Tag Archives: double drive

Doing the impossible – a double drive puzzle

(updated 28 January)

First, an acknowledgement – Shan Wong and Lorraine Cross have been very helpful with his project. It wouldn’t have got anywhere without them. So please consider them co-authors.
Mary Knox

If someone tells you something can’t be done, without a clear and convincing reason why, do you immediately want to try it? I do, specially if it’s a spinning something.

Some friends and I have been experimenting.

Double drive is a system in which a single long drive band in a figure-of-eight goes around the drive wheel twice, and once each around the flyer whorl and the bobbin whorl. It’s not used as often in New Zealand as in many other countries, but it’s a lovely spinning setup, smooth and fast. This Rappard Mitzi in double drive is Shan’s current favourite.

Double drive winds the yarn on because the bobbin has a smaller-circumference groove on its end than the groove on the flyer whorl. That means when you let the twisted yarn feed in, the bobbin turns faster than the flyer (think gears on a bicycle) so yarn is wrapped around the bobbin.
When you stop feeding in and hold the yarn while twist accumulates, the flyer and bobbin become ‘locked’ together and turn as one (this is called yarn lock).

Why is it always done that way round? Why not have the bobbin groove bigger and the flyer groove smaller? We’ll call that, for convenience, Reverse double drive (RDD). If it’s the difference between the two groove circumferences that matters, surely it would work either way? Apparently not – most authorities say a correct double drive wheel has a smaller bobbin groove than its flyer whorl groove.1

Eric Corran calls RDD ‘inoperable’ and is the only writer we’ve found who tries to explain why.2 Shan and I both find his explanation unconvincing, partly because he fails to take into account slippage of bobbin and flyer and the build-up of twist during yarn lock.

Alden Amos is more balanced, saying it can be done but there’s no good reason to.3 Curiosity was sufficient reason for us. My Ron Shearman wheel has bobbins with several different groove sizes and the smallest groove on the flyer is quite small. I assembled an RDD setup, grabbed some horrible wool and tried spinning. The drive band wouldn’t stay on … the wheel was obviously trying to tell me something, but what? Eventually there were a few metres of awful ‘yarn’ which I’m not going to show you.

So I made a really fine drive band, some sort of embroidery cotton I think, and found some nicer wool.
The new drive band stayed on and I actually made a little yarn, on two bobbins so I could try plying later. It wasn’t easy – takeup was mostly very slow and jerky. The wheel would take the yarn when it felt like it but didn’t want to be fed. It was  better after I figured out that a short backward draft worked best.

Shan had been experimenting too, at first with little success (the yarn kept unwinding from the bobbin). We agreed that we should try a bigger differential between bobbin and flyer whorl sizes. I had been using a bobbin groove circumference of 17cm and a flyer groove circumference of 15cm, giving a ratio of 1:1.13 – definitely on the low side. No wonder there was poor takeup and overtwist.

We both made changes. Shan put a bobbin with a 22cm groove on her Mitzi (flyer whorl circumference 14.5cm) for a ratio of just over 1:1.5, which should be very adequate. She commented ‘There’s a funny sensation that translates to “bobbin takes when it wants to”. I can feel an occasional tug, which for me means the bobbin says, “Give me the yarn!” The sweet spot is very narrow and if the drive band is slightly too loose, it falls off.’

