The Speed Demon 30

by Adrian Duncan

Click on images to view larger picture.
Hover over images for captions.

    Origin and Numbers
    The Speed Demon in the Contemporary Media
    The Head-Fixture Issue
    Running the Speed Demon

Time for a nostalgic ramble down Barstock Boulevard to take a look at a relatively obscure motor from the late 1940's—the rather inappropriately-named Speed Demon from Long Island, New York.

The Speed Demon 30 was a 4.85 cc (.296 cu. in.) sideport diesel engine produced to very high standards in relatively small numbers for less than a year beginning in late 1947. It made little if any impression upon the contemporary marketplace, but it is a very interesting motor indeed in technical terms. Furthermore, a number of aspects of the history of this engine have been the subject of debate among engine aficionados over the years.

All of this makes the Speed Demon a particularly interesting subject for research! Let's start by considering a few of those debatable points.

Origin and Numbers

To begin with, questions have been raised regarding the geographic origin of the engine. It has generally been assumed to have been produced in the USA, but some have speculated (based on its design features plus a surviving letter which accompanied a replacement needle valve assembly with Bert Striegler's example) that it might in fact have been made in Europe and imported to the USA by its Eastern US promoters.

In my personal view, two factors argue very strongly against this notion. Firstly, there is no record of the engine ever appearing on the European market or even being the subject of commentary in the European modelling media. If it was produced there, we would surely expect to find some evidence of a European marketing effort, or at least of some awareness of the engine in Europe? Secondly, the fact that all measurements and threads are based upon US (as opposed to metric) standards argues strongly against a European origin, as does the use of what appears to be a made-in-USA prop driver from an OK 29 on the later versions.

Wherever it was actually manufactured, there is no debate at all regarding the fact that the promoters of the engine were based in the eastern United States. One of my two examples still has its guarantee certificate and spare parts price list, and both documents name the makers as the Eastern Model Engineering Co. of 40-16 83rd Street, Jackson Heights, Long Island, New York. Today this location is included in the predominantly residential Queens district of New York, and the address is now (2010) shared by several dental businesses. It's quite possible that the site was occupied by a completely different building in 1947, perhaps even a residence.

The New York State location plus the fact that the later models of this engine sport the apparent OK 29 pressed steel prop driver noted previously combine to suggest some kind of tie-in with the OK factory at Herkimer, which was also located in the State of New York. In addition, the bore and stroke measurements, the use of compression ignition and certain other data to be discussed below imply a degree of influence from the contemporary Drone fixed compression diesel manufactured in nearby New Jersey.

Now the town of Herkimer is nowhere near Long Island (or New Jersey), being located well to the northwest, albeit still well within the New York state boundaries. So the geographic connection with OK appears to be rather tenuous. Nonetheless, Bert Striegler has recalled reading somewhere that the Eastern Model Engineering Co first became involved with the model trade as suppliers of parts to OK under contract and were looking at the Speed Demon as a potential new cash-flow generator which the American precision engineering industry had experienced during the recent WW2 boom. This seems quite credible to me. Overall, I think that the notion that this engine might have been an import is not readily defendable when all the surrounding evidence is considered.

Quite apart from the issue of the engine's origin, there has also been some discussion regarding the number of these motors that were actually made. The late OFW Fisher claimed a figure of 2,000 to 3,000 examples, but I've always been highly sceptical—on what basis did he make this claim? Moreover, if he was anywhere near right, where are they now?

All of the known authentic Speed Demons carry stamped-on serial numbers, invariably appearing at the rear of the mounting plate which separates the cylinder casing from the main crankcase. My information (from knowledgeable old-time sources such as Gus Munich, Art Suhr and others) is that the serial numbering started at 1000 and went on up from there—a not uncommon practise. I have two of these motors—nos. 1008 (a very early example, assuming that they started at 1000) and 1427. The one illustrated by Fisher was reportedly no. 1355.

OK, so how many were made?? Well, I am presently unaware of any reported examples with serial numbers under four figures, so it seems highly likely that they started at 1000 as claimed by my various informants. Furthermore, no. 1427 is the highest serial number that I have ever encountered, so I would guess that only about 500 examples were actually made, if that. The relative rarity of the engine today certainly supports such a low figure. Still, 500 engines was an achievement that commands respect! Bert's example is right in the middle at 1255, and don't forget Fisher's 1355.


