Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Effect of oil on inner tube

370 views
Skip to first unread message

carl...@comcast.net

unread,
May 14, 2005, 5:12:02 PM5/14/05
to
Back in January, Joe Riel explained why some riders fear
that petroleum-based oil from lubricated spoke nipples might
damage their butyl inner tubes:

"Note that Isobutylene Isoprene, which is, I believe, the
rubber used in inner tubes, is classified as having poor oil
resistance. Presumably, then, its tensile strength is
reduced by more than 60%; how much more I have no idea."

http://groups-beta.google.com/group/rec.bicycles.tech/browse_thread/thread/ac6ffc7aaf212339/94f2e67d73d5045d?q=carl+oil+rubber&rnum=3&hl=en#94f2e67d73d5045d

(Joe's post is up a little ways from this link.)

Curious, I cut a pair of 2-inch sections out of an inner
tube, set one aside, recklessly dunked the other in a cup of
10w-40, and promptly forgot about the experiment.

Today, I noticed the cup on a garage shelf. A mud wasp had
perished in the oil, presumably investigating its effect on
the rubber.

At first, I thought that a few months of holding its breath
in the oil had not affected the inner tube. It stretched
sideways and lengthwise without breaking.

But when I sat down at the computer to reveal that oil has
no more effect on rubber than garlic on werewolves, I
stretched the piece of inner tube sideways a few more times.

It broke.

More stretching kept breaking it. The smaller the pieces of
oiled rubber were, the easier they broke--it was almost as
satisfying as popping bubble wrap.

Then I tried stretching the dry section of inner tube. It
wouldn't break.

After a while, I got tired and cut the dry piece in half. It
still wouldn't break.

So I cut it in half again. Same result--even the thinner
strip of dry rubber wouldn't break despite repeated
stretching, while pieces of oiled rubber the same width
broke easily with a single pull.

Naturally, I went back to breaking the oiled rubber into
even smaller pieces. Here's a picture of the carnage:

http://home.comcast.net/~carlfogel/download/tube_oil.jpg

The three dry strips that I cut and couldn't break are on
the left (the largest is flattened with two staples). The
fragments of the oiled strip are on the right.

Maybe those guys worrying about oil on their tubes aren't
spouting pure bugle oil.

Carl Fogel

jobst....@stanfordalumni.org

unread,
May 14, 2005, 5:39:30 PM5/14/05
to
Carl Fogel writes:

> Back in January, Joe Riel explained why some riders fear that
> petroleum-based oil from lubricated spoke nipples might damage their
> butyl inner tubes:

> "Note that Isobutylene Isoprene, which is, I believe, the rubber
> used in inner tubes, is classified as having poor oil
> resistance. Presumably, then, its tensile strength is reduced by
> more than 60%; how much more I have no idea."

http://tinyurl.com/e3v5j

> (Joe's post is up a little ways from this link.)

> Curious, I cut a pair of 2-inch sections out of an inner tube, set
> one aside, recklessly dunked the other in a cup of 10w-40, and
> promptly forgot about the experiment.

> Today, I noticed the cup on a garage shelf. A mud wasp had perished
> in the oil, presumably investigating its effect on the rubber.

> At first, I thought that a few months of holding its breath in the
> oil had not affected the inner tube. It stretched sideways and
> lengthwise without breaking.

> But when I sat down at the computer to reveal that oil has no more
> effect on rubber than garlic on werewolves, I stretched the piece of
> inner tube sideways a few more times.

> It broke.

I think you should consider that a properly sized tube doesn't suffer
any stress in use other than preventing air from diffusing through its
walls that are supported by the rim and tire casing against stretch.

Besides, the amount of oil of interest, that of putting a drop of oil
at the spoke nipple to eyelet interface is minuscule and cannot reach
the tube if a rim strip is being used. In addition, centrifugal
action makes it tend to the other direction when the bicycle is in
use. This is a classic exaggeration in bicycle maintenance and is not
simulated by inner tube immersion.

> More stretching kept breaking it. The smaller the pieces of oiled
> rubber were, the easier they broke--it was almost as satisfying as
> popping bubble wrap.

I don't understand the "smaller pieces" routine. I find pulling on a
piece of rubber easier if it is larger. That way I can get a grip on
it.

> Then I tried stretching the dry section of inner tube. It
> wouldn't break.

> After a while, I got tired and cut the dry piece in half. It still
> wouldn't break.

> So I cut it in half again. Same result--even the thinner strip of
> dry rubber wouldn't break despite repeated stretching, while pieces
> of oiled rubber the same width broke easily with a single pull.

> Naturally, I went back to breaking the oiled rubber into even
> smaller pieces. Here's a picture of the carnage:

http://home.comcast.net/~carlfogel/download/tube_oil.jpg

You can do better than that. Those fuzzy images don't show anything a
few snips of dark paper wouldn't do as well. Let's get some focus.

> The three dry strips that I cut and couldn't break are on the left
> (the largest is flattened with two staples). The fragments of the
> oiled strip are on the right.

> Maybe those guys worrying about oil on their tubes aren't spouting
> pure bugle oil.

If you go to such lengths to do the test, how about dunking a good
tube in the same oil for a selected time, then wash it off with dish
detergent and put it in the tire for a road test. Then I think we
could discuss the merits of the oil effect. Tearing the tube in half
after that can always be added as a bonus after we hear whether the
tube worked or not at holding air.

Jobst....@stanfordalumni.org

carl...@comcast.net

unread,
May 14, 2005, 7:14:19 PM5/14/05
to

Dear Jobst,

The reason for trying to break thinner and thinner strips
and smaller pieces is that they're easier to tear.

The idea was that I might have been mistaken about how hard
I was pulling on 2-inch wide sections, but when I cut the
dry tube into strips as narrow as you see, I still wasn't
breaking the dry rubber, but was easily breaking oiled
rubber strips of the same width, dozens of times.

Concerning the "fuzzy focus," I think that the picture makes
the width of the three dry strips that refused to break
quite obvious, compared to the dozens of pieces into which
the same-size oiled piece broke quite easily.