I discovered that the bobbins of my Gib Wilson wheel could be used on the Shearman wheel, though they rattled a bit. Their bobbin groove circumference is 20cm, and now the flyer to bobbin ratio was 1:1.33. That was so much better that I could even do a gentle long draw, until I got a little over-excited and oops…
(The problem is clear: the bobbin is too short for the flyer and has drifted so that the flyer ends hit the drive band. That was easily solved with a piece of thick felt from an unsuccessful effort in a workshop, sandwiched between two washers – you can see it below, in front of the bobbin.) Soon I was spinning again, not loving it because takeup was still somewhat erratic, but not hating it either. This time I spun the second little batch on a partly-filled bobbin, which didn’t change anything.
Meanwhile I’d emailed Lorraine in Australia about what we were doing, and she happily turned to a new (to her) Schacht Ladybug which has a good choice of whorl sizes. First she spun a little in normal double drive, and then reversed the bobbin to use its wider-circumference end. Her flyer whorl groove is 13cm around; the normal double drive end of the bobbin has a 10cm circumference groove and the other end (intended for spinning in scotch tension) has 17cm. So in normal double drive (on the left below) she had a ratio of 1:1.3 between the grooves, and in RDD (on the right) roughly the same. You can see the difference in spinning quality.
Lorraine wrote that in RDD ‘this very easy, responsive wheel was hard to restart each time I stopped spinning, but once it got going, it actually filled the bobbin pretty well. Most obviously, it was very hard to get enough twist – this fibre to the right of the photo is very underspun by my standards, even after a lot of tweaking. It was hard to slow the pull enough to allow sufficient twist – a very fine line between slowing the wheel/flyer down enough, and the wheel stopping … It certainly is possible to spin DD the wrong way round … but it was not really pleasurable.’

Now it was time to tackle plying. On my 1:1.33 RDD setup I plied each of my pairs of experimental spinning, in my normal fashion with back hand at my hip controlling the threads and front hand feeding in. I expected woe and misery, and was astonished that it worked OK.
In the top skein, the overspinning of the singles is noticeable, with poor takeup at the ratio of only 1:1.13 between the flyer and bobbin grooves. For a close inspection, as always, click on the photo. The bottom skein wouldn’t win any prizes, but it’s quite a bit better. Singles in skein 2 and all plying were at a ratio of 1:1.33.

Lorraine also tried plying and initially found it very hard to control. Later she wrote ‘With a bit more practice, both spinning and plying results were quite acceptable – apart from the unusual starting process, lesser twist, and periodic jerking of the fibre.’

So why is RDD takeup troublesome? Here is one theory, which owes quite a lot to Alden Amos.4

In yarn lock, when twist is being added but no winding on is happening, flyer and bobbin are held together by the yarn between them and turn at the same speed, relatively slowly with the drive band slipping a certain amount:
When the spinner starts to let the yarn wind on (breaking yarn lock) the bobbin, with only minimal friction between it and the well-polished flyer shaft, can speed up quickly. I wondered if, because of several factors, the flyer might take longer to get up to speed and start winding the yarn on.

First, the flyer has more weight to get moving – arms, hooks, orifice, whorl, spindle (though Shan weighed her two flyer-and-whorl combinations and found the difference to be 30gms, not enough to explain their different behaviour). Second, there’s more friction to overcome (the bearings at each end of the shaft, the friction of the yarn passing over the hooks). Third is air resistance slowing the flyer arms. And fourth, the drive band has less contact with the flyer whorl groove than it does with the larger bobbin groove, so slipping might take longer to overcome.
Could the flyer (now left behind by the bobbin) pull a little of the already-wound yarn off the bobbin before it finishes speeding up and normality is resumed? In this moment does the opposite of winding on take place, and ‘spit out’ a little of the wound-on yarn?

Lorraine, who says she’s spun on ‘far too many different, often cantankerous wheels’ mentions that she has learned to compensate for deficiencies, and enjoys getting the best out of any wheel. ‘After that first tendency for the bobbin to pull too soon, I was automatically/subconsciously starting the wheel using the treadles, and simultaneously flicking the sluggish flyer arm into action. So I had already unwittingly prevented this particular problem you had detected. Restarting the wheel using only treadles confirmed your observation.’

Shan on the other hand found that jerkiness was more likely to happen after a bit of winding on, and ‘Too loose tension results in no wind on, and too much results in strong jerk.’ Her interpretation is that the flyer winding on alters the bobbin rotation during takeup.

In any case, we now have tangible evidence that spinning with RDD is not always impossible. ‘Not always’ because I have read comments by a couple of highly knowledgeable spinners, whose expertise I respect, who when confronted with an RDD setup have found they could get no takeup at all no matter what they did. And if they say that at that time on that wheel it was impossible, I totally believe them.