Now that we seem to have come to a few provisional conclusions regarding the US origin and relative rarity of our subject, it's time to take a closer look at the engine itself. In a number of ways, the Speed Demon seems somewhat out of step with its time and place—it's an American-made sideport diesel engine which was competing in a marketplace which had by and large resisted the advent of the model diesel. The sole exception to this was the Drone fixed-compression FRV diesel which enjoyed a period of considerable popularity primarily among control-line stunt fliers in the USA during 1947 and 1948. And even that was not to last—America's headlong rush towards the general adoption of the rotary valve glow-plug motor was poised to begin when the Speed Demon was introduced. The timing could certainly have been better.

Another distinguishing feature of the engine is the fact that it was constructed entirely from bar stock—no castings were used at any point. Furthermore, the engine used the "moving liner" system of compression adjustment whereby a blind cylinder liner with a fixed head (as opposed to a contra-piston) is moved up and down in its entirety by the action of the comp screw, which engages with a blind tapped hole drilled part-way through the fixed top of the cylinder.

In this design, the compression lever doesn't move up and down with changes in compression setting—the comp screw stays at the same elevation relative to the deck and it's the entire cylinder liner that moves, head and all. This system had previously been used quite successfully on the early French Delmo diesels, and the makers of the Speed Demon may well have been influenced by examples of the Delmo engines which had been brought to the USA by returning US servicemen.

The moving liner system is effective enough as far as it goes—it certainly allows for changes in the compression ratio, but a few seconds' thought will lead to the realization that this system of control inevitably changes the port timing along with the compression! Faster running of course requires that the ignition timing be advanced. In a model diesel this is accomplished by increasing the compression setting. However, as you increase compression and hence advance the ignition timing by lowering the cylinder in a design of this type, the ports necessarily move downwards also.

As a result, the exhaust and transfer ports open later and close earlier (the opposite of what is ideal for faster running) while the induction port opens earlier and stays open longer. That of course is good for a higher running speed, but compromises abound... Oh well, the thing works, so I guess we can't ask for much more!

In any event, this probably wasn't much of an issue with the Speed Demon given the relatively low and narrow speed range within which the engine was at its best. Because of this factor, the required range of compression adjustment wasn't that great. In fact, as the Drone diesel proved, at such low speeds you can get away with fixed compression and still get a useful performance! Regardless, the fixed openings in the outer cylinder casing of the Speed Demon are suitably enlarged so that they don't blank off the cylinder ports at any particular compression setting.

Another factor to consider with an engine of this design is the necessary freedom of the cylinder liner to move within the outer casing. As a result of this freedom, there's nothing to restrain the liner from revolving axially within the casing so that the ports no longer align correctly with the corresponding openings in the casing. The Speed Demon gets around this problem by having a vertical hole drilled off-centre through the top of the outer casing to the rear of the comp screw. This hole corresponds with a similar hole drilled part-way through the actual cylinder head itself—in fact, the two holes were almost certainly drilled at one setting with the engine part-assembled. A 1/16 inch diameter metal dowel is fixed in place in the hole within the inner cylinder head, and this dowel passes through the corresponding hole in the outer casing, in which it is free to move up and down. It thus prevents axial rotation of the cylinder liner while leaving it free to move up and down as the compression is adjusted.

The compression screw is threaded 10-32 UNF below a flange which locates the screw vertically in the top of the outer cylinder casing to resist operating forces and also limits the degree to which the compression can be reduced by contacting the top of the internal cylinder "head". Above the flange, the compression adjusting shaft is smooth, with a diameter of 1/4 inch. This smooth section protrudes through a inch diameter hole in the top of the outer cylinder casing. An alloy collar fits over this smooth section to maintain the vertical location of the comp screw in the outer casing and thus create a push-pull action. The adjusting lever is a 1/8 inch diameter piece of good-quality steel. This is threaded 5-40 on the end that passes through the alloy collar and threads into a hole drilled and tapped transversely through the compression adjusting shaft, thus securing the collar in place.

