July 2009

I came across a Gearhart Knitting Machine cylinder that, even though unused and stored in its crate for the last 85 years, had tiny hairline cracks and some very small chips on ridges of the tracks that hold the needles. I’m not sure if you can make it out, but here is a photo of the cylinder. The cracks do not circle all 360 degrees of the cylinder.


This leads me to Richard Candee’s statement in his book titled, The Hand-Cranked Knitter and Sock Machine. In chapter 9, page 234, he says:

Late Gearhart cylinders were made of bad pot metal, a silent testimony to the company’s coming bankruptcy.

So, did this cylinder have bad pot metal? Based on the discovery of this cylinder with hairline cracks, I do agree that somewhere along the line, the company received a batch of cylinders that did not hold up well after 85 years. For the second part of the statement, the company actually went out of business due to a hostile takeover, not a batch of bad cylinders from the Herman Doehler’s Die Casting Company. I think the health of the cylinder points more to production problems with the Doehler company or the cylinder’s subsequent storage then to the Gearhart Knitting Machine company. But, more on hostile takeovers and die-casting processes in future posts.

Richard also noted that he had observed Gearhart cylinders that had sagged or bent. And one of them was even molded to the square shape of the box it came in.

Wow! A square cylinder. If a cylinder was received by the Gearhart Knitting Machine company per the photo above, I wonder if the company would have shipped them out? I’m inclined to say they would have rejected the batch and demanded a replacement. So far, I haven’t found anything which indicates the company knew about bad cylinders. My own cylinders are perfect. no chipping, no hairline cracks, perfectly round.

I’d assume that the cylinder deteriorated over time. Either bad metal or low casting pressure might explain the progression of hairline cracks. Sagging? I don’t know. If I were stored in an attic that could reach over 150 degrees (like my attic), I might end up square and sagged too. Its hard to say without knowing more about the environment.

Pot metal is a slang term that covers many alloys of aluminum, zinc, copper and a few others in various combinations. It’s also referred to as Die Cast metal. It’s hard to just look at it and differentiate it from an aluminum alloy except in most cases, it’s heavier because of the zinc.

A scratch and daub test will usually tell you if it’s die cast alloy. A drop of battery acid (sulfuric) will foam if it has zinc. If it has zinc, it’s most likely a die cast alloy or “pot metal”. So, I performed this test on one of my cylinders, and sure enough the metal did contain zinc.

While Pot Metal does melt at low foundry temperatures, its most important property is it’s strength right out of the mould. It is this property that makes it useful in high speed die casting. Casted objects can be ejected out of the mould almost immediately after the molten alloy is injected into the mould. However, with die casting, high pressure is important because too little pressure will leave pin holes and chips, and produce a casting suseptible to gradual cracking and separation over time due to low density.

There are many slightly different alloys that have the desirable character of Pot Metal but none are strong. Probably the strongest is one called ZYMAC, an alloy I believe made up of zinc, magnesium, aluminium and copper. The Alcoa Aluminium Company was started in 1888, in Pittsburgh. So, the Doehler company in Cleveland would have had access to the material at the time they contracted with the Gearhart Knitting Machine Company to cast their cylinders. This contract was awarded in 1915. I’m not sure if the cylinders were a ZYMAC alloy, but I’m sure a metallurgist could tell me if one of the ingredients is Aluminium.

Pot metal is not junk. Zinc alloy is the least expensive metal for making light castings. Steel rusts faster than zinc alloy pits, so in that sense it beats steel. Zinc oxide turns to a white-ish powder. It forms pits and can turn into crumbly cubes before disintegrating.

So, I think in the absence of white powder, crumbly cubes, distortion or abuse, the cylinder in the photo might have been a victim of low pressure during casting. This produced a low density cast, which could have developed cracks over the years. That’s my theory.

As for the square cylinder, this is a mystery. It takes 650 degrees to melt pot metal. I wonder if large swings in temperature over years would cause this? From the posting titled Assignment of Refund Guarantee (click here), we know that the company guaranteed their product for 10 years. So, I think they had sufficient confidence in the cylinders at least over the short term.

The original question was, did the cylinders have bad pot metal? I don’t have any evidence yet that points to this, although I do have evidence that indicates a low casting pressure.

The Home Earners were always busy all across the country, knitting socks on their Gearhart Knitting Machines and sending those socks into the Gearhart Knitting Machine Company. The company, in turn, distributed the socks to various retailers and mailed out checks to the Home Earners for their work.

The company kept detailed records on this, since the Hosiery business was a major contributor to the company’s income in addition to machine sales. Shown below is the incoming hosiery tally for the month of February, 1928, broken down by day and by type of hosiery. There were two types of hosiery. The first was CSH, which stood for Childrens Support Hosiery. The second was WOOL, which stood for Wooladdie Hosiery. Wooladdie was adult hosiery, which was sturdier and thicker than the children’s hosiery.


I really couldn’t find anything remarkable on this list yet as far as volumes go. I’ll need to collect more information before I can tie this list volume-wise to the outgoing shipments to the retailers described in an earlier posting titled Some Hosiery Customers (click here). I’ll also need to get a good idea of the seasonal variations in supply and demand before I can tie everything to warehouse quantities.