So what was the difference? Could it possibly be to do with the shape of the grooves in flyer and bobbin? There is a widespread belief that for double drive, the groove in the bobbin should be U-shaped allowing more slippage, and the groove in the flyer whorl should be V-shaped to grip the drive band more.5
New Zealand makers have never bothered with this apparent necessity, and Shan’s wheels and mine all have fairly similar V-shaped grooves on flyer whorl and bobbin. Not everyone agrees that it matters, but what if it does, just a bit? In RDD, could the difference between groove shapes, now doing the opposite of what it was supposed to do, be just enough to transform RDD from unsatisfactory to impossible? Well no, there must be more to it than that, because Lorraine’s Ladybug wheel has a flat bottomed, wide groove on the bobbin whorl, and a V-shaped flyer whorl groove. She didn’t find it a problem.

We still don’t know for certain the exact reasons for the behaviour of our wheels in RDD. Subtle differences between wheels, and between spinners, certainly play a part. But this we can say –
We found that to spin in RDD isn’t too hard as long as the bobbin whorl is substantially greater than the flyer whorl for sufficient winding on. That part is definitely conclusively true. The operational range of tension too is much smaller than spinning double drive: it needs to be on the tight side.

So if you are stuck for bobbins, then by all means put on a large groove bobbin and give it a whorl! We’d love to hear how you get on. But now, for us, normal (not reverse) double drive will resume.

Quick update, 28 January:
Lorraine doesn’t give up easily. She writes ‘I tried again, using some lightly washed Romney fleece (far more enjoyable to spin than the dirty-very-fly-away-alpaca). I spun quite successfully until I had absolutely no takeup as the bobbin filled.
I then even tried changing to a larger whorl (close to the bobbin whorl size) to increase the band tension further, and to see what effect that might have… it actually very briefly helped with bobbin filling, but I have now given up. As you can see, with practice, and increasingly as the takeup reduced, I also managed to achieve more twist in the fibre.’
There are many ‘Why?’ questions still unanswered! But at least we have a better idea of ‘What happens if …’

1. For example David Bryant (p.44), Judith Buxton (pp.109-115), Eliza Leadbeater (p.6), Mabel Ross (p.60).
2. Page 47
3. Page 218 note 3
4. Pages 218-220
5. Authorities who state this include Amos (p.226), and Corran (p.109). Buxton (p.113) thinks it’s important but should be the other way round. Mabel Ross on the other hand says the difference in groove shape is traditional rather than necessary (p.60) and David Bryant doesn’t show such a difference in any of his plans.

Alden Amos The Alden Amos Big Book of Handspinning (Interweave Press 2001)
David Bryant Wheels and Looms (Batsford, London 1987)
Judith Buxton (or Buxton-Keenlyside) Selected Canadian Spinning Wheels in Perspective (National Museums of Canada 1980)
Eric Corran Understanding the Spinning Wheel (self published Australia 1997)
Eliza Leadbeater Spinning and Spinning Wheels (Shire Publications, no date)
Mabel Ross Encyclopedia of Handspinning (Interweave Press 1988)

Drive bands

Here  is the section on fitting a double drive band that was missing from last month’s Under Tension. I’ve added a few other thoughts too – they are full of “might” and “maybe” for a reason. Please don’t take any of those thoughts as more than suggestions and opinions. My golden rule of spinning is “If it works, it’s not wrong!”

Tension – but no stress ?

The next article for you, from a couple of years ago, was inspired by realising that a lot of spinners in New Zealand don’t know what kind of tension they are using (here, it’s practically always scotch tension). They would often be afraid to try any other system (particularly double drive). In fact they may have little understanding of how their system actually works.

When it was published, several people told me they disagreed with the first sentence: “We all know that our spinning wheels work because the flyer twists the yarn and then winds it on the bobbin.” It was the “we all know” part that they disagreed with.

I don’t know whether any spinners are better informed after reading it, but I like to think that some might be.

The article included, in a sidebar, a brief explanation of how to tie a double drive band. I’ll put that up separately a little later, and I’ll add a bit about string and knots and other drive band considerations.