Sunday, October 13, 2013

DRS Revisited

Differential Rotation Speed (DRS) is a way of setting up a double drive spinning wheel.  I first came across it Alden Amos, Big Book of Handspinning.  It is also discussed in the introductory chapters in several of the books on industrial spinning published early in the 20th century but the technology is given other names.  Some of these are free on Google Books.  There is also some material and pictures from earlier in this blog.

Normally, in modern double drive spinning wheels, there is a 1:1.2 ratio between the flier whorl and the bobbin whorl. The drive band is kept loose enough to allow some slip and whorl profiles facilitate this slip.  "Yarn lock" causes the flyer and bobbin to rotate at the same speed. During yarn lock the drive band slips on the bobbin whorl.  When yarn lock is released, the bobbin speeds up and yarn is rapidly wound on to the bobbin. This set up allows the drive band to transfer power to the flyer/bobbin assembly through two whorls and allows slightly more power transfer, thereby driving the flyer bobbin assembly at speeds significantly higher than Scotch Tension. Thus, this arrangement is favored by lace spinners that need to insert a lot of twist.  However, total power transfer and thus top speed is limited by drive band slip. Moreover, continuous drive band slip causes wear on the drive band and the whorls.

In this arrangement, rapid take-up is always available and the spinners' tension on the yarn determines whether take-up occurs.  The feel of such a wheel is very like that of Scotch Tension or Irish Tension.  And, almost any yarn can be spun with very minor adjustments in the drive belt tension.

With DRS, the ratio of take up to inserted twist it controlled by the ratio between flier whorl and the bobbin whorl. This ratio is selected to produce a particular yarn.  The drive band tension and whorl profiles are selected to limit drive band slip, and the differential rotation speed between the flier whorl, the bobbin whorl with the effective diameter of the bobbin controlling the amount of inserted twist. If the spinner wants to produce a different yarn, then whorls with a different ratio must be selected or the effective diameter of the bobbin changed.  As the bobbin fills up the effective diameter of the bobbin changes, and it tends to wind yarn on faster, thereby reducing inserted twist. (With high grist yarns, this effect is trivial.)  Then, yarn must either be wound off, or a different whorl ratio selected, or a thicker/ lower twist yarn will be produced as the bobbin fills.  When producing low grist yarns, the effective diameter of the bobbin changes rapidly and the DRS approach is not useful.

With DRS, if the spinner drafts too thick for the set whorl ratio, the yarn breaks off.  If the spinner drafts too thin for the set whorl ratio, the yarn drifts apart.  If the yarn winds on, it is good competent yarn of the grist set by the effective diameter and the ratio of the whorls.   The system will not spin anything else, and  the spinner that tries to spin anything else will be frustrated.  No drive band slip is required, so power transfer by the dive band to the flier/bobbin assembly can be much greater and the flier/bobbin assembly can be driven at speeds 2 or 4 times greater than typical modern modern double drive spinning wheel.

The feel of the DRS  wheel is very different. Techniques such as long draw, that use twist accumulation followed by rapid take up do not work.  Both Alden Amos and Henry Clems have made such wheels, and all of those wheels were returned as "non-functional".  I agree that the approach is not useful for a recreational spinner that spins small batches of yarn.  And, here a small batch of yarn is 25,000 yards of the same grist single.

The concepts starts to be useful at grists of  ~ 5,600 ypp (9 tpi). By using a flier whorl with 4 different drive band slots allowing 4 different flier/bobbin differential rotation speeds, and bobbin length of just under 4" (Ashford) a full hank (560 yd/ 1.6 oz) can be spun with a twist/grist variance of  just about 10%, which I am told is not bad.  Spinning 11,200 ypp singles again using a flier whorl with 4 different drive band slots the twist/grist variance can be much less. By the time you get to 40s (22,400 ypp) The change in effective bobbin diameter for a hank (0.4 oz) is so small that a flier whorl with a single drive band slot can be used to produce a full hank with minimum twist/grist variance.

Sustained 2,400 rpm is as about as fast as a great wheel with an accelerator can insert twist.  However, with a flier/bobbin assembly both hands can be used to draft, and thereby produce worsted yarn.  Since, worsted yarn required 5 -10% less twist than woolen the  flier/bobbin assembly can produce  5 -10% more yarn at the same average rate of twist insertion.  Thus, I am no longer seeking a great wheel.  Also, I have made a driven spindle assembly for my Traddy that is very fast (twice as fast as the Ashford spindle).  Peter Teal was correct, one can easily spin worsted on a driven spindle.

That makes DRS seem perfect for spinning fines (60s - 80s).  However, at that point the ratio of the whorls is less than 1:1.01, so with a 1" whorl diameter, we are talking about a difference of  less than 1/100 of an inch between the diameters of the two whorls, and when I make whorls that are only 1/2" in diameter, the difference between the whorl diameters is less than 1/200".  I made some flier/bobbin assemblies for such spinning.  At that point, seasonal changes in humidity can change the diameter of wooden whorls enough to affect performance. For such whorls, details such as the kind of wood and the orientation of the grain of the wood matters a lot.  Or, the build up of a film of drive belt dressing on one whorl can cause the thing not to work, provoking much profanity. This summer I have not been spinning a lot of fines, so I have been using Irish Tension on the #0 flier or conventional DD with a DRS of 1.01 and belt slip when I needed to sample fines.

Tolerances for whorls designed for spinning 5,600 ypp are on the order of 1/32".    This is doable with no profanity. The 9 tpi of these singles needs the speed of a DRS system.  DRS is the only way that a spinner can spin 5,600 ypp singles as fast as they spin 1,800 ypp singles.  A man's sweater from 2-ply worseted requires about a million twists to make the yarn.  A sweater from 5-ply takes about 4 million twists.  If one is spinning gansey yarn, a faster wheel is better.  For singles in the range of 10s to 40s, DRS is worth the effort.  And, I admit that it is a lot of effort.

One other point is that the math for DRS is a bit counter intuitive, so one must do the algebra to select correct whorl diameters.  I made up little tables for different effective bobbin diameters and those tables are in both my shop journal and my spinning journal for reference. I check them frequently.








No comments: