Tuesday, June 11, 2013

Twist

Twist is inserted when one end of a string is rotated relative to the other end of the string.

The string is tied to a spindle, the spindle goes round, the spinner holds the other end of the string, and twist is inserted.

String is tied to the spindle on a great wheel, the wheel/spindle goes round, the spinner holds other end, and twist is inserted. A bobbin on the spindle makes no difference.  If the string is tied to the bobbin, the bobbin goes round and twist is inserted when the spinner holds the other end.

On a flyer/ bobbin assembly, the string is tied to the bobbin.  When the spinner holds one end of the string, then twist is inserted when the bobbin goes round.  The flyer is not tied to the string. The flyer cannot insert twist unless the bobbin rotates.  The bobbin and flyer may rotate together, but it is the bobbin that is fixed to the end of the string; and therefor,

it is the bobbin that inserts twist.  

If the flyer rotates, but the bobbin does not, then string is wound onto the bobbin, but no net twist is inserted.

(Some small amount of twist will be inserted, but this is NOT enugh to hold the yarn together, so no actual yarn is formed. If no yarn is formed, then the bobbin is simply rotating in space, and there is no yarn to accept the twist, and "spinning" does not occur.)

If the bobbin rotates but the flier does not, then twist will be inserted, at the rate of one twist per effective circumference of the bobbin.  One twist per per effective circumference of the bobbin is not likely enough twist to make a competent yarn.  It may not be enough to even hold a soft roving together.

Confusion comes when the rotation of the flyer causes the bobbin to rotate (via yarn lock.)  In this case, the flyer drives the rotation of the bobbin, but it is the rotation of the end of the string tied to the bobbin that causes the insertion of the twist. If rather than a mass of fiber, filaments are passed through the flyer, then it can insert some small amount of twist. However, with normal fiber this is not enough twist to form a yarn, so there is no yarn to accept the twist.

One end of the string is held by the spinner, and the other end of the string is tied to the bobbin.   Only rotation of the bobbin or rotation of the spinner (or the counter rotation of both) can insert enough  net twist to form yarn from fiber.

Twist is the essence of spinning.  If one does not understand twist, then one does not understand spinning.

11 comments:

Anonymous said...

For someone who self-admittedly has written numerous reports and high-level documents, your spelling is pretty atrocious. Therefor is spelled therefore, by the way. In today's English. Maybe that's how it was spelled in all of the spinning books from the 1800's that you read. You try to act so learned and intelligent, but most grade-schoolers can spell better than you. If you care about the subject about which you are writing, how about proofreading your diatribes? Poor spelling renders them laughable.

LeslieJM64 said...

Okay, let's try this explanation.

I've re-read Priestman's Principles of Worsted Spinning. He doesn't really go into the workings of what he calls "Saxon wheels", but he does examine several different set-ups for industrial spinning. He makes it quite clear that in "fly spinning", the flyer acts as a crank which inserts the twist. In fact, in this set-up, the bobbins are not even driven. They are caused to rotate on the spindle shaft because of friction on their bearings, and they stand on cloth or felt washers which provide a braking action.

He does say that the only significant difference between the fly spindle in an industrial setting and the one on the Saxon wheel is that the wheel lacks a traverse motion for evenly filling the bobbin.

All in all, I think that Priestman agrees with me, so if you feel that he "got it right"...

Aaron said...


The bobbins are caused to rotate --

The bobbins are in fact, "driven" by "yarn lock". It is just like a Scotch Tension Saxon wheel.

His language in this case is imprecise, because he is using engineering terms and you are understanding the terms in their general meanings.

If the bobbin does not rotate, Then the total twist inserted is one twist per effective circumference of the bobbin. I use bobbins with a 5/8" diameter which would give one twist inserted every 2 inches. That is about right for a very soft roveing, and you have larger bobbin diameters, and thus would get less twist. When our bobbin are full, we would be getting 1 twist in 9 inches. In comparison singles for 2-ply worsed need ~ 4 tpi and my 5,600 ypp singles need 9 tpi.


Try it! Get out your ST wheel, set the ST brake so tight that the bobbin does not rotate at all, and try to spin. It does not work. Never! Never! Never! You are so accustomed to having the bobbin rotate that you have not tried locking the bobbin because you know that does not work. But, that is the very point. It is the rotation of the bobbin that inserts useful twist and allows spinning.

