Sewing Machine Research
ISMACS International
International Sewing Machine Collectors' Society

Union Restoration

Martin Gregory
ISMACS News
Issue 41

MY LATEST restoration project has been a small two-thread chain-stitch machine called the Union. I acquired a very battered example plus access to a complete machine from Graham. Now, many hours of labour later, it is working again and looks more respectable.

Initially the machine appeared not only to be seized solid but also to be totally covered in rust. Once freed, with penetrating oil, careful removal of the surface rust with very fine steel wool revealed that the remains f some decoration was still in place.

Although discoloured by the rust, I have managed to save some decoration and stabilise it under a layer of varnish. For what looks at first sight like an overgrown Raymond or Weir machine, why was it so worth the effort?

So far there seem to be four survivors of this machine. One has been seen in a small museum in America and the other three were at the Milton Keynes ISMACS regional meeting.

Two machines in the UK are numbers 34, 100 and 15. 34 has the name Union on the top of the overarm, neither of other others has any trace of a name. 34 and 156 have similar decoration and are much closer to each other than either is to 1.

Almost every major component is numbered as parts are very far from interchangeable; this was certainly not a mass-produced machine! As yet, we do not know who produced it.

Figure 1 shows the general arrangement of the machine. it looks very like a slightly-enlarged early Raymond New England single-thread chain-stitch machine. The hand wheel and gears are the same, so is the needle-bar drive with the reel (R) at the top, so is the scuffing presser-foot feed (D) raised by lever (C).

Obvious differences are the much larger brass needle plate, the cloth plate with a hump to carry the bevel gear (B2) and the second reel (R2). Looking underneath we see a stitch-forming mechanism quite different from the ordinary Raymond machine.

The Union makes a stitch which is equivalent to the Grover & Baker stitch. It will not unravel and uses thread from two ordinary reels. How does it work?

The top shaft provides the up-and-down motion for the upper needle in the normal way. The lower shaft rotates two bevel gears (B1, B2) which drive the connection rod A (figure 1) carrying a looper with an eye like a needle. The lower thread from (R2) is passed through the eye of the looper which is guided by a pin in a slot attached to the needle plate. Figure 2 shows four stages in the stitch-forming process (not to scale):

1.  In A the upper needle is down through the cloth and rising, leaving a loop of upper thread; the looper is advancing to pick up the loop on the right-hand side of the needle.

2.  In B the upper needle is above the cloth and the presser foot is feeding the cloth along; the looper retains the loop of upper thread it picked up in A.

The machine does not sew very well because there is only a very rudimentary tension system for the threads and neither thread has any take-up mechanism to pay out thread to make a loop and then pull it tight to form the stitch. However, it is such an elegant and simple solution to the two-thread chain-stitch problem that I am surprised that nobody developed it into a serious machine.

Now for a bit of speculation about its history: 1855-65 was a boom time for two-thread chain-stitch (double-loop-stitch) patents. In the UK there were 70 compared with only 29 for the period 1865-75.

My of the early ones represented new ideas on how to form a locked stitch from two threads fed from standard  reels.

The Union machine bears a remarkable family resemblance to the early Raymond machines. Grace Rogers Cooper's book tells us that Nettleton & Raymond started off with a two-thread chain-stitch machine before Raymond patented his New England single-thread machine in 1858/61.

Other manufacturers also made the New England machine, including Grout and Folsom. Raymond himself patented an improved two-thread chain-stitch machine and both Grout and Folsom held patents for two-thread chain-stitch machines.

T S Washburn's US patent for a two-thread chain-stitch machine was registered in the UK by G Whight (later to manufacture his own Excelsior two-thread chain-stitch machine in 1860. The patent drawing shows the front end of a New England machine, like the Union, but with a different and clumsier looper assembly underneath.

Initially the machine appeared not only to be seized solid but also to be totally covered in rust. Once freed, with penetrating oil, careful removal of the surface rust with very fine steel wool revealed that  some decoration was still in place.

Although discoloured by the rust, I have managed to save some decoration and stabilise it under a layer of varnish. For what looks at first sight like an overgrown Raymond or Weir machine, why was it so worth the effort?

So far there seem to be four survivors of this machine. One has been seen in a small museum in America and the other three were at the Milton Keynes ISMACS regional meeting.

Three machines in the UK are numbered 34, 100 and 15. 34 has the name Union on the top of the overarm, neither of the others has any trace of a name. 34 and 156 have similar decoration and are much closer to each other than either is to 100.

Almost every major component is numbered as parts are very far from interchangeable; this was certainly not a mass-produced machine! As yet, we do not know who produced it.

Figure 1 shows the general arrangement of the machine. It looks very like a slightly-enlarged early Raymond New England single-thread chain-stitch machine. The hand wheel and gears are the same, so is the needle-bar drive with the reel (R) at the top, so is the scuffing presser-foot feed (D) raised by lever (C).

Obvious differences are the much larger brass needle plate, the cloth plate with a hump to carry the bevel gear (B2) and the second reel (R2). Looking underneath we see a stitch-forming mechanism quite different from the ordinary Raymond .

The Union makes a stitch which is equivalent to the Grover & Baker stitch. It will not unravel and uses thread from two ordinary reels. How does it work?

The top shaft provides the up-and-down motion for the upper needle in the normal way. The lower shaft rotates two bevel gears (B1, B2) which drive the connection rod A (figure 1) carrying a looper with an eye like a needle. The lower thread from (R2) is passed through the eye of the looper which is guided by a pin in a slot attached to the needle plate. Figure 2 shows four stages in the stitch-forming process (not to scale):

1.  In A the upper needle is down through the cloth and rising, leaving a loop of upper thread; the looper is advancing to pick up the loop on the right-hand side of the needle.

2.  In B the upper needle is above the cloth and the presser foot is feeding the cloth along; the looper retains the loop of upper thread it picked up in A.

The machine does not sew very well because there is only a very rudimentary tension system for the threads and neither thread has any take-up mechanism to pay out thread to make a loop and then pull it tight to form the stitch. However, it is such an elegant and simple solution to the two-thread chain-stitch problem that I am surprised that nobody developed it into a serious machine.

Now for a bit of speculation about its history: 1855-65 was a boom time for two-thread chain-stitch (double-loop-stitch) patents. In the UK there were 70 compared with only 29 for the period 1865-75.

Many of the early ones represented new ideas on how to form a locked stitch from two threads fed from standard  reels.

The Union machine bears a remarkable family resemblance to the early Raymond machines. Grace Rogers Cooper's book tells us that Nettleton & Raymond started off with a two-thread chain-stitch machine before Raymond patented his New England single-thread machine in 1858/61.

Other manufacturers also made the New England machine, including Grout and Folsom. Raymond himself patented an improved two-thread chain-stitch machine and both Grout and Folsom held patents for two-thread chain-stitch machines.

T S Washburn's US patent for a two-thread chain-stitch machine was registered in the UK by G Whight (later to manufacture his own Excelsior two-thread chain-stitch machine in 1860). The patent drawing shows the front end of a New England machine, like the Union, but with a different and clumsier looper assembly underneath.

Who's idea was the Union - improved Raymond, Grout, Folsom, improved Washburn?

I would like to give credit to a very ingenious mechanic.                            Martin Gregory

GF Comments: There is somewhere deep within the stored MS collection, a much-larger Parmitter machine made under Nettleton & Raymond patents which, I seem to remember, uses the same, or similar two-thread mechanism.

I will explore and report in the next issue.