>
> The pain is that I have a linux box, so I need to first upload the
> stuff to my wife's laptop then transfer it my machine, which takes a
> bit of hacking.

IIRC, 2.4 had USB support if it was compiled into the kernel. You may
need to manually mount the USB device (I actually preferred this, since
I didn't have to guess what different directory every single new USB
device ended up mounted in).

OTOH, cygwin and ssh rule. scp or sftp to your heart's content!

Pat