Bicycle tire-making: the valiant struggle continues.... with wire-coiling

[DougC]

Episode #I forget....

Still going on the tire-making project. The last few days I've found out
how hard it is to coil a long piece of twisted cable without adding
additional twist.

I have this machine that does a step in making the tire's steel belt.
The cable is about equivalent to shifter cable. For a 559-size tire, the
piece of cable needed is about 58 meters long. The cable can't be under
much of any tension as it is drawn into the machine, so I thought it
best to have the cable "laying" in a coil and be drawn off sideways.
This (I thought) is much simpler than adding any kind of
tension-sensitive feeding spool.

If you just try to coil this long of a piece around a simple round path,
somehow it causes additional twist in the cable as it is drawn off.
Originally I figured I'd just coil it inside a bucket. One can see this
because the cable begins to twist up as you try to feed it into the
bucket. At some point it begins to wrap whole twisted bights into the
bucket, that tangle and won't feed properly through the belt-making machine.

If you turn the bucket as you feed, you can successfully hand-feed the
cable into the bucket so that it all lies flat and nice. But then it
still twists again, the OTHER way, coming off,,, and it jams inside the
belt-making machine because of that.

It coils naturally REALLY NICELY inside a small bucket, but there's no
way to avoid the twisting problem if it only turns one direction.

So I took a board and stuck two dowels into it, and tried laying the
cable around the dowels in a figure-8 pattern--and it worked, and did
not cause twisting!!!
:D
But the cable got tangled because of the different size loops getting
caught on each other. So that won't work.

So what I am going to do (tomorrow) is go out and buy a few small cheap
paint buckets and cut a chord of 2-3 inches off each one with a band
saw, and try to bolt them together somehow so that the interior space is
joined in a figure-8 pattern. This way I can coil the cable inside both
buckets, in a figure-8 pattern. Maybe.

Oh well.
At least I don't gotta order buckets from China and wait 3 weeks to get
them.

[Tosspot]

Didn't you have a web page up with a blog. If it's up to date I'd like
a gander.

[DougC]

On 9/25/2016 2:01 PM, Tosspot wrote:
>
> Didn't you have a web page up with a blog. If it's up to date I'd like
> a gander.
>
>

Still there but not updated since Sept 2015.
http://www.norcom2000.com/users/dcimper/assorted/inanities/recumbent/tire_making/tire_making_main.html

My website managing software stopped working and it took me a while to
figure that one out (the ISP changed their FTP server address by 1).

Also there's just not been much to look at.

I was all ready last week to try making a couple more tires, I had a few
more issues solved--but the wire-twisting issue arose. I had run shorter
pieces (10 ft) through the belt making machine and it worked, but the
short pieces don't show the twisting problem. It only happens with long
pieces.

[John B.]

On Sun, 25 Sep 2016 13:32:28 -0500, DougC <dci...@norcom2000.com>
wrote:

Wrap the cable on a horizontal drum and pull it off keeping everything
in the same plane. i.e. a horizontal drum and a horizontal wheel or
tire.

Note that normal steel cable, due to its twisted construction, tends
to twist when tension is applied to it. There is a non-twist cable
made for special purposes.

Steel cable also has a sort of built in bend due to having been rolled
onto a drum for handling upon manufacturer and stored that way.

This is usually handled as I mention above by storing the cable on a
drum and not inducing any twist during use. From the top of one drum
to the top of another drum, or pulley, for example.

Another method of handling this "normal twist" is to lay the loose
cable out on the flow in a straight line and do not fasten the end.
Then although winding the cable will induce a twist but as the cable
is free to turn it will not kink.

see: http://tinyurl.com/h7u3fcn
which will download a pdf titled "08_oper_char_ropes_cables.pdf"

Quote:
Even if a cable has been designed to have good torque balance, it
may still develop some torsional energy if it has experienced any
twisting. Induced twist can occur during the lowering or raising of a
nonsymmetrical payload, by maneuvering of a tethered vehicle so as to
accumulate turns in the cable, or by the cable handling techniques.
For example, if a cable is deployed manually and is allowed to pull
out of a coil which is lying on the deck, it will develop one turn of
twist for each wrap in the coil. Similarly, a cable handling system
which does not incorporate a drum, but which allows the cable to lie
in a cage or basket, will produce one complete twist of the cable for
each loop of cable in the basket. Depending on the diameter of the
cable and on its inherent torsional stiffness, this twisting may be
sufficient to produce a hockle if a slack loop should be allowed to
form.
Unquote
--
cheers,

John B.

[DougC]

On 9/25/2016 7:25 PM, John B. wrote:
...

> Wrap the cable on a horizontal drum and pull it off keeping everything
> in the same plane. i.e. a horizontal drum and a horizontal wheel or
> tire.
>

That would work but would have other issues I think.
The figure-8 bucket is really cheapest and easiest to try first.