The compression screw arrangements described above have the effect of adding an unusual challenge to the running of these beasts—the right-hand threaded comp screw works backwards (think about it!). Turning the comp screw anti-clockwise (viewed from above) gives an increase in compression—the opposite of the usual arrangement! You soon get used to this, so it's not really an issue. The one practical operating problem is the need to use some kind of friction lock on the comp screw, because there's no "stiction" in the system and the setting doesn't hold at all in actual operation. I use a bit of thick plastic tubing on the arm—this compresses between the arm and the head with sufficient friction to hold the setting, and it doesn't mar the surface of the head. The Delmo used a very neat tensioning spring to get around this problem—the makers of the Speed Demon could have emulated this with some advantage. Ken Croft came up with a somewhat similar device when making his superb 2 cc Speed Demon replica.

The crankcase is a simple turning from aluminium alloy bar stock with a flat area milled into the upper surface to facilitate the mounting of the upper cylinder components. One interesting feature is the fact that the main bearing and crankcase cavity are turned at an eccentric setting from the outer case. This creates more material thickness at the top of the case where it's needed for the accommodation of the mounting plate and turned aluminium cylinder casing. The latter component incorporates the cooling fins. The crankcase, mounting plate and upper cylinder casing are secured together as a unit by four screws, with thin paper gaskets being used to ensure a seal. A screw-in backplate is employed, and this requires a pin spanner for its removal—please use the correct tool if you ever have to take one of these apart! The main crankshaft bearing appears to be a stock 1/2 inch O/D oilite bushing which is pressed into the main bearing housing.

Turning now to the working components, the cylinder is very simple. It is made from a straight piece of what appears to be leaded steel tubing with a 5/16 inch deep "head button" pressed into the top in place of the more usual contra-piston. This " head button" has a central hole that is 1/4 inch deep and is tapped for the 10x32 UNF comp screw. It also has a 1/8 inch thick flange which locates it on top of the cylinder, and this flange has the same 3/4 inch outside diameter as the cylinder.

Two exhaust ports are cut in the cylinder wall, one on each side, and there are single transfer and induction ports cut fore and aft respectively in the cylinder as well. The transfer port at the front of the cylinder is fed by a bypass channel which is internally milled into the front of the outer cylinder casing using a 1/2 inch diameter milling cutter. The induction port at the rear is fed by a straight venturi tube which screws into the outer cylinder casing on a tapered thread which maintains the orientation of the needle without a lock-nut. A conventional spraybar and needle are used for mixture control.

The piston, crankshaft and con-rod are all massive affairs. The heavy piston is made of cast iron, with little attempt having been made to lighten it. The steel gudgeon pin is pressed in, as it has to be to avoid fouling of the induction and transfer ports. The crankpin and gudgeon pin are 7/32 inch and 3/16 inch diameter respectively—very respectable sizes indeed! The steel rod has a massive look about it, having a column diameter of 7/32 inch. Basically, this engine is so over-built that you'd have to kill it with a sledge-hammer!

Both ends of the rod are bronze bushed, with the bushing on the big end being over-length to limit movement of the rod towards the rear of the crankpin. This extended bushing prevents easy disassembly of the engine and must presumably have been pressed in with the shaft and rod in place during the final assembly process.

The crankweb is a full 1/4 inch thick. The crankshaft journal has a diameter of 3/8 inch and the prop-shaft is 5/16 inches in diameter with a 5/16-24 UNF thread for the prop nut. The early models used a conventional split collar system on a round prop-shaft extension for mounting a "normal" centre-hub prop driver. Quite early on, they seem to have got hold of a supply of OK prop drivers, presumably by arrangement with the Herkimer factory. Certainly, at that point they started machining a flat into one side of the prop mounting portion of the shaft and fitting what appears to be a standard OK 29 driver! No. 1008 is of the first kind, while number 1427 is of the second type. I like the split-collar system better, I must say—the single-flat OK-style arrangement tends to develop a "wobble" pretty quickly.

The engine was supplied with a double set of prop mounting fastenings. One is a conventional nut and washer, while the other is a spinner nut machined from hexagonal bar stock. This is recessed on the rear face to accommodate the nut, so it's clear that both fixtures were intended to be used together. It's my impression that the nut and washer were the main fixtures, with the spinner nut being more for show—although it may have a secondary safety function as a lock-nut.