A microscope might show interesting details, but I can see
the point with my glasses off and my good eye shut--three
widths of dry rubber didn't break, while the oiled rubber
turned into confetti.

As for "going to such lengths," dunking a piece of inner
tube into oil, pulling it out a few months later, and
pulling it to pieces with my fingers is not exactly
difficult testing.

(Of course, if Hugh Hefner offers to hire me because I
thought of stapling the widest piece of dry tube to the
paper to flatten it out, I'm ready to discuss my photography
fees.)

The surprising loss of tensile strength when rubber is oiled
suggests that not all of the concerns about getting oil onto
inner tubes are mistaken.

I agree that the local effect of a single drop might not
matter, but I'm no longer willing to dismiss the idea out of
hand--the tube is stretched inside the tire and under
pressure, and your own business-card-inside-the-tire test
shows that there's constant stretching.

When removed, most tubes show local deformation, either
where the spokes stick up into the tube under the rim liner
or where the tube presses the liner into the hole where the
recessed spoke head lurks.

Those spots seem like bad places to add a drop of oil, so I
plan to wipe as much of it up as I can in the future (or use
non-petroleum oil).

In any case, it turned out to be an easy and amusing
demonstration of Joe Riel's point about petroleum oil
reducing the tensile strength of butyl rubber.

Carl Fogel

Jim Smith

unread,
May 14, 2005, 7:48:28 PM5/14/05
to
carl...@comcast.net writes:

A quick search turns up plenty of evidence that butyl rubber is in
fact affected by petroleum oil. The results of your demonstration
seem to be exactly what one would expect. The correct line of
inquiry, as Jobst correctly points out, is not whether oil affects
butyl rubber (it does), but whether this has any effect on its use in
bicycle tires.


As an aside, I am having more difficulty finding information on
exactly what is going on at the chemical level when oil comes in
contact with butyl rubber. Does anyone reading this have that
knowledge? I would imagine it is a well studied phenomenon.

jobst....@stanfordalumni.org

unread,
May 14, 2005, 8:54:19 PM5/14/05
to
Carl Fogel writes:

> The reason for trying to break thinner and thinner strips and
> smaller pieces is that they're easier to tear.

I suspect you mean narrower strips which doesn't necessarily mean
smaller. However from the fuzzy picture it seems the got smaller in X
and Y. I can't really tell.

> The idea was that I might have been mistaken about how hard I was
> pulling on 2-inch wide sections, but when I cut the dry tube into
> strips as narrow as you see, I still wasn't breaking the dry rubber,
> but was easily breaking oiled rubber strips of the same width,
> dozens of times.

Aha, the way you described it I took it to mean that you kept tearing
the single existing sample into smaller and smaller rectangles which
would have been progressively more difficult to grasp and clumsier to
tear. I was wondering why the tweezers didn't tear a hole in the
ends.

> Concerning the "fuzzy focus," I think that the picture makes the
> width of the three dry strips that refused to break quite obvious,
> compared to the dozens of pieces into which the same-size oiled
> piece broke quite easily.

I couldn't tell if they were rubber or whatever from the picture, nor
was the torn edge visible.

> A microscope might show interesting details, but I can see the point
> with my glasses off and my good eye shut--three widths of dry rubber
> didn't break, while the oiled rubber turned into confetti.

There is a long way between a sharp close-up photo and microscope
photography. I would feel better if the picture was recognizable as
pieces of rubber from a tube.

> As for "going to such lengths," dunking a piece of inner tube into
> oil, pulling it out a few months later, and pulling it to pieces
> with my fingers is not exactly difficult testing.

OK, so do the real test of whether it affects tube performance rather
than just performing a tensile test.

> (Of course, if Hugh Hefner offers to hire me because I thought of
> stapling the widest piece of dry tube to the paper to flatten it
> out, I'm ready to discuss my photography fees.)

> The surprising loss of tensile strength when rubber is oiled
> suggests that not all of the concerns about getting oil onto inner
> tubes are mistaken.

As I said, it's all in the fears of the beholder. I have never had a
tube failure or even an oil mark from oiling spoke nipples on a
rebuilt wheel, and I have been doing this for many years as I have
already mentioned. I just exhumed a cast off MA-40 from my stack of
worn out rims that someone pretzled years ago and left here. I spoked
it up, straightened it and it runs like a dream. I'll see how soon it
develops cracks, but then it was a free rim.

Anyway, I don't believe that such oiling presents the problem that it
is made out to be in various discussions here.

> I agree that the local effect of a single drop might not matter, but
> I'm no longer willing to dismiss the idea out of hand--the tube is
> stretched inside the tire and under pressure, and your own
> business-card-inside-the-tire test shows that there's constant
> stretching.

There is a spook behind every corner for some folks. I like to asses
the probability of these lurking hazards and make a weighted judgment.
If I followed the common thread here I wouldn't descend mountain roads
at the speeds I and my pals do either.

> When removed, most tubes show local deformation, either where the
> spokes stick up into the tube under the rim liner or where the tube
> presses the liner into the hole where the recessed spoke head lurks.

Sooooo? How many blowouts have we seen reported here on that count?

> Those spots seem like bad places to add a drop of oil, so I plan to
> wipe as much of it up as I can in the future (or use non-petroleum
> oil).

And be sure to use "Sea Salt" when tossing it over your shoulder, that
stuff from salt mines (Salzburg) is no good.

> In any case, it turned out to be an easy and amusing demonstration
> of Joe Riel's point about petroleum oil reducing the tensile
> strength of butyl rubber.

Hold it. The problem is substantially greater with tubulars and latex
tubes. I have seen failures there, even on new tires that were
probably stored in an automotive oily exhaust atmosphere. Those
yellow tubes took on a slightly greenish color and could be pulled
apart with a slight tug. That is, when we opened them for repair we
could not pull the tube out without tearing it even though it was not
stuck in the casing. That's when we got good at re-tubing tubulars.

