Puma Elfin 2.49 Repro
by Chris Murphy
|Click on images to view larger picture.|
The British Elfin 2.49 of 1949 must be in line for the award of 'most replicated diesel'. Since the original, produced by Aerol Engineering in Liverpool, there have been replicas produced by Dunham Engineering in the UK, Gordon Burford in Australia, Targos in the UK, and CS in China. RJL in the USA currently claim the rights to several Elfin models such as the 1.8 and 2.49 radial mount, and 2.49 beam mount, but like their other vintage acquisitions, have yet to do anything with them.
Another Elfin 2.49 replica has now joined the ranks from Hobby Club in the US. Their "Puma" range of engines have been associated with the sometimes roughly executed but workable Typhoon F2D combat motor and the seemingly endless stream of Elfin 1.49 replicas coming out of Russia. The latest effort is currently available (October, 2007) for US$129.95. This review is based on my impressions and experience with this engine.
First and foremost, this new replica is a well executed high quality production. Whilst adhering very closely to the original layout, careful attention has been paid to subtly strengthening parts of the original which were prone to damage without creating a bulked out look to the replica. The engine is based around a solidly executed crankcase casting of 29.5g, with a thin-walled bronze bush main bearing. It is unclear whether the bearing was cast in, or shrunk in during the machining process. Externally the finish is a soft matt light grey, obviously produced by fine bead blasting. Internally the case is fully machined with no 'as cast' surfaces visible anywhere. Cylinder and backplate threads are particularly well formed, and cylinder and backplate screw in smoothly with no trace of snatching or binding. The key areas are very solid. The main bearing OD is 14mm resulting a main bearing thickness of 3mm. The cylinder housing has a wall thickness of 1.85 mm. Likewise the rear mounting flange is over 4mm thick and the 3.6mm (9/64") clearance mounting holes are inset a good 2.5mm from the edge of the mounting flange on 26.5mm centres. The mounting layout is square, which is useful, allowing the mounted engine to be reoriented without having to re-drill the mounting holes in the model.
Cylinder liner is hardened steel with the traditional Elfin internal transfer grooves which terminate at exhaust port level for a zero timing overlap. The liner screws into the crankcase and seals on a steped, recessed flange in the crankcase, sealing without a gasket. The traditional Elfin feature of a huge amount of sub piston induction has been retained—the lower edge of the piston fully clears the exhaust port depth of 2.3mm at TDC. Exhaust port depth is generous by radial ported diesel standards. The piston is of 'modern' design extensively milled out on the inside to lighten, with a very shallow conical crown, which matches a similar profile on the contra piston. Gudgeon pin is 3.5mm dia, blind drilled 2mm ID from the front to save weight, and retained with wire circlips. The conrod is a massive affair with ball ends, a 4.5mm diameter shank, and a 4.5mm dia big end.
As the photos show, the crankshaft is a little odd to say the least. It has an 8mm journal, a 5.4 mm ID gas passage fed from a 3/16" dia circular shaft valve port. The crankweb is 3mm thick, but it has had the sides ground away parallel on either side of the crankpin! The constant width across the flats is 15mm. This highly unusual feature provides no more counterbalancing effect than a circular web. While it does reduce the weight, it also increases the crankcase volume. The crankpin is quite short and stubby at 5mm long by 4.5 mm diameter. The prop driver locates on the usual short taper section, beyond which the shaft diameter steps down to a short 8.5mm long 5mm metric threaded length. The main bearing length is 25mm and there is only a small amount of shaft end play.
The prop driver is a simple turning from alloy with a narrow 3mm knurl around the outer edge of the driving face. The annoying feature is the large 10mm diameter x 8mm long centre boss on the prop driver. This requires props to be drilled out to this diameter and depth—not always easy to do accurately. The short shaft threaded length limits choice to relatively thin hub props unless a sleeve nut is used. The maximum prop hub thickness is 8mm with the prop washer hard up against the face of the prop drive boss.
The updraft carb has a straight 5mm bore with no venture effect at all, relying on the restriction posed by the 3.5mm diameter brass spraybar. This is relieved to 2.95mm diameter in the vicinity of the jet. Needle is 1.5mm diameter wire, with a short 6.4mm tapered portion. A soldered split brass thimble completes the assembly.
A nicely machined, finely finned screw-on head and fin assembly completes the engine. This has an internal boss to provide additional comp screw thread depth—which is just as well, as the 5mm thread single sided compression screw is a relatively slack fit in the head threads. A locking bar will probably prove necessary. Comp screw is the traditional radial Elfin engine—slightly tapered, with the adjusting lever angled slightly upwards.
So the Puma Elfin 2.49 looks nice and is well made, but how does it perform? The short answer is very well indeed! I mounted it up and started running-in using the classical method: a large diameter, heavy prop for flywheel effect and lots of short 2-3 min runs allowing time to cool between runs. The first start took barely 30 secs. I just guessed at a needle setting—3 turns open—choked it a few times, gave it a small exhaust prime and had it running in about half a dozen flicks.
The first 30 minutes running was on a Master 11x4, using Model Technics D1000 homebrew mix. Revs were around 8200-8400 on this prop, running rich, but the comp set right. It prefers an exhaust prime for positive starting. It will start with finger choking, but starts are less positive. Contra-piston fit was good, just on the tight side of easy to adjust. Turns out that despite the loose screw thread, it did not need the locking bar I had made up just in case! Both needle valve and compression seemed relatively insensitive and the engine runs happily over a range of adjustment without either misfiring badly or sounding laboured.
The vibration level was remarkably low despite the lack of any effective counterbalancing as mentioned earlier, but backing off proved to be a problem. If any attempt was made to reduce compression once the engine had warmed up, the cylinder and fins unscrewed as a unit! After a full 30 minutes of running on the Master 11x4, I tried a Topflite nylon 10x3-1/2 and a Kavan 9x4 nylon. Revs at this stage were:
|Master Airscrew||11x4||8,400 rpm|
These look quite promising after only 30 minutes running on D1000 mix. With a bit more running and a change to D2000 fuel, a few more revs could be expected. The prop sizes tested represent the most useful sizes for F/F—with a 9x4 being likely to correspond to the BHP peak. A 9x5 may prove a useful size as well, with an 8x6 the obvious choice for C/L
There are a few debit points to offset the positives. The short shaft meant I had to machine up a sleeve nut to fit the 11x4 and reaming out prop centers for the 10mm boss is a pain. The unscrewing cylinder a nuisance too. The Russian Elfin 1.49 replicas suffer from the same problem. The cure is thread lock compound. Simply tightening up really tight seems to induce some distortion. When I tried this the engine became tight turning over bottom dead centre, a condition which was instantly corrected by loosening the liner slightly in the case, and gently snugging up firmly, as distinct from tightly. Just like the originals!
So overall, I'm happy. It is a quality production that appears to have no obvious vices and is built to last. At US$130 it is not that much more expensive than the CS replica—which is no longer available.
This page designed to look best when using anything but IE!
Please submit all questions and comments to email@example.com
Unless otherwise expressed, all original text, drawings, and photographs created by
Ronald A Chernich appearing on the Model Engine News web site are licensed under a
Creative Commons Attribution-Noncommercial-Share Alike 3.0 License.