My own engine number 1427 is little used and even still has both its guarantee certificate and spare parts list as well as the neat little separate plastic fuel tank with which these motors were originally supplied. I've actually never seen a mounted metal tank like the one on Bert Striegler's example, and in fact all but two examples of the engine which I have encountered over the years have had their needle valve assemblies set horizontally when correctly tightened—the two exceptions being Bert's example and the tankless example illustrated by Fisher. It is of course a simple matter to adjust the venturi thread to give a vertical needle valve alignment as required for the use of a back-tank like Bert's.

However, I have seen other examples which were accompanied by tanks just like the one that I have—Art Suhr had one, for instance, of which he kindly sent me the attached photograph many years ago, and the late Ted Enticknap confirmed the authenticity of the tank that came with my example from his own knowledge. There was a short illustrated nostalgia article about the engine in the 1980's in "Model Airplane News" (I have it somewhere, although who knows where??), and it too mentioned the plastic tank. So I'm certain that it's authentic. The tank on Bert's example may be an add-on carried out by a previous owner. Alternatively, it might have been an undocumented option offered by the manufacturers.

The plastic tank is interesting in that it doesn't mount on the motor at all—it has to be mounted elsewhere on the model. Just an accessory that was supplied as an extra as opposed to being an integral part of the motor. Note that the cost of the tank is shown on the spare parts list as 65 cents! Note also that the tank listing specifically says "tanks" (plural), indicating that there may indeed have been more than one type on offer.

The quality of construction of these engines was superb—right up to the best "model engineering" standards. All fits and finishes are beyond reproach. The manufacturer's claims that they were "custom-built" and were "truly a quality product" are well supported by an examination of an as-new example. The engines were guaranteed for a period of one year provided that they were not taken apart or damaged in a crash, but the manufacturers claimed that if well cared for and respectfully used they should last "indefinitely". In the context of a normal "modelling lifetime", I see no reason to doubt this.

The Speed Demon in the Contemporary Media

All right, now we know quite a bit about what it is, where it was made and how many were made. So, on to the next question—when wuz it made? Here the contemporary media constitutes our best source of information. The Speed Demon is not mentioned in the article by Jack Bayha entitled "Care and Feeding of Diesels" which was published in the June 1947 issue of Model Airplane News (MAN). Nor does it appear in the later piece by the same author entitled "Diesels Grow Up" which appeared in the December 1947 issue of MAN. Neither does it feature in the article by Jim Noonan entitled "The Development of Diesels" which appeared in the September 1947 issue of Air Trails and Science Frontiers. That covered all the then-current diesels, including American products. However, the Speed Demon is included in the January 1948 article by Edward Ingram entitled "Model Engines for 1948" which appeared in MAN for that month. Allowing for publication lead-times, this dates the introduction of the engine pretty closely to the final few months of 1947.

However, by the time Ingram's 1949 survey was published in the June 1949 issue of MAN, the Speed Demon had already disappeared without trace from the scene. So it looks to me as if production of these motors began in late 1947 and that all of the rest were made in 1948. In fact, given the very small numbers produced, I'd guess that production had probably ceased by mid-1948 after it was found that the engine couldn't compete with the Drone and accordingly wasn't selling. Perhaps the introduction of the Mk II ball-bearing Drone was the final straw!

Assuming that production carried on until say June 1948, the implied average monthly production rate of only some 50 engines confirms that we're not talking about a major factory operation here. Production rates may in fact have been considerably less than that, since a fair proportion of the engines produced may well have been part of the pre-release stockpile which any prudent maker would have on hand before launching his new design onto the marketplace.

Such a short production life isn't really all that surprising—the engine is very heavy and lacking in power by comparison with the glow and even diesel opposition as of 1948. Rather a boat anchor in many ways. I really can't see it ever having gained much of a foothold in the contemporary US market, and there's presently no evidence to suggest that it was ever exported on a commercial scale.