Jobst....@stanfordalumni.org

jim beam

unread,
May 14, 2005, 9:17:30 PM5/14/05
to

for public record, state the drive side spoke tension for this wheel
jobst. you have a tensiometer.

eh? further above, you say oil doesn't ruin tubes, yet here you state
that oil /does/ ruin tubes. which do you want us to believe?

> Jobst....@stanfordalumni.org

carl...@comcast.net

unread,
May 14, 2005, 9:28:15 PM5/14/05
to

Dear Jobst,

Cut a two-inch section out of a dry inner tube. Try to pull
the thin-walled rubber pipe apart with your fingers.

Now cut the thin-walled pipe lengthwise and try to pull the
strip apart--it's easer to grasp.

Now cut the strip in half lengthwise. It's narrower or
thinner or however you like to describe it. Try to pull it
apart, grabbing each end between a thumb and forefinger.

Now cut it lengthwise again and try to pull the narrower
strip apart.

It's awfully hard to do with a normal dry tube.

But it's awfully easy to pull the tube marinated in motor
oil to pieces and then pull those pieces apart again--the
repeated failures show that it's not just a single weak spot
that just happened to be in that section of tube.

Perhaps I need to explain more clearly that the picture
shows two sections, each two inches long, cut from the same
inner tube:

http://home.comcast.net/~carlfogel/download/tube_oil.jpg

On the left is a section of dry tube cut into three strips,
the widest showing two staples that flatten it out.
Obviously, they didn't break when repeatedly pulled both
lengthwise and sideways. The tensile strength is impressive.

On the right is a section cut from the same tube, dunked in
oil as described, and easily pulled into more than three
dozen pieces. Obviously, it lost lots of tensile strength.

The difference isn't as dramatic as matchsticks and uncooked
spaghetti, but it's rather striking. Anyone who's curious
can easily duplicate it in a few weeks with a tube, some
motor oil, a cup, and a dead wasp. Anyone can pull the oiled
tube to pieces with bare fingers.

And--hold it--what's that at the end of your post? After all
that quibbling, are you now saying that there is a
substantially greater problem with getting oil on tubulars
and latex inner tubes, just from an "oily atmosphere"?

Carl Fogel

jobst....@stanfordalumni.org

unread,
May 14, 2005, 11:05:06 PM5/14/05
to
Wild Turkey writes:

>> OK, so do the real test of whether it affects tube performance rather
>> than just performing a tensile test.

>>> (Of course, if Hugh Hefner offers to hire me because I thought of
>>> stapling the widest piece of dry tube to the paper to flatten it
>>> out, I'm ready to discuss my photography fees.)

>>> The surprising loss of tensile strength when rubber is oiled
>>> suggests that not all of the concerns about getting oil onto inner
>>> tubes are mistaken.

>> As I said, it's all in the fears of the beholder. I have never had
>> a tube failure or even an oil mark from oiling spoke nipples on a
>> rebuilt wheel, and I have been doing this for many years as I have
>> already mentioned. I just exhumed a cast off MA-40 from my stack
>> of worn out rims that someone pretzled years ago and left here. I
>> spoked it up, straightened it and it runs like a dream. I'll see
>> how soon it develops cracks, but then it was a free rim.

> for public record, state the drive side spoke tension for this wheel
> jobst. you have a tensiometer.

Holy spoke gauge! What's this public record of which you speak. My
front wheel doesn't have a drive and its tension is the same as that
of the MA-2 that this rim replaced. That rim was worn practically
through in the last three and a half years... and get that shift key
fixed. You can do it.

Please quote that paragraph... and who is "us"?

Jobst....@stanfordalumni.org

jim beam

unread,
May 14, 2005, 11:18:50 PM5/14/05
to

perhaps the lack of shift key has you confused? read your own work.

jobst....@stanfordalumni.org

unread,
May 14, 2005, 11:23:44 PM5/14/05
to
Carl Fogel writes:

> Cut a two-inch section out of a dry inner tube. Try to pull the
> thin-walled rubber pipe apart with your fingers.

> Now cut the thin-walled pipe lengthwise and try to pull the strip
> apart--it's easer to grasp.

> Now cut the strip in half lengthwise. It's narrower or thinner or
> however you like to describe it. Try to pull it apart, grabbing each
> end between a thumb and forefinger.

> Now cut it lengthwise again and try to pull the narrower strip
> apart.

Hey, that's enough. I said I understand now what you were trying to
convey about your test method. It wasn't clear from the original text
which the unclear pictures reinforced. A notion that you kept on
working with ever smaller rectangular pieces that should have gotten
progressively harder to tear.

> It's awfully hard to do with a normal dry tube.

> But it's awfully easy to pull the tube marinated in motor oil to
> pieces and then pull those pieces apart again--the repeated failures
> show that it's not just a single weak spot that just happened to be
> in that section of tube.

> Perhaps I need to explain more clearly that the picture shows two
> sections, each two inches long, cut from the same inner tube:

> http://home.comcast.net/~carlfogel/download/tube_oil.jpg


A (good) picture is worth 10,000 words except in this case. I still
don't understand how these pieces that seem to be too small for mu
fingers, tore into rectangular micro rectangles. Your description
doesn't clarify the picture.

I got the impression from your second explanation that you cut a tube
into long and progressively narrower strips and pulled them apart. I
don't see that in the picture. Why?

> On the left is a section of dry tube cut into three strips, the
> widest showing two staples that flatten it out. Obviously, they
> didn't break when repeatedly pulled both lengthwise and sideways.
> The tensile strength is impressive.

Well, I'm sure it isn't any more than the tubes I use and I don't find
that impressive having used butyl tubes for many years on bicycles,
motorcycles and cars.

> On the right is a section cut from the same tube, dunked in
> oil as described, and easily pulled into more than three
> dozen pieces. Obviously, it lost lots of tensile strength.

How did you tear them apart. I know you explained it but I can't seem
to reconcile the picture with the description. With what did you pull
and how did they tear into neat rectangles?

> The difference isn't as dramatic as matchsticks and uncooked
> spaghetti, but it's rather striking. Anyone who's curious can
> easily duplicate it in a few weeks with a tube, some motor oil, a
> cup, and a dead wasp. Anyone can pull the oiled tube to pieces with
> bare fingers.