Lock the bobbin so it cannot rotate, and you will end up with a flyer/bobbin assembly under a pile of loose fiber as the flyer pulls the roving/sliver apart and bats it to pieces.

In this spinning equipment, the bobbins do rotate. Some slippage of the bobbin is required because as effective diameter of the bobbin increases as it fills, there is a change in the required differential rotation speed (DRS). Hence the bobbins are not "driven" by a belt or gear. Instead they are caused to rotate with a certain amount of slip by yarn lock, just like your ST wheel.

LeslieJM64 said...

I’m trying really hard not to be insulted by your assumption that I either a) don’t know how to read, b) am too ignorant to understand what I’m reading, or c) have not actually looked at or experimented with my spinning wheel.
In your original post you said: “If the flyer rotates, but the bobbin does not, then string is wound onto the bobbin, but no net twist is inserted. If the bobbin rotates but the flier does not, then twist will be inserted, at the rate of one twist per effective circumference of the bobbin. One twist per per effective circumference of the bobbin is not likely enough twist to make a competent yarn. It may not be enough to even hold a soft roving together.”
I repeated my first experiments (the ones I had documented in a Ravelry thread, as you well know). If the flyer rotates but the bobbin does not, I get one twist per rotation of the flyer (also, there is wind-on as the flyer wraps the yarn around the bobbin, which is the entire point of braking the bobbin in a scotch tension system).
If the bobbin turns, but the flyer does not, there is no twist added. None.
In your most recent comment you said: “If the bobbin does not rotate, Then the total twist inserted is one twist per effective circumference of the bobbin. “
This is what I’ve been saying all along, and it is exactly the opposite of what you said in your post. It is also true, but it is true because the wind-on allows that much twist, not because the bobbin is responsible for twist insertion. I did an experiment (again) with two yarns of contrasting colors. I tied them to the bobbin, and measured the length of yarn between where it was tied to the bobbin and the eye of the orifice. This leader portion was 10 inches. Then I turned the flyer 10 times while the bobbin was immobilized. I pulled the yarn back off the bobbin and measured—I had 10 twists in 35 inches of yarn. If I eliminate the leader portion (because my assertion is that this amount was wound onto the bobbin before the flyer was able to insert any twist into it), I get 10 twists in 25 inches. Guess what the circumference of my bobbin is? Yep, 2.5 inches. Oh, and I did move the yarn down the hooks so that the effective circumference of the bobbin was unchanged during my experiment.
So, the flyer inserted 1 twist with each rotation, and it wrapped 2.5 inches of yarn around the bobbin with each rotation, but it was only able to insert twist into the yarn that was on my side of the orifice. That 10 inch leader portion did not get any twist at all.
If the bobbin was not immobilized, it would be caused to rotate by friction of its bearings on the flyer shaft. When yarn lock is achieved, the bobbin effectively disappears. In other words, when the flyer and bobbin are rotating at the same speed, the system behaves no differently than the spindle that the flyer assembly is derived from. If you were to turn a top-whorl spindle on its side, would you say that the twist was inserted at the hook, or that it was inserted where the yarn is tied to the spindle shaft?

Aaron said...

It takes more than 10 twists in 25 inches to make competent yarn.

Put only 10 twists into 25 inches of yarn, and all you have is a pile of fiber.

That is not spinning, (in the sense of making yarn).

The goal here is to understand how to spin competent yarn.

Twist is inserted when the yarn crosses the axis of rotation.

LeslieJM64 said...

"It takes more than 10 twists in 25 inches to make competent yarn.

Put only 10 twists into 25 inches of yarn, and all you have is a pile of fiber.

That is not spinning, (in the sense of making yarn)."

Seriously? What does this have to do with what I said? I'm fairly certain that you understood precisely what my experiment was intended to illustrate. If you don't, in point of fact, understand what I'm saying I can direct you to some photographs which demonstrate how the experiment was conducted and why I draw the conclusions I have drawn.

“The goal here is to understand how to spin competent yarn.”

No, the goal of this discussion is to demonstrate where twist is created on a flyer wheel. I’m going to take your refusal to address the points I’ve made, coupled with your previously noted contradictions, as an admission that you know that I am correct.

Anonymous said...

Hi Aaron,

Great blog post! I very much enjoyed reading this post.

I think LeslieJM64, what Aaron is trying to say is that it is the bobbin which allows the yarn to be twisted in the first place.