The January 1948 write-up by Ingram on the then-new Speed Demon is quite informative. It confirms that the engine was manufactured in the USA by the Eastern Model Engineering Co and was entering the market after "...about a year of experimental and design work". This comment is interesting insofar as it begs the question of what the makers were doing in order to eat while they did this work. In fact, this comment strongly suggests that the Speed Demon was most likely the product of a one or two-man "small workshop" operation like many small producers in Britain at the time. The very low production figures for the engine are certainly consistent with this scenario. The avoidance of the use of castings as well as the very high quality of construction both suggest a "model engineering" background in keeping with the company's name.

If this was the case, the individuals concerned were presumably model engineers who earned a living in other ways, perhaps working for a larger business at the same location, while they developed and manufactured the Speed Demon in their spare time or as a business sideline. Possibly the supply of parts to OK as suggested by Bert Striegler was their main business during the Speed Demon's reported one-year development phase. Unless someone with first-hand knowledge of these long-ago events steps forward, we'll probably never know.

Ingram's article describes the method of compression adjustment in considerable (and accurate) detail—old Edward clearly found it highly intriguing! He then does some bragging about the close limits to which the engines are constructed (with some justification, if surviving examples are anything to go by). The rod and main bearings are correctly stated to be oilite items.

A subtle advantage claimed by Ingram for the engine which I hadn't previously thought about but which is undoubtedly true is the fact that the engine is mounted using a flat "plate" which separates the upper and lower halves of the main case and incorporates the bearer lugs. This arrangement is clearly visible in the attached images. If the lugs are damaged in a crash, all that is required is to replace the plate rather than replace the entire case. True—and a very good point!

Manufacturer's recommended fuel given in the write-up is 4 parts S.A.E. 40 oil, 4 parts ether and 2 parts "steam-distilled wood turpentine"! Wot, no kerosene?!

Since the commentary by Ingram appears to represent the Speed Demon's sole starring appearance in the contemporary modelling media, it's worth reproducing here in full. This commentary appeared as one paragraph in a much longer article entitled "Model Motors for 1948" which appeared in the January 1948 issue of "Model Airplane News". The paragraph on the Speed Demon reads as follows:

"A unique feature of the Class B Speed Demon Diesel, placed on the market by Eastern Model Engineering Co. after about a year of experimental and design work, is the method of varying the compression. What may be termed the outer cylinder is of aluminum alloy with an integral head. Within this unit is a steel sleeve, into the upper end of which is pressed a cap. The latter unit is the true combustion chamber. The compression ratio is changed by movement of the inner sleeve unit up and down through the provision of an adjusting screw which engages with the cap and projects through the aluminum head where a lever is attached for turning it. It is stated (that) the clearance between the sleeve and outer cylinder is held to less than .0005 in., and as the surfaces are lubricated no leakage can occur. Another distinctive feature is the provision of a separate engine mounting plate which does away with the necessity of replacing the whole crankcase when the mount is damaged. All main engine components are machined from solid stock, and oilite rod and main bearings are used. The engine is operated on 4 parts S.A.E. 30 oil, 4 parts ether, and 2 parts steam distilled wood turpentine. It is stated (that) a Mercury plane in which a Speed Demon was installed showed good performance and a steady climb despite the fact that this model aircraft has a 6 ft. wingspan and is designed for Class C engines".

The comment about leakage seems a bit obscure at first, but think about it; unlike a conventional engine, there's no gasket to ensure a seal between the cylinder liner and the case because of the moving liner! An O-ring could have been used, but wasn't. So the base compression seal is in fact completely dependent upon a close fit being maintained between the moving liner and the case below exhaust level, as well as an associated oil seal. The makers must have anticipated being challenged on this point, and were ready with an answer! This must be what Ingram was referring to. The fit on my two engines is undeniably very close indeed.

From the style of the above comments, I think it highly likely that Ingram took much of this info directly from some kind of promotional blurb prepared by the makers—it certainly reads that way with the repeated "it is stated" comments. Wish we could find a copy .. but I think that a good chunk of the manufacturer's sales pitch is embodied in the above paragraph. The reported use of a Mercury test aircraft incidentally seems to me to offer further evidence to confirm the North American origin of this motor.

Ingram also included a quite comprehensive table of data on the various motors covered in the article. The data provided for the Speed Demon are generally fairly accurate, as follows:

As all previous commentators have noted, the engines are very well made indeed. Both of mine run extremely smoothly and are a snap to start. They are definitely not "speed demons", though—they like big props and low revs! Respect that, and they'll run for ever for you, I reckon—they're built like tanks!