> And--hold it--what's that at the end of your post? After all that
> quibbling, are you now saying that there is a substantially greater
> problem with getting oil on tubulars and latex inner tubes, just
> from an "oily atmosphere"?

As I said, a whole shipment of tubulars came in one day with "brittle"
tubes that smelled as though they got oiled. We replaced the tubes.

Just the same, I've been oiling spoke nipples on wheels with tires
mounted for many years and never detected an oil spot on a tube or had
any tube failures, with tubulars or clinchers, from this practice. I
am certain that the oil bugaboo is a spook without merit.

Just imagine how much oil you could get into a rim cavity past the
nipple of a tensioned spoke. Capillary attraction would take care of
any excess and beyond that you can't even see it on the inside of the
rim. Maybe that would be a better test rather than to asses what
would happen IF oil got on the tube.

Jobst....@stanfordalumni.org

PanFan

unread,
May 14, 2005, 11:42:06 PM5/14/05
to
jobst....@stanfordalumni.org wrote in news:meuhe.473$W51.4470
@typhoon.sonic.net:

> I think you should consider that a properly sized tube doesn't suffer
> any stress in use other than preventing air from diffusing through its
> walls that are supported by the rim and tire casing against stretch.

Oh. When the sidewall of a properly inflated tire deflects outward under
load, I always assumed the tube deflects with it. Silly me.

jim beam

unread,
May 14, 2005, 11:53:00 PM5/14/05
to
jobst....@stanfordalumni.org wrote:
> Wild Turkey writes:
>
>
>>>OK, so do the real test of whether it affects tube performance rather
>>>than just performing a tensile test.
>
>
>>>>(Of course, if Hugh Hefner offers to hire me because I thought of
>>>>stapling the widest piece of dry tube to the paper to flatten it
>>>>out, I'm ready to discuss my photography fees.)
>
>
>>>>The surprising loss of tensile strength when rubber is oiled
>>>>suggests that not all of the concerns about getting oil onto inner
>>>>tubes are mistaken.
>
>
>>>As I said, it's all in the fears of the beholder. I have never had
>>>a tube failure or even an oil mark from oiling spoke nipples on a
>>>rebuilt wheel, and I have been doing this for many years as I have
>>>already mentioned. I just exhumed a cast off MA-40 from my stack
>>>of worn out rims that someone pretzled years ago and left here. I
>>>spoked it up, straightened it and it runs like a dream. I'll see
>>>how soon it develops cracks, but then it was a free rim.
>
>
>>for public record, state the drive side spoke tension for this wheel
>>jobst. you have a tensiometer.
>
>
> Holy spoke gauge! What's this public record of which you speak. My
> front wheel doesn't have a drive and its tension is the same as that
> of the MA-2 that this rim replaced. That rim was worn practically
> through in the last three and a half years... and get that shift key
> fixed. You can do it.

oh, i missed this dodge. what a peach.

jobst, you want to hold up your "new" ma40 as a cracking candidate, but
to do that, you need to post here, on this forum, the spoke tension with
which you've built that wheel. building wheels with spoke tension in
excess of the rim manufacturer's tension spec /will/ lead to cracking.
just in case you have not understood any of our previous discussions on
this subject, rim cracking is *not* a function of anodizing. it *is* a
function of spoke tension.

post your figures.

carl...@comcast.net

unread,
May 15, 2005, 12:09:19 AM5/15/05
to

[snip]

Dear Jobst,

Smaller (narrower) pieces are easier to pull apart, as long
as they're still big enough to grip between thumb and
forefinger--the bigger (wider) the strip, the stronger it
is.

Think of rubber bands--the wider, the stronger. The original
single dry piece was cut lengthwise. Eventually, it cut into
three increasingly narrow bands, all of which refused to
pull apart:

http://home.comcast.net/~carlfogel/download/tube_oil.jpg

The similar single oil-marinated piece was cut lengthwise
and pulled apart. Then I pulled each piece apart again and
again and again--the oil had weakened the rubber, so it was
easy.

So what's puzzling you?

First you say "Hey, that's enough" when I explain that I
pulled a piece of oil-weakened inner-tube into smaller and
smaller pieces with my fingers.

But then you say that you "still don't understand how these
pieces that seem to be too small for my fingers tore into
rectangular micro rectangles."

The secret is dunking the rubber in oil for a few months,
which as Joe Riel suggested can reduce its tensile strength
by more than 60%, and then pulling the rubber apart with
your fingers into smaller and smaller pieces.

Maybe you're not wondering why the rubber tears, but rather
why it tears into rectangles? I don't know, but it does
indeed pull apart abruptly in the middle between your
fingers in a nice, straight line.

I suppose that a crack develops somewhere and then grows too
fast to be noticed. It's not a visibly ragged tear like
paper or a tissue--in fact, the breaks are nice and clean
even under magnification, with extremely sharp corners.
They're not perfect right angles, but they're darned close.

The three long strips of the dry piece just stretch and
narrow without breaking. With a really hard pull, they'd
presumably either slip out from between my fingers, snap
cleanly in a similar fashion, or else rip in more ragged
manner.

But the oil-weakened pieces just crack, snap, break, bust,
fracture, or whatever the right term is. And they do so
fairly easily.

If you want to measure the size of the pieces, those are two
standard staples in the widest dry strip on the left. They
measure about 13 mm on my calipers. The three long strips
joined together and rolled into a cylinder were originally a
section of ordinary inner tube.

Carl Fogel

carl...@comcast.net

unread,
May 15, 2005, 12:22:07 AM5/15/05
to
On Sun, 15 May 2005 03:42:06 GMT, PanFan <nom...@none.org>
wrote:

Dear P.,

Yes, that's what I was thinking when I reminded Jobst of his
squirm demonstration, the one in which a business card
between the tube and tire is cracked into confetti.

I even tried it myself on a whim one day. After a 15-mile
ride, I took the card out of my rear tire. At first, I
thought that it was intact and wondered if 125 psi had
reduced the squirm, a Kevlar belt had kept things from
stretching, or if fifteen miles wasn't enough.