> Note that normal steel cable, due to its twisted construction, tends
> to twist when tension is applied to it. There is a non-twist cable
> made for special purposes.
>

I am using the smallest size of stainless twist cable commonly available
(.8mm). Lots of places make and sell it, but it's all the same for the
two strand types (1x7 and 7x7).

The only kinds of non-twist cable I ever found were ones with different
layers wound different directions. This is too small for that, since it
is only 7x7 strand.

I wondered for a bit if I should try to just get the single-strand wire
used for this tiny cable, and try that--but no place sells it at a
useful price it seems. At least, not in relatively small quantities. You
can buy tiny little 100-yard spools easily, but I would want spools a
thousand times longer, at least.

> Steel cable also has a sort of built in bend due to having been rolled
> onto a drum for handling upon manufacturer and stored that way.
>

This stuff doesn't, but then it isn't bent much for as thin as it is and
the spool that it comes on.

> Another method of handling this "normal twist" is to lay the loose
> cable out on the flow in a straight line and do not fasten the end.
> Then although winding the cable will induce a twist but as the cable
> is free to turn it will not kink.
>

That aint gonna work for a tiny thin piece so long. And I've not got the
room to do it anyway.

> .... Depending on the diameter of the

> cable and on its inherent torsional stiffness, this twisting may be
> sufficient to produce a hockle if a slack loop should be allowed to

> form. ...
I did not know the term for this (hockle). Thank you.
-------
Damn the hockles.

[Tosspot]

On 25/09/16 22:08, DougC wrote:
> On 9/25/2016 2:01 PM, Tosspot wrote:
>>
>> Didn't you have a web page up with a blog. If it's up to date I'd like
>> a gander.
>>
>>
> Still there but not updated since Sept 2015.
> http://www.norcom2000.com/users/dcimper/assorted/inanities/recumbent/tire_making/tire_making_main.html

Did I read that correctly, you set fire to your tire making machine!?

<impressed>

[John B.]

On Sun, 25 Sep 2016 21:24:53 -0500, DougC <dci...@norcom2000.com>
wrote:

>On 9/25/2016 7:25 PM, John B. wrote:
>...
>> Wrap the cable on a horizontal drum and pull it off keeping everything
>> in the same plane. i.e. a horizontal drum and a horizontal wheel or
>> tire.
>>
>That would work but would have other issues I think.
>The figure-8 bucket is really cheapest and easiest to try first.
>
>> Note that normal steel cable, due to its twisted construction, tends
>> to twist when tension is applied to it. There is a non-twist cable
>> made for special purposes.
>>
>I am using the smallest size of stainless twist cable commonly available
>(.8mm). Lots of places make and sell it, but it's all the same for the
>two strand types (1x7 and 7x7).

That is normal wire rope construction.

>The only kinds of non-twist cable I ever found were ones with different
>layers wound different directions. This is too small for that, since it
>is only 7x7 strand.
>
>I wondered for a bit if I should try to just get the single-strand wire
>used for this tiny cable, and try that--but no place sells it at a
>useful price it seems. At least, not in relatively small quantities. You
>can buy tiny little 100-yard spools easily, but I would want spools a
>thousand times longer, at least.

It doesn't make much difference what you use. If you coil something up
and lay it on its side and then pull it off it will form loops that
will tangle.

>> Steel cable also has a sort of built in bend due to having been rolled
>> onto a drum for handling upon manufacturer and stored that way.
>>
>This stuff doesn't, but then it isn't bent much for as thin as it is and
>the spool that it comes on.
>
>> Another method of handling this "normal twist" is to lay the loose
>> cable out on the flow in a straight line and do not fasten the end.
>> Then although winding the cable will induce a twist but as the cable
>> is free to turn it will not kink.
>>
>That aint gonna work for a tiny thin piece so long. And I've not got the
>room to do it anyway.
>
>> .... Depending on the diameter of the
>> cable and on its inherent torsional stiffness, this twisting may be
>> sufficient to produce a hockle if a slack loop should be allowed to
>> form. ...
>I did not know the term for this (hockle). Thank you.
>-------
>Damn the hockles.

"Hockles" is a term commonly used by those that work with wire rope.

hockle, (plural hockles) A knob in cordage caused by twisting against
the lay.
Photo:
http://charles.hamel.free.fr/knots-and-cordages/img/super-coil.jpg

--
cheers,

John B.