The Head-Fixture Issue

In the original write-up on this site regarding the Speed Demon, Bert Striegler reported that it was his impression that there was no mechanical fixture of the pressed-in "head button" at the top of the cylinder. This matter has now been clarified thanks to the fact that I was forced to disassemble the upper half of engine number 1008 to fix some major issues. One immediate benefit of this was the opportunity to confirm by direct measurement that this example has the published bore and stroke figures of 16.66 mm x 22.22 mm for a confirmed displacement of 4.85 cc. Both of my examples weigh in at a hefty 11.675 ounces bare, minus tank.

Number 1008 had been the unhappy victim of some clown who had tried to convert it to glow operation without having the slightest understanding of what he was doing! I had a devil of a time sorting out what he'd messed up!! He'd bored a bloody great hole in the top of the aluminium cylinder casing just for starters—I had to make a press-fit aluminium plug and TIG-weld the seam, then re-machine the unit inside and out to eliminate the excess weld. You can hardy tell now. I also had to make accurate copies of all of the compression adjustment components, which I was able to do using no. 1427 as my pattern.

But perhaps we should thank this unknown butcher, because one very useful thing that I did learn as a result of this enforced restoration effort was the method of securing the pressed-in "head button" which seals the bore. This is another point which has been debated over the years, and now it's sussed out! You have to look very closely to see this, but after the bore was finished internally (except perhaps for the final lapping), the flanged "head button" (blind-drilled and tapped from the top for the comp screw) was press-fitted into the top of the cylinder and staked in place with two thin ( 1/16 inch diameter) piano-wire or silver steel dowels which are a very close fit (probably a press-fit) in a pair of holes drilled diametrically through the cylinder wall into the "head button" on diametrically-opposing sides, presumably after the fitting of the button. The subtle part is that these holes stop short of breaking through into the threaded hole provided for the compression screw. The final external finishing of the cylinder was done after this set-up was installed, and the very close fit of the dowels in their holes plus the fact that their outer ends are finished flush with the cylinder surface makes it very difficult indeed to see the evidence once that has been done! But it's effective—once those stakes are in place, that "head button" isn't going anywhere, as I learned the hard way!

I only discovered this myself because I had no choice but to remove the old button on number 1008 thanks to the previous owner having drilled completely through it and tapped it 1/4-28 UNF, not apparently knowing that glow-plugs are threaded 1/4-32 UNS and also not realizing that this wouldn't work in any case due to the lack of any liner fixing! AARRRGGGHHHH!! I clearly had to make a new "head button", but couldn't shift the old wrecked one for love nor money (you can understand why after reading the above description!). So I ended up very carefully machining it out!

Immediately, the reasons for my troubles became plain—there were the remains of the two stakes plus the two holes through the liner where the stakes had gone! I made a new "head button" (copying my other one) and installed it just as per the original using a snug press-fit followed by staking from the sides using the original holes as a location guide for the blind holes in the "head button itself. I then had to re-lap the bore and make a new replica piston to fit, thanks to the abuse that I was forced to heap upon the poor old cylinder. But I did so, and everything now looks fine, works fine and stays together. I subsequently examined number 1427 very carefully and (knowing where to look and what to look for) was finally able to spot the ends of the stakes on that one as well. But you have to look hard in just the right light!

Running the Speed Demon

Having two examples of the engine on hand, I thought that the least that I could do was set one up and give it a run or two for this article. I wasn't worried about doing this given the engine's very sturdy construction plus the fact that I'd previously run-in the new piston in number 1008 following its rebuild described above and had also put a few test runs on number 1427. I elected to use number 1008 for this test since I wished to preserve number 1427 in as pristine a condition as possible.

I also elected to test a Drone Mk I (number 441, which is well run-in) at the same time on the same props to provide a direct comparison between the Speed Demon and the American engine that must surely have been its main contemporary competitor in the US large-diesel market. When the Speed Demon appeared (and for most of its production lifetime), the Mk I plain-bearing Drone was the 5 cc American diesel to beat—the Mk II ball bearing Drone only appeared later in 1948 when the Speed Demon was at or near the end of its short production life. So this comparison seemed fair to me.