Then I looked at the business card more closely.

It was smudged and curled to the shape of the tire, but what
I thought at first were just three dirty crosswise creases
were actually cracks in the middle of the card--when I
flexed it slightly, they opened and I could see daylight
through them. (They're all about a centimeter long--I still
have the card.)

Closer inspection showed the start of a lengthwise tear and
some more weakening spots where crosswise tears were
beginning form on one edge.

Interestingly, the three tears were roughly in the middle of
the card and hadn't reached any edge yet.

Carl Fogel

Neil Brooks

unread,
May 15, 2005, 12:45:23 AM5/15/05
to
carl...@comcast.net wrote:

>But when I sat down at the computer to reveal that oil has
>no more effect on rubber than garlic on werewolves, I
>stretched the piece of inner tube sideways a few more times.
>
>It broke.
>
>More stretching kept breaking it. The smaller the pieces of
>oiled rubber were, the easier they broke--it was almost as
>satisfying as popping bubble wrap.
>
>Then I tried stretching the dry section of inner tube. It
>wouldn't break.
>
>After a while, I got tired and cut the dry piece in half. It
>still wouldn't break.
>
>So I cut it in half again. Same result--even the thinner
>strip of dry rubber wouldn't break despite repeated
>stretching, while pieces of oiled rubber the same width
>broke easily with a single pull.

My theory? You tired yourself out on the oiled piece ;-)

Zog The Undeniable

unread,
May 15, 2005, 4:11:21 AM5/15/05
to

I once booted a slashed sidewall (piece of glass bottle - major blowout)
with a piece of cigarette carton I found nearby. It only lasted 4 miles
before blowing out again, but that was enough to get within a mile of
home. Apparently plastic crisp packets are much better, but I now carry
Park tyre boots to be on the safe side.

It's a bit much to expect card to flex as much as rubber, I suppose.

jobst....@stanfordalumni.org

unread,
May 15, 2005, 9:51:01 AM5/15/05
to
Someone writes:

>> I think you should consider that a properly sized tube doesn't
>> suffer any stress in use other than preventing air from diffusing
>> through its walls that are supported by the rim and tire casing
>> against stretch.

> Oh. When the sidewall of a properly inflated tire deflects outward
> under load, I always assumed the tube deflects with it. Silly me.

The load is held by the casing. We aren't talking flex here but
tensile stress.

Jobst....@stanfordalumni.org

Bill Sornson

unread,
May 15, 2005, 10:48:19 AM5/15/05
to
Zog The Undeniable wrote:

> I once booted a slashed sidewall (piece of glass bottle - major
> blowout) with a piece of cigarette carton I found nearby. It only
> lasted 4 miles before blowing out again, but that was enough to get
> within a mile of home. Apparently plastic crisp packets are much
> better, but I now carry Park tyre boots to be on the safe side.
>
> It's a bit much to expect card to flex as much as rubber, I suppose.

Menthol works better.

(Hope this gets past...filters.)

Matchless Bill


carl...@comcast.net

unread,
May 15, 2005, 3:03:36 PM5/15/05
to

Dear Jobst,

What's the difference between flex and tensile stress?

Doesn't any bending put one surface into tension?

Carl Fogel

michael overton

unread,
May 15, 2005, 3:55:56 PM5/15/05
to

"Neil Brooks" <Neil...@yahoo.com> wrote in message
news:7rkd81t30u66j7nq6...@4ax.com...

I agree. You might lube your spoke nipples say 4...okay maybe 6 times in a
riding season (and I do) with TriFlow or some such other petro based lube.
As someone pointed out, that tube simply contains air in an already confined
space..you are not stretching it out or working it like "bubble wrap". It
begs the bigger question...will that tube last the season regardless of
petroleum contact? Possibly but the odds aren't super. There surely must be
bigger concerns about ones tires and spokes other than a bit of petroleum
contact. I mean you are riding the bike over asphalt, tar, macadam, etc.
Another petroleum based medium. Gads, what the hell is that doing to your
tires as they roll over it? For that matter what is that pair of Pearl Izumi
shorts doing in contact with your skin. My god...your ass might go flat!

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Mark Hickey

unread,
May 15, 2005, 4:25:41 PM5/15/05
to
carl...@comcast.net wrote:

>Back in January, Joe Riel explained why some riders fear
>that petroleum-based oil from lubricated spoke nipples might
>damage their butyl inner tubes:

<snip>


>Curious, I cut a pair of 2-inch sections out of an inner
>tube, set one aside, recklessly dunked the other in a cup of
>10w-40, and promptly forgot about the experiment.

Great work!

I've often postulated that the failure of tubes around the stem could
be from oil getting on the tube (it's happened to me on new rims that
I built using oil to lubricate the nipples).

It occurs to me that it wouldn't be much of a stretch to assume that
the occasional drop of oil (even one carried from the chain by rain
water) could score a "bullseye" on the inflator valve, and would then
be flung onto the tube by centrifugal force.

Hence, the odd tubing failures that are reported on these forums over
and over...

Mark Hickey
Habanero Cycles
http://www.habcycles.com
Home of the $695 ti frame

Leo Lichtman

unread,
May 15, 2005, 4:48:36 PM5/15/05
to

<carl...@comcast.net> wrote: (clip) Doesn't any bending put one surface
into tension?
^^^^^^^^^^^^
It certainly does, but I would assume, without doing the calculations, that
the stress of bending a tube would be far below the strength of the
material. The issue really is: how much tensile stress does the tube
develop due to inflation? If, as Jobst assumes, the tube is a close fit to
the inside of the tire, then the inflation pressure is matched by the forces
of the tire pressing against the outside of the tube, and there will be no
"inflation stress." However, most tubes are made to "fit" a range of tire
sizes, so it is possible that a tube may have to stretch out to fit the
tire, and will have some "inflation stress." As the tire rolls, there will
be a little additional stess developed at the sidewall, due to bending, as
Carl suggests, but this is in an area where oil is extremely unlikely to
reach.

