Could I try an IC engine?

Hello,

I am not a complete beginner, I have built and operated a 2 inch scale traction engine. I have been vaguely thinking about trying a 4 stroke IC engine, however it seems that the precision needed is at a higher level than for a steam engine. In particular, I am wondering about making the camshaft. I have read some posts on this forum, including the post from David K and the replies to him back in 2019..

My specific question now is about the precision needed for angular location of the cams relative to each other and to the register which locates the shaft in the timing gear. I do not have a milling machine or a rotary table. I reckon I could have a go at filing the cam flanks to a horizontal guide, using a micrometer, with the shaft in the lathe chuck, but setting the angular positions accurately would not be possible without some kind of dividing head.

I wonder how camshafts were produced in the 40s and 50s when milling machines and rotary tables would not have been possessed by many amateurs.

There are other precision tasks which would be difficult for me, such as making the valves tight, and getting the crankshaft bearings in alignment..

I would be interested in any comments, thanks.

Paul

Hi Paul,

Have a look at Andrew Whale's "hit & miss" engine. It might not be exactly what you're looking for, but an interesting IC project non the less.

In truth, as long as one is not looking at a high performance ("racing" )  engine, they don't need to be very accurate to get a running engine - maybe 5 degrees, or so? You could probably get away with 10.

There's a way of cutting certain cam profiles fairly accurately on a lathe using a series of arcs which is straightforward, but a little tedious.

Edited By Andy_G on 11/02/2021 17:25:29

A lot will depend on the type of IC engine. A high performance multi cylinder engine will need it's cams to be quite accurate and all lined up correctly. On the other hand a slow reving old school engine or Hit & miss type could use an atmospheric inlet (opens due to vacuum caused by piston going down) so only needs an exhaust valve. The single cams are not too fussy on timing so could be filed up and just use a grub screw to locate it on the cam shaft or peg.

Valve sealing is no different to getting a regulator, clack or steam valve to seal on a steam engine.

Thanks, gents, for the replies, they were most helpful for me.

I was considering the Westbury Seagull two