I ran the variable compression Speed Demon on conventional diesel fuel, while the Drone was supplied with its own custom fixed-compression brew of 70% ether and 30% oil (my usual mix for the Mk I Drone).

The Speed Demon proved to be a cinch to start, although it did require a prime—choking merely had the effect of flooding the crankcase. Initially, I had to keep reminding myself about the reverse-direction operation of the comp screw, but I soon got the hang of it. The engine was very responsive to both controls, making the establishment of the optimum settings a relatively simple matter. Suction was excellent and running was extremely smooth across the entire speed range tested. In common with many larger diesels, it was found that compression had to be somewhat increased to start from cold and then backed off as the engine warmed up.

I only had a handful of large props that were drilled out to suit the 3/8 inch diameter hub on the taper-fitted prop drive on this early example, so the tests were a bit scanty in terms of data obtained. Still, the available props did appear to cover the engine's preferred operating speed. Results were as follows, with indicated BHP figures based on the power absorption coefficients for APC propellers kindly supplied by Gordon Cornell:

PropellerSpeed (rpm)Implied BHP
12x8 APC4,7000.104
12x6 APC5,4000.129
11x6 APC6,3000.142
10x6 APC6,9000.122
Speed Demon Mk I Test Resulus, June 2010

Not a stunning performance by any means, but the engine certainly swings quite a large prop and moves a fair bit of air in doing so. It would certainly do a good job of flying quite a large sports free-flight model. Plus the handling and running characteristics are first-rate, especially for a large diesel. The above admittedly sparse figures imply a peak output of around 0.144 BHP at 6,400 rpm.

However, the above performance pales besides that of the Drone using the same props on the same day! The Drone was its usual good-mannered self, requiring only a couple of choked flicks on a full fuel line and a slightly leaned-out mixture followed by a short series of energetic flicks to start. It actually seemed to be having a "good" day, and produced some quite impressive rpm figures as follows:

PropellerSpeed (rpm)Implied BHP
11x10 Zinger w.5,3000.207
12x8 APC6,1000.226
12x6 APC6,5000.226
11x6 APC7,2000.212
10x6 APC7,8000.176
Drone Mk I Test Results, June 2010

Running was extremely smooth on all props except the 10x6, on which the engine (as usual) displayed clear signs of under-compression. But since that prop turns well above the engine's peak, this is of little consequence. The above figures imply a peak output of some 0.228 BHP at 6,300 rpm—slightly in excess of the maker's claim for the Drone albeit at the maker's claimed peaking speed. As I said, it seems to have been a good day for the engine!

On the basis of this comparative test using the same props on the same day, the Drone wins hands down. Clearly the Drone's combination of front rotary valve induction and more efficient loop scavenging makes a lot of difference. There's no doubt at all that anyone trying both engines would plump for the Drone with its greater power, lighter weight and equally easy handling. The later Mk II ball-bearing Drone would have had an even greater edge. However, indications are that the makers of the Speed Demon didn't stay in production long enough to find out.


If the Speed Demon had been released in Europe a year or two earlier, it would probably have sold quite well for a time at least. Quality was right up there, and the indicated performance is within sight of that found by Lawrence Sparey in his September 1948 test of the British-made ETA 5 diesel (0.1805 BHP @ 6,250 rpm). As it was, the Speed Demon had to compete against the established Drone diesel opposition in the US market, and as we have shown above it was strictly no contest—the Speed Demon never stood a chance in that particular race.

Still, a very cool motor! Not a powerhouse by any means and rather overweight, but a very well-made and dependable unit with excellent handling characteristics and buckets of torque as well as a few interesting design features. A very worthy effort by the unknown designer, even if it was somewhat distanced from its appropriate time and place.

As noted at the outset, there are no castings used anywhere and it would not actually be at all difficult to make a pretty good repro—they undoubtedly exist. I've thought of making one myself, in fact—just never got around to it! Plus there's less incentive when you already have two originals; in any case, Motor Boy Ken Croft already beat me to the punch by making a half-size (displacement) replica Speed Demon! This worked out splendidly and may in fact re-inspire me to have a go myself sometime, someday...



This page designed to look best when using anything but IE!
Please submit all questions and comments to

Creative Commons License