One thing that is surprizing, is the rejection rate for the incoming hosiery. We can see a 13% rejection rate on Children’s Sport Hose and a 16% rejection rate on Wooladdie Hose. Does that sound like a lot?

This means that, on average, for every 12 socks sent in 1 sock was rejected. I ran this number past some others. Most say its not too bad, and they would have expected more rejections. It would be interesting to see the distribution between new Home Earners and older, more experienced Home Earners. I don’t think I’d be able to figure that out from the archives, since I don’t have the customer list of machine sales. I do think the experienced Home Earner would have a rejection rate approaching 0%.

I suspect that I’ve just acquired the only original Gearhart Knitting Machine sock in existence. If any others exists they may be tucked away in an attic somewhere, or they may be an article of clothing on one of those very old Smithsonian exhibits. We’ll never know, because I don’t think there is any way to distinguish a 1920’s sock as having been made by a Gearhart Knitting Machine.


Just think, this sock was made when Calvin Coolidge was President, and when my 90-year-old barber was only 6-years-old, and when everyone probably wished for a brand new 1925 Ford Model T for Christmas! Its amazing to think, that by whatever kind of fate, this last surviving sock could have made it over so many years to eventually wind up all the way back to one of the Gearhart descendants and the last spot for for most of the Gearhart Knitting Machine company artifacts. You probably have to be a Gearhart to appreciate this, so no doubt I’m making too much out of it.

This sock was given to me by Kathy Roletter, as part of the original shipment from her Gearhart Knitting Machine. Details of this machine can be found in my post titled The Brand New 1924 Gearhart Knitting Machine (click here). There are actually two articles, a sock and the sample which was knit on the machine and included with the shipment. I’ve got the articles stored in a wooden case, so they should be safe and sound for another 85 years. I can’t really think of anything else more valuable to the archives than this, other than a mint-condition machine.



Its quite a unique addition to the archives, because we see for the first time a Gearhart Knitting Machine sock as it was intended to be, at the time the company was in business. I suspect the sock was taken from inventory and included with the machine when shipped. I’d have a hard time thinking that the company would take the amount of time necessary to knit an entire sock in order to check out the workings of the machine. I may be wrong but we’ll likely never know for sure.

However, the knitted sample which was on the cylinder was definitely knitted on the machine that was shipped, probably by one of the men shown in the photo from my posting titled Assembling Machines for Shipment (click here). It would take just a few minutes to do this, and would certainly suffice to verify the correct operation. The articles have a few holes in them, since something got into the crate sometime between 1924 and 2009. There are no droppings in the crate, so it was probably some kind of small bug rather than a mouse. Regardless, the sock and sample are in really good shape considering their age, but nevertheless very fragile. I also have four skeins of wool that were also included with the shipment, just as the inventory sheets in the archives state.

I think it would be pretty nice to make a new sock from the same color and type of wool, so that we could see the 1924 sock and an identical 2009 sock side by side. Add this to the list of things to do…


One last thought. So why was Kathy’s new machine so hard to turn? The machine was obviously checked out and used before shipment. When she tried it out after assembling it, the cylinder was binding against the Collar Ring. Kathy had to file down the cylinder slightly in order to reduce the binding. Hmmm. Is it the properties of the cylinder metal, after sitting for 85 years, which caused the cylinder to expand just a tiny amount? The machine was otherwise brand new, so it hadn’t been dropped or abused in any way whatsoever. All cylinders at that time (1924) would have been the same size since they were die-cast from a mold, so its not like this was a individually-machined part. Any thoughts on this?


I have several dozen pricing sheets for the Gearhart Knitting Machine. Most of them are undated, so it would be hard to put any weight behind them when it comes to dating the price of the machines. However, there are a few sheets that are clearly dated and have illustrations as well. The prices from these sheets are shown below. In all cases, the price includes shipping.

There were three different models manufactured between 1888 and 1925. The original model had a very small crank, no cam ring, no ribber, and operated by rotating the Cylinder inside the Crown Ring. The improved model resembled the modern model shown to the right, but it also operated by rotating the Cylinder inside the Crown Ring. The modern model was the first to reverse the operating mechanism by instead turning a Cam Ring inside the Crown Ring, thereby leaving the Cylinder stationary. There were also many other improvements over the years, including a ribber as well as different Cam and Carrier mechanisms.

By 1924, the machine had reached the height of its evolution. I have several letters describing suggested improvements. Having sold over 200,000 machines, the company was very careful to scrutinize every suggested change to the machine. From what I can tell, very few suggestions made it into production. Even some obvious changes, like having the crank turn one revolution for each 360 degrees of cylinder rotation, was declined for good reason. I’d be hard-pressed after reviewing the letters to override the company’s judgement at that time, other than for those things which would soon become easy to implement due to big industry milestones like ball-bearings, electronics, and so on. But then, how would they have know at that time? To their credit, they did take advantage of modern die-casting techiques invented by Herman Doehler in 1905. A motor drive (declined by the company in 1914), and sealed bearings (declined by the company in 1922) are just two of the other things that might have stayed on the table given developments which were just then appearing on the horizon.