I think the argument is moot, but it is fun to watch both your minds work.

Carl, one thing I do object to is the infliction of needless suffering on
mud wasps in the name of research.


Michael Press

unread,
May 15, 2005, 9:01:57 PM5/15/05
to
In article <ttbf811d7a6ampkss...@4ax.com>,
Mark Hickey <ma...@habcycles.com> wrote:


... and one in the eye for those who throw away their valve stem
retainer nuts. :)

--
Michael Press

carl...@comcast.net

unread,
May 15, 2005, 10:42:07 PM5/15/05
to

Dear Leo,

If I hadn't removed so many permanently dimpled inner tubes,
I'd agree that nothing is happening at the spokes.

But something deforms tubes there on the inside of the
cylinder.

I do agree that there's no rash of tube failures likely to
be attributed to oil on the rubber.

But once you pull a normally indestructible piece of inner
tube to pieces with your fingers, you may decide to be more
careful about getting oil on the tube. Odd as it seems, the
oily tube has become brittle, so it will fatigue and crack
much more easily.

Notice how hard it was for Jobst to come to grips with what
the picture represented--the stumbling block may have been
that none of us expects a bicycle tube to lose so much
tensile strength that it can be pulled into confetti with
bare fingers. (It certainly startled me.)

As for the insect, that mud wasp got exactly what it
deserved. If it had been a cow, I'd have barbecued it, using
chunks of its own oil-soaked carcass for fuel. Possibly it
would have been quite tender, too, the oil having reduced
the tensile strength of the meat to where it could be cut
with a fork, like Kobe steaks.

Brutally,

Carl Fogel

PanFan

unread,
May 16, 2005, 1:17:09 AM5/16/05
to

Well, if we're talking about tubes being weakened by oil, then flex is in
the picture. If the tire is deflecting under load, then the tube is
deflecting too (I assume together as a system). All that flex is generating
some complex stresses, strains and heat in the tube wall. That means an
oil-weakened tube is more likely to fail in use.

data...@yahoo.com

unread,
May 16, 2005, 11:02:11 AM5/16/05
to
yup. oil on cheap rubber is not good maintenance but one could use
linseed oil: grease and oil do not have an extreme effect on "real
rubber" the experiment-i screwed up the cranks twice after using the
copper anti-seize then did a BB repack and see the copper goes dry
between the crank and spindle-BUT no rust!! the spindle surface looks
clean as a...
yet i'm too old fashioned to believe a dry compund protects from
moisture intrusion! time for a bolt nut with copper left outside in
summer's thunderstorms.

data...@yahoo.com

unread,
May 16, 2005, 12:29:52 PM5/16/05
to
whoa! my examples of "real rubber" are covered with grease not
detergent motor oil

Zog The Undeniable

unread,
May 16, 2005, 1:33:38 PM5/16/05
to
Mark Hickey wrote:

> I've often postulated that the failure of tubes around the stem could
> be from oil getting on the tube (it's happened to me on new rims that
> I built using oil to lubricate the nipples).

That's often due to an excessively large diameter valve hole (Presta
valve in Schrader rim being the worst example), which allows the valve
to sink too far into the rim and puts strain on the rubber. The other
reason is careless pumping, especially with no valve nut.

jobst....@stanfordalumni.org

unread,
May 16, 2005, 2:50:11 PM5/16/05
to
Carl Fogel writes:

>>> I think you should consider that a properly sized tube doesn't
>>> suffer any stress in use other than preventing air from diffusing
>>> through its walls that are supported by the rim and tire casing
>>> against stretch.

>> Oh. When the sidewall of a properly inflated tire deflects outward
>> under load, I always assumed the tube deflects with it. Silly me.

> Yes, that's what I was thinking when I reminded Jobst of his squirm


> demonstration, the one in which a business card between the tube and
> tire is cracked into confetti.

> I even tried it myself on a whim one day. After a 15-mile ride, I
> took the card out of my rear tire. At first, I thought that it was
> intact and wondered if 125 psi had reduced the squirm, a Kevlar belt
> had kept things from stretching, or if fifteen miles wasn't enough.

> Then I looked at the business card more closely.

> It was smudged and curled to the shape of the tire, but what I
> thought at first were just three dirty crosswise creases were
> actually cracks in the middle of the card--when I flexed it
> slightly, they opened and I could see daylight through
> them. (They're all about a centimeter long--I still have the card.)

> Closer inspection showed the start of a lengthwise tear and some
> more weakening spots where crosswise tears were beginning form on
> one edge.

> Interestingly, the three tears were roughly in the middle of the
> card and hadn't reached any edge yet.

We're drifting... The business card test only shows that paper is
not elastic and that flexing of a bias ply tire causes microscopic
distortions that in time will cause paper to separate into tiny fuzzy
rectangles. This is all within the range that even an oil weakened
butyl inner tube can sustain.

As I suggested, a valid test would be to soak a whole tube in the same
oil, all except the valve stem area and ride it after cleaning it of
residual oil. We don't want stress on the stem during inflation to
cause a failure before the test can begin. After all, a drop of oil
on a spoke nipple would probably not travel to the valve stem in real
life.

Jobst....@stanfordalumni.org

jobst....@stanfordalumni.org

unread,
May 16, 2005, 2:59:53 PM5/16/05
to
Carl Fogel writes:

>>>> I think you should consider that a properly sized tube doesn't
>>>> suffer any stress in use other than preventing air from diffusing
>>>> through its walls that are supported by the rim and tire casing
>>>> against stretch.

>>> Oh. When the sidewall of a properly inflated tire deflects outward
>>> under load, I always assumed the tube deflects with it. Silly me.

>>The load is held by the casing. We aren't talking flex here but
>>tensile stress.

> What's the difference between flex and tensile stress?

In bending there is a neutral axis of no stress, usually in the center
of the bending cross section. Bending a rubber sheet in a radius of
more than several sheet thicknesses produces only light stress in the
rubber while tension puts the entire cross section under stress and a
lot more than a 15 degree bend with large radius.

