Lewis had a long history of using bought-in engines in its motorcycles. Among those used in the early days were Kelecom, De Dion and Minerva, and the 1910 models used 3 h.p. air- and water-cooled motors from the Stevens Motor Manufacturing Company in Wolverhampton, and a 3 1/2 h.p. air-cooled motor either built, or at least used, by A.W. Wall Limited of Birmingham in their ROC machines.
For 1911, Lewis adopted new engines across its model range. These motors were built by F.E. Baker Ltd. from 1910, with the works first at Moorsom St, Newtown, Birmingham and later at "Precision Works", Kings Norton, Birmingham. We can let them introduce themselves by quoting from their undated (but c1914) catalogue:
So F.E. Baker were wholesalers, and indisputably they were a major force in the Australian motorcycle industry before the first war. A.G. Healing in Melbourne were very large users of the Precision engine, and contemporary photographs show Precisions engines lined up on benches in considerable numbers: I count 29 of what look to be 3 1/2 h.p. units in one photo in A to Z of Australian-made Motorcycles. In the same book, Rob Saward reports that Healings were the sole agents for Precision engines sold in Australia.
I've already had one go at writing the Precision chapter of the Lewis motorcycle story (I've saved it here for reference), but as I did so I came up with a number of key - but unanswered - questions. Thanks to input from a number of local enthusiasts, I can now have another go...
The air-cooled 3 1/2 h.p. Lewis engine
Despite speculation, the evidence that 1911-on Lewis engines were imported directly from the UK and originated in the works of F.E. Baker Ltd in Birmingham is compelling. Contemporary press reports describe the motors as being "British" and "imported", and the engine numbers of Lewis engines fall into the general Precision sequence. Even though the timing chest is customised for Lewis, many are clearly stamped "Made in England".
However although the motors used by Lewis were made by F.E. Baker they were not "Precision" motors. "Precision" was the brand used by Baker for their main range of proprietary engines, but we'll show here that not all motors from the Precision Works were called Precision.
Let's start by looking at similarities and differences between contemporary Precision and Lewis engines. Here we see the 1914 air-cooled offerings from Lewis (on the left) and Precision (on the right):
If we start with the bottom end, we find that the motors are essentially identical except that the Precision timing cover is labelled "FEB" between the tappet bosses (absent on the Lewis version) and "PRECISION" across the bottom where the Lewis motor has "LEWIS" cast in.
The motors also differ in the magneto chain case, with "The Lewis" replacing "The Precision" on the Lewis version - but customisation of this sort was a common thing for Precision. Many Australian and British manufacturers - for example Bullock in Adelaide - had their brand names cast into the magneto chain case of their Precision engines.
Things get interesting when we look at the cylinder barrels. Looking across the range of British and Australian machines that used Precision motors in the 1911 - 1916 period, it's clear that the 3 1/2 h.p. (499 cc) Precision motor was supplied with a "square" fin pattern. By contrast the air-cooled Lewis motors have quite different cylinders: not only are the cylinder fins essentially "round" in plan, but also the bare, downward-angled exhaust port is quite different from the standard Precision offering. This cylinder is not unique to Lewis but was used by a number of other Baker customers including Calthorpe, Mead, Dunkley and ASL, confirming that the cylinder is of Baker, and not Lewis, origin. Here we illustrate with a 3 1/2 h.p. machine from the 1914 Calthorpe catalogue, showing the "non-Precision" Baker engine.
The most likely explanation is that F.E. Baker had two distinct ranges of engines: their proprietary engines branded "Precision", and another range that they offered directly to selected (large-scale?) customers, bypassing local agents. To avoid conflicts with agents who had "exclusive" rights to sell the "Precision" engine in a particular territory, these engines used different cylinders, and did not carry the "Precision" brand. In the Australian case, A.G. Healing had exclusive rights to distribute "Precision" engines, but Bakers sold "Lewis" engines directly to Lewis.
As I pondered the Lewis/Precision question, I could think of only one example of a Lewis machine fitted with a square barrel. In the photo below of noted competition rider Edgar Ferguson, his c1914 machine sports what is obviously a square-finned motor:
At least the machine looks like a Lewis, and given that Edgar Ferguson is seen elsewhere in Lewis photographs it would make sense for his machine to be a Lewis. However consulting the registration records, we find that 5901 was registered to Edgar Furguson as a 3 1/2 h.p. "Precision"!