Date Price Model Description Today’s Price
1889 $3.50 original Wood. 1 cylinder, no ribber $82.00
1890 $5.00 original Metal. 1 cylinder, no ribber $118.00
1891 $8.00 improved 1 cylinder, no ribber $189.00
1891 $10.00 improved 2 cylinders, ribber $236.00
1900 $12.00 improved 2 cylinders, ribber $306.00
1908 $14.00 modern 2 cylinders, ribber $331.00
1914 $14.00 modern 2 cylinders, ribber $297.00
1921 $47.00 modern 2 cylinders, ribber, 2-lb yarn incl. $561.00
1924 $67.00 modern 2 cylinders, ribber, 2-lb yarn incl. $831.00


I have included a column which shows the price of the machine in today’s dollar, adjusted for inflation. In general, we see a price increase as accessories are added and the machine evolves. It is interesting to note, that unlike other products of the time which were mass-produced, the Gearhart Knitting Machine did not experience a price decrease other than for the period between 1908 and 1914. The Ford Model T, for example, was priced at $850 in 1909, but fell to $300 by 1920 due to the improvements in manufacturing processes.

Possibly in the case of the Gearhart Knitting Machine, the parts were not manufactured in a volume high enough to realize much of a cost savings. Also, if the competitors had inflated prices, then I suspect that the Gearhart Knitting Machine company would have little incentive to out-price them, since according to the volume of machine sales, the Gearhart Knitting Machine was by far the preferred machine at any price.

Out of curiosity, I made some inquiries about the machine which is in production today. We currently have a single manufacturer. The price ranges from $1695.00 to $3000.00. You can sure buy a lot of socks at Sears for that price. It is definitely for the hobbyist since it has no historic value. This machine is comparable to the 1924 Gearhart Knitting Machine and has a similar parts list.

So, I’m now even more curious. If you have a 1980’s Harmony Knitter, I’d be very interested in hearing how much you paid for it, if you bought it new. I’d also be interested to know if anyone has other pricing numbers for the Gearhart Knitting Machine for the years not listed above. And, it would also be useful to compare the Gearhart Knitting Machine prices to their competitor’s prices. I don’t have any of this information, but would very much welcome input.

Of course, buying vintage machines on eBay is another subject entirely. This might make an interesting posting…

I’m off to see Kathy Roletter in Virginia, so no more postings for a few days. Kathy is the proud owner of a brand-new Gearhart Knitting Machine, which she bought in its original unopened crate. Her story was in the posting titled The Brand New 1924 Gearhart Knitting Machine (click here).

I’ve got my camera and a copy of Miss Ada Weiner’s instruction book (click here) for her. Hopefully, I’ll come back with some more photos in addition to all nice photos (click here) from Kathy.



Where do they find these children? Here is Little Miss Addington. She looks to be about 10 years old. She has her PhD in Knitting Machines, as do most other little kids appearing in Gearhart Knitting Machine advertisments or residing near the Gearhart Knitting Machine factory or an actual Gearhart Knitting machne. I suspect they were shipped off at an early age to a knitting machine school, where they were taught to make it look so easy and put all the grown-ups to shame. I even have some advertisements noting that a child of 6 years old is fully qualified to use the machine. 6 years old. Hmmm.

Miss Addington has apparently gone into business with her mother over there in Georgia. It looks like she is doing a very good job. I wonder if she is still alive. If so, I should mail this photo to her…


Oh, I just found another one! She looks to be about the same age as Miss Addington. I think the kids back then must have been either really smart, or they didn’t have a chance to get distracted by the blue flicker of a TV screen.


Here is a photo of a lady using the 1892 Gearhart Knitter. According to the caption, the photo was taken in 1923, and the machine at that time would have been in use for 31 years. This is the only photo I can find so far of one of the very early machines before they took on their final appearance with a large crank and a uniform cylinder diameter. I have seen that by late 1892, the Gearhart Knitting Machine Company was selling the newer model, so I think the machine in this particular photo would have been one of the last of the early machines ever produced.


This particular model is very simple compared to the later models. The handle is very small, and the cylinder diameter is 5 inches, which is much larger than the 4.625 inch diameter cylinders of the later machines. The ribbing attachment hadn’t been invented yet, and the weight appears to be a large round disc inside the sock.

I have Joseph Gearhart’s machine, as shown below. It is surprizingly smooth turning. There is very little friction and it is quite easy to turn considering the weight of the cylinder. I haven’t tried to use this machine to knit a sock, but I assume it would work quite well once I installed the correct needles. The slots for the needles are very narrow – too narrow for the needles used in the later machines. Unfortunately, I do not have a complete set of needles for this machine so I cannot get to a point where it would be completely operational.


I do have the manual for this machine, as shown below. The manual describes how to set and operate the machine in the same way as the newer manuals. Naturally, this manual is only eight pages long, which matches the simplicity of the machine and might explain why someone would keep the machine for 30 years before upgrading to something more complex and more expensive.


Next Page »