A good test for this is how much force the two processes take when
performed by hand. You can barely notice any resistance when bending
a sheet of rubber.

> Doesn't any bending put one surface into tension?

Yes, but as in most of these technical matters, degree is more
important than the basic event, just as immersing a tube in oil is to
putting a drop of oil on the outside of a rim at a spoke nipple.

Soap bubbles explode and so do atomic bombs. It isn't the same.

Jobst....@stanfordalumni.org

jobst....@stanfordalumni.org

unread,
May 16, 2005, 3:25:06 PM5/16/05
to
Mark Hickey <ma...@habcycles.com> writes:

>> Back in January, Joe Riel explained why some riders fear that
>> petroleum-based oil from lubricated spoke nipples might damage

>> their butyl inner tubes.

>> Curious, I cut a pair of 2-inch sections out of an inner tube, set
>> one aside, recklessly dunked the other in a cup of 10w-40, and
>> promptly forgot about the experiment.

> Great work!

> I've often postulated that the failure of tubes around the stem
> could be from oil getting on the tube (it's happened to me on new
> rims that I built using oil to lubricate the nipples).

Well that ain't so. That was a manufacturing error that didn't occur
before that time nor has is done so since, at least in the massive
numbers of those tubes whose stems separated more than a year ago. At
that time the masochists claimed it was their foolish use of valve
stem (knurled) nuts that caused separation, and now we have oil on
spoke nipples as the cause. Do I hare another spook around the
corner?

Recall the test I propoased: Screw the nut on a tight as you can get
it by hand, inflat the tire to 100 psi and notice the nut is loose and
free of the rim. Tire ptressure is a far greater force tha the hand
tightned knureld nut. However, that doen't seem to bother the
faithful in their beliefs.

> It occurs to me that it wouldn't be much of a stretch to assume that
> the occasional drop of oil (even one carried from the chain by rain
> water) could score a "bullseye" on the inflator valve, and would
> then be flung onto the tube by centrifugal force.

That's a huge stretch considering how we have never seen a tube
failure from oil reported here, nor have I in many years of patching
tubes seen any other than the tubulars with latex tubes that came from
the factory (like the separating stems) in a chemically weakened
state. I've trued many wheels that required a drop of oil on spoke
nipples and have not seen any tube damage, or for that matter a trace
of oil on a tube.

In fact, if spoke wrench users would put a drop of oil on the spoke
nipple interface they wouldn't need to use clumsy four corner grip
spoke wrenches. A slot type spoke wrench being adequate to twist off
any spoke I have tightened when its nipple isn't galling.

As an interesting sidelight, welding of dissimilar metals
(brass/steel) was perfected in Russia as inertia welding, where two
dissimilar metal parts are pressed together while one of them is
spinning, much like a spoke nipple and eyelet.

> Hence, the odd tubing failures that are reported on these forums
> over and over...

You had best cite one of these. My recollection is that these were
always loud blowouts that were claimed to have occurred inside the
tire casing. As is often explained, "if you heard the bang, the tube
was outside the tire". End of mystery. These people are not aware
that tires drop back into the rim after lift-off and lead the
inexperienced rider to believe the bang somehow got outside the tire
to be heard.

And you say you run a bicycle shop??? This IS spooky!

Jobst....@stanfordalumni.org

Jasper Janssen

unread,
May 16, 2005, 3:47:10 PM5/16/05
to
On Mon, 16 May 2005 05:17:09 GMT, PanFan <nom...@none.org> wrote:

>Well, if we're talking about tubes being weakened by oil, then flex is in
>the picture. If the tire is deflecting under load, then the tube is
>deflecting too (I assume together as a system). All that flex is generating
>some complex stresses, strains and heat in the tube wall. That means an
>oil-weakened tube is more likely to fail in use.

I have never seen a tube fail without either a hole or foreign object
through the tyre, or a sharpish edge at the rim side.

Jasper

carl...@comcast.net

unread,
May 16, 2005, 11:11:31 PM5/16/05
to

Dear Jobst,

Sorry, the tube is already soaking.

As pointed out in this thread, chain oil may well get down
through the space between the rim and the valve stem.

And what a strange idea--let's not test one part because it
might fail?

Carl Fogel

carl...@comcast.net

unread,
May 16, 2005, 11:21:29 PM5/16/05
to

Dear Jasper,

O horror, horror, horror! Tongue nor heart
Cannot conceive nor name thee!

Confusion now hath made his masterpiece!
Most sacrilegious tube failure hath broke ope
The Lord's anointed temple, and stole thence
The air o' the tire!

Approach the chamber, and destroy your sight
With a new Gorgon: do not bid me speak;
See, and then speak yourselves.

Inner tube bumps formed into spoke holes:

http://home.comcast.net/~carlfogel/download/tubebump.jpg

Inner tube failure near bump by valve:

http://home.comcast.net/~carlfogel/download/tubebump2.jpg

(Possibly anointing the tube was a bad idea.)

Macduff

jobst....@stanfordalumni.org

unread,
May 17, 2005, 12:19:09 AM5/17/05
to
Carl Fogel writes:

> Sorry, the tube is already soaking.

> As pointed out in this thread, chain oil may well get down through
> the space between the rim and the valve stem.

Whoa! That was the most far fetched oil transfer I've seen yet and as
another contributor curtly stated, we haven't seen a tube fail from
oil in regular use.

> And what a strange idea--let's not test one part because it might
> fail?

Mainly because getting oil on the stem and surround is really out of
the question under the conditions we are investigating here.

My point, besides that soaking a tube in oil doesn't occur in reality,
is that there are no significant forces on the tube inside a tire, its
elasticity being enough to withstand the small distortions encountered
in use.

Jobst

jobst....@stanfordalumni.org

unread,
May 17, 2005, 12:25:42 AM5/17/05
to
Carl Fogel writes:

> Inner tube bumps formed into spoke holes:

http://home.comcast.net/~carlfogel/download/tubebump.jpg

> Inner tube failure near bump by valve:

http://home.comcast.net/~carlfogel/download/tubebump2.jpg

> (Possibly anointing the tube was a bad idea.)