Another piece of information is a surviving machine, c1911, that has all the Lewis hallmarks but is fitted with a square-fin Precision motor with "The Precision" in script on the magneto chain case, and has "The Precision" sign-written on the tank. I'm chasing more information about this bike - does it have a Lewis frame number?
A possible explanation for both the Ferguson machine and the mystery survivor is that Lewis, on running out of "Lewis" engines, built some machines with "Precision" motors (possibly sourced from A.G. Healing the Australian Precision agent), and sold them branded "Precision". I'll keep chasing this.
The water-cooled 3 1/2 h.p. Lewis engine
Much of the above discussion also applies to the water-cooled motor of the day, although here the situation is complicated by the long history of water cooled engines at Vivian Lewis Limited. As for the air-cooled motors, the water-cooled cylinder offered by Precision in their 1914 catalogue bears little resemblance to the Lewis item:
By contrast, the cylinder on the early-pattern water-cooled Lewis-Precision motor shows a close family resemblance to cylinders on the earlier Lewis offerings, which came from the Stevens Motor Manufacturing Company. From 1911 through to the final Precision engines of 1915-16 the changes in cylinder design were by evolution, rather than revolution.
Tracking our thoughts on Baker's air-cooled motors, we might expect to find two lines of water-cooled motor: the proprietary "Precision", and a slightly different version for direct sale to selected customers.
Reading about motorcycle engine production in the Midlands of the U.K., it seems that many (or possibly most) manufacturers outsourced the production of cast items like cylinders and crankcases to specialist foundries in the region. Given the similarities between the cylinders on the last of the water-cooled Stevens-built motors and the first of those built by F.E. Baker, a possible explanation is that Baker sourced their cylinders from the same firm that had been making them for Stevens. Given that the Stevens Motor Manufacturing Company had recently been wound up (to make way for production of the new AJS machine) it is unlikely that Stevens would have any objection to their design continuing development with another manufacturer.
The air-cooled 4 & 6 h.p. twin-cylinder Lewis engine
These engines, shown on the Lewis air-cooled engines page, were essentially identical to proprietary Precision items. The only known variation was that the Lewis versions of these engines did not carry the Precision name anywhere on the crankcase or timing chest, where it was variously found over the production span of the Precision twins. The Lewis name was only on the magneto chain case on the twins.
The 3 h.p. air-cooled and water-cooled Lewis engine
From the registration records, it would seem that Lewis sold a large number of machines powered by the 3 h.p. (75 x 86 mm giving 380 cc) variant of the Baker-made engine, which was offered in both air- and water-cooled form. I have never seen this engine elsewhere, and it is not listed in contemporary Precision catalogues that I have seen. Was it built exclusively for Lewis?
Examining one of these 3 h.p. engines, the crankcase appears identical to that of the 3 1/2 h.p. motor, but the timing cover is modified to give a valve spacing of 45 mm, a dimension shared with the 6 h.p. twin. On the 3 1/2 h.p. single the valve spacing is about 60 mm. Note that with equal cam spacing but different valve spacing the cam followers are likely to be different for the two engines - watch out for this if you're assembling an engine from the proverbial "box of bits".
The cylinders of the 3 h.p. single and the 6.h.p. twin are almost identical. Both have a bore of 75 mm, and as the photos below show, there is a strong resemblance between the two when comparing fins, shape, dimensions etc. The major difference is the exhaust port pointing more steeply downwards on the 3 h.p. single. The twin is on the left, the single on the right:
So for F.E. Baker the recipe for the 3 h.p. single was a relatively simple one: a new timing cover, new cam followers, and some modifications to the existing pattern for the front cylinder of the twin. Or perhaps it was the other way around: was the twin built using modified 3 h.p. cylinders? Over to you, Precision experts.
So that seems to cover most bases, even though there are still a few outstanding questions. Any additional information would be most welcome, particularly relating to the origin of the cylinders used by Stevens and Baker. Watch this space.