[DougC]

On 9/25/2016 11:58 PM, Tosspot wrote:

>
> Did I read that correctly, you set fire to your tire making machine!?
>
> <impressed>
>

It seems like I explained that bit before:

I have a 4' x 1' x 1' metal box with a heating element inside, that I
use to mold strips of rubber for tire tread. The first version was
(electrically) just a 1200-watt heating element connected to a wall
cord. I'd plug it in for a minute or so while checking the temperature
with a separate thermometer to keep the box surface at 125°F or so. The
heating element was rated for 1500°F, so it will get WAY hotter than
that if you leave it plugged in long enough. I knew that, but this
(heater + cord setup) was quick and easy to build and I figured I'd
always be watching it when it was running anyway.

One day I plugged in the heater box, when I thought I was plugging in
something else entirely. After a couple minutes the heater box began to
smoke and pop. It just burned up some pieces of lamp cord inside the
box, nothing else got hurt--but then I built an Arduino setup to
automatically run the heater to prevent this again.

On the Arduino you have to set a temperature and a time limit and
[start] it, it maintains the temperature for the set amount of time, and
then it shuts off automatically at the end. The Arduino turns on but the
heater doesn't turn on at all when you just plug the whole thing in, and
you can't tell it to stay on indefinitely either.

Plus it just maintains the temperature better--it heats up as quick as
possible, and then checks the temp every 15 seconds or so and runs the
heater as needed.

[Gregory Sutter]

On 2016-09-25, DougC <dci...@norcom2000.com> wrote:
> Episode #I forget....
>
> Still going on the tire-making project. The last few days I've found out
> how hard it is to coil a long piece of twisted cable without adding
> additional twist.

Doug,

Can you over-under coil the cable? That would avoid creating
additional twist as the machine draws it in. (It's the topological
equivalent of a figure-8, but once coiled, you can feed the machine
from a single bucket.)

Here's a video showing how:
https://www.wired.com/2013/08/tnhyut-coil-cable/

Cheers, and thanks for keeping us updated on your interesting project.

--
Gregory S. Sutter Mostly Harmless
mailto:gsu...@zer0.org
http://zer0.org/~gsutter/

[DougC]

On 9/27/2016 12:36 PM, Gregory Sutter wrote:
> On 2016-09-25, DougC <dci...@norcom2000.com> wrote:
>> Episode #I forget....
>>
>> Still going on the tire-making project. The last few days I've found out
>> how hard it is to coil a long piece of twisted cable without adding
>> additional twist.
>
> Doug,
>
> Can you over-under coil the cable? That would avoid creating
> additional twist as the machine draws it in. (It's the topological
> equivalent of a figure-8, but once coiled, you can feed the machine
> from a single bucket.)
>
> Here's a video showing how:
> https://www.wired.com/2013/08/tnhyut-coil-cable/
>
> Cheers, and thanks for keeping us updated on your interesting project.
>

The problem I'm seeing is that the wire has its own occasional random
twist in it, and it won't lie flat at all unless that twist is removed.

I made a couple of the figure-8 bucket things but getting the wire to
lay in them properly is too time consuming as it is. I may try one more
thing before abandoning that method tho. A couple tall funnels fastened
pointing up in the middle of each bucket should help speed it up a lot.

I can just build automatic feeder and take-up reels too, if I have to. I
would not have thought that this would require it, but maybe it does.

[DougC]

On 9/27/2016 12:36 PM, Gregory Sutter wrote:

> Can you over-under coil the cable? That would avoid creating
> additional twist as the machine draws it in. (It's the topological
> equivalent of a figure-8, but once coiled, you can feed the machine
> from a single bucket.)
>
> Here's a video showing how:
> https://www.wired.com/2013/08/tnhyut-coil-cable/
>

I tried doing this today, and it doesn't work.
It SHOULD work, but it doesn't.
The cable has tiny burrs on it or something, if you have loops lying on
each other and start to draw them off, they will not slide apart from
each other cleanly. They end up pulling each other into a tangle.

I'll have to build the feeding and take-up spools, because the only way
to keep this stuff from tangling in long lengths is to keep it wound
around a spool all the time.

[Gregory Sutter]

On 2016-09-28, DougC <dci...@norcom2000.com> wrote:
> On 9/27/2016 12:36 PM, Gregory Sutter wrote:
>
>> Can you over-under coil the cable? That would avoid creating
>> additional twist as the machine draws it in. (It's the topological
>> equivalent of a figure-8, but once coiled, you can feed the machine
>> from a single bucket.)
>

> I tried doing this today, and it doesn't work.
> It SHOULD work, but it doesn't.
> The cable has tiny burrs on it or something, if you have loops lying on
> each other and start to draw them off, they will not slide apart from
> each other cleanly. They end up pulling each other into a tangle.

Hmm, that's too bad. I can't imagine how expensive your tire beads
would be if you had to use polished wires, though:

https://www.universalcycles.com/shopping/product_details.php?id=73532

Yes that is $26 for a single 1700mm inner cable. Suitable for
sponsored riders only?