What sort of rim tape were you using and how much oil got on this
tube?

You need to read the close-up instructions for your camera. This is
all so fuzzy that no detail of interest can be seen. How does this
tube failure next to the stem look in a frontal view. If it isn't a
dimple-concentric starburst than it is a chafing failure induced by
movement during inflation.

Let's try to keep cause and effect aligned.

Jobst....@stanfordalumni.org

carl...@comcast.net

unread,
May 17, 2005, 1:04:48 AM5/17/05
to

Dear Jobst,

First, Jasper's comment, to which I was replying:

"I have never seen a tube fail without either a hole or
foreign object through the tyre, or a sharpish edge at the
rim side."

Now Jasper can say that he has seen a tube failure that had
nothing to do with a hole, a foreign object, or a sharpish
edge.

Second, you've already indicated that no amount of oil can
possibly matter, so why worry? It must be myth and
lore--except if tubulars are left in the oily atmosphere of
a garage, in which case you sing a different tune, or if it
gets near the valve stem, which you have decreed does not
matter.

Third--oh, why bother?

Carl Fogel

jobst....@stanfordalumni.org

unread,
May 17, 2005, 1:04:35 AM5/17/05
to
Carl Fogel writes:

> First, Jasper's comment, to which I was replying:

> "I have never seen a tube fail without either a hole or foreign
> object through the tyre, or a sharpish edge at the rim side."

I'm not convinced. I couldn't see anything on that picture that
showed what occurred.

> Now Jasper can say that he has seen a tube failure that had nothing
> to do with a hole, a foreign object, or a sharpish edge.

> Second, you've already indicated that no amount of oil can possibly
> matter, so why worry? It must be myth and lore--except if tubulars
> are left in the oily atmosphere of a garage, in which case you sing
> a different tune, or if it gets near the valve stem, which you have
> decreed does not matter.

Could you show me where I said no amount of oil could damage a tube?
I think you are confusing Old Crow's assertions with what I wrote. He
claims I wrote all sorts of things that I never wrote for convenience
of argument.

> Third--oh, why bother?

I see you are between a rock and a hard place. It's not that hard.

Jobst....@stanfordalumni.org

carl...@comcast.net

unread,
May 17, 2005, 2:05:21 AM5/17/05
to
On Tue, 17 May 2005 05:04:35 GMT,
jobst....@stanfordalumni.org wrote:

[snip]

>I'm not convinced. I couldn't see anything on that picture that
>showed what occurred.

Dear Jobst,

You never are. You never do.

Carl Fogel

waxbytes

unread,
May 17, 2005, 2:04:07 AM5/17/05
to

Micheal Overton wrote (in part):

" I mean you are riding the bike over asphalt, tar, macadam, etc.
Another petroleum based medium. Gads, what the hell is that doing to
your
tires as they roll over it? For that matter what is that pair of Pearl
Izumi
shorts doing in contact with your skin. My god...your ass might go
flat!"

To which I reply:

I can assure everyone that no matter what the shorts I wear are made of
or by
whom they are made, there is no chance at all that my ass might go
flat, no
matter how much I might wish it to do so.


--
waxbytes

Bad Idea

unread,
May 17, 2005, 10:45:03 PM5/17/05
to
This whole thread is based on the false premise
that an inner tube can fail in tension. Jobst
was almost correct when he wrote:

... a properly sized tube doesn't suffer any stress in use...

A tube is only stressed when it is squeezed between
the rim and the contact patch and is subjected
to *compression*...by the spokes. ;)

Bad Idea

unread,
May 17, 2005, 11:06:55 PM5/17/05
to

Leo Lichtman

unread,
May 17, 2005, 11:34:47 PM5/17/05
to

"Bad Idea" wrote: (clip) A tube is only stressed when it is squeezed
between the rim and the contact patch and is subjected to *compression*...by
the spokes. ;)
^^^^^^^^^^^^^^^^
That can't be right. You seem to be saying that the tube is compressed
between the road and the spokes at the bottom of the wheel. First, you have
to realize that the pressure inside the tube does not change as the tire
rolls. Second, if you isolate a piece of the tube in contact with the
spokes at the bottom of the wheel, and draw a free-body diagram, you will
see that the forces pressing down on the tube element must be equal to the
air pressure x area of the element pressing up. This does not change as the
element moves to different positions around the wheel. The spokes don't
have a different effect on the tube at the bottom position than at all other
rotational positions.


Mark Hickey

unread,
May 18, 2005, 10:19:39 AM5/18/05
to
jobst....@stanfordalumni.org wrote:

>Mark Hickey <ma...@habcycles.com> writes:

>> I've often postulated that the failure of tubes around the stem
>> could be from oil getting on the tube (it's happened to me on new
>> rims that I built using oil to lubricate the nipples).

<snip>


>> Hence, the odd tubing failures that are reported on these forums
>> over and over...
>
>You had best cite one of these. My recollection is that these were
>always loud blowouts that were claimed to have occurred inside the
>tire casing. As is often explained, "if you heard the bang, the tube
>was outside the tire". End of mystery. These people are not aware
>that tires drop back into the rim after lift-off and lead the
>inexperienced rider to believe the bang somehow got outside the tire
>to be heard.

Same planet, differnt worlds I guess. Did you REALLY miss all the "my
tube failed right by the valve posts"? It's happened to me many times
as well - what that has to do with loud blow-outs will, I'm afraid,
remain a mystery known only to you.

>And you say you run a bicycle shop??? This IS spooky!

I don't recall ever saying that I run a bicycle shop. That IS spooky.

Jasper Janssen

unread,
May 19, 2005, 7:02:35 PM5/19/05
to
On Wed, 18 May 2005 03:34:47 GMT, "Leo Lichtman"
<l.lic...@worldnet.att.net> wrote:

>The spokes don't
>have a different effect on the tube at the bottom position than at all other
>rotational positions.

If the spokes are beyond ridiculously undertensioned, they'll poke out a
little when they're pointing to the contact patch, though. I don't think
tube failure is what a wheel that badly made has to worry about..

Jasper

0 new messages