It actually stands to reason. Imagine threading the spinning wheel without a bobbin... all you have is a spinning flyer and a leader sitting idly not being twisted at all. Add a bobbin, and you'll get twist insertion. the bobbin is associated physically to the flyer by the yarn, and Aaron is right when he says the yarn lock is what allows twist insertion. Without it you'd just be back to the initial situation, so yes, whilst the flyer does spin and provide the initial rotation, it is the bobbin that enables the rotation to be transferred to the yarn in the form of twist.
It's physics. if you think about it like that, it does work.

great blog Aaron please keep posting! I really enjoy reading what you write!

Aaron said...

Twist is not inserted when an object rotates in space. Twist is inserted into yarn or string or fiber when there is a yarn or fiber, and one end is rotating relative to the other end of the yarn. If there is no yarn, then an object rotating in space is not inserting twist.

In spinning, if there is not enough twist to form a yarn, then there is no yarn, and the flyer/bobbin assembly is simply rotating in space. Without yarn attached to the rotating flyer/bobbin assembly there is no point of twist insertion.

LeslieJM64 said...

Both ends of the yarn need to be secured in order for twist to be formed---this is true. In the flyer/bobbin array, one end of the yarn is secured to the bobbin---this is also true. It is true, but it is not the whole story.

It is also true that twist will be inserted quite nicely if the bobbin is removed and the non-working end of the yarn is secured to the flyer shaft, the flyer, after all, being simply a variation on a spindle. Securing the yarn to the bobbin allows for the tension that holds the yarn against the eye of the orifice--WHERE THE TWIST IS INSERTED. The flyer is analygous to a high whorl drop spindle in this respect: the yarn must be secured to the shaft of the spindle, but it is clear that the twist is formed at the hook of the spindle. When the tension is released by the spinner, the bobbin will spin (for lack of a better term) independently of the flyer shaft, and the differential between the bobbin's speed and that of the flyer will allow the yarn to wind on (and this is true regardless of the drive system of the wheel).

You have repeatedly asserted that the rotation of the bobbin is what creates twist. You have gone as far as to say that rotating the bobbin without rotating the flyer will create twist (though you did contradict yourself later and say the exact opposite of this). This is simply incorrect. The rotation of the bobbin INDEPENDENT OF THE FLYER will NEVER create twist. It is not physically possible, due to the angle at which the yarn meets the bobbin. Twist is created during yarn lock, when the bobbin is effectively eliminated (i.e., the rotation of the bobbin is the same as the rotation of the flyer shaft).

All of this is easily observable. The logical progession from spindle, to spindle wheel, to flyer wheel is fascinating and quite clever, but at the heart of it, even on a flyer wheel we are still spinning off the point of a horizontal driven spindle. The flyer/bobbin array just eliminates the extra step of winding on.

Anonymous said...

No. Now I see the problem. Aaron is often a little unusual in the way he uses common terms.

Aaron, in the way most people use the words, the bobbin does not INSERT twist. The bobbin ANCHORS the yarn to enable twisting and STORES twist once it's been added. Inserting, for most, describes the action of twisting - and that's definitely the flyer.

Maybe this will help.

Aaron said...

In a typical flyer/bobbin assembly, the flyer will insert on the order of 0.5 tpi to a preexisting string connected from the bobbin to the spinner. So, by a strict physics definition, that IS inserting twist.

However, a typical modern fat yarn yarn needs 4 tpi to become competent, so the flyer is inserting 12% of the required twist for actual spinning & 88% of the required twist comes from the rotation of the bobbin.

My 5,600 ypp singles require 9 tpi, so the flyer is inserting 5% of the twist. When I am spinning fines, the flyer inserts much less than 1% of the total twist in the yarn.

From an engineering standpoint, the twist inserted by the flyer is not adequate to form a competent yarn. It there is no yarn for the twist to be inserted into, then no twist is inserted. Thus, by a strict physics definition if no yarn is being formed between the bobbin and the spinner, then no twist is being inserted or accumulated and spinning is not occurring. Without the rotation of the bobbin, "spinning" will not occur.

If you depend on the twist inserted by the flyer, you cannot spin wool, cotton, or linen. The best you could do is twist silk filaments. If you want to limit the wheel to twisting silk, then I would agree that you can use a flyer to insert twist.

On the other hand, as you calculate DRS ratios, you need to consider the twist inserted by the flyer or your design grist will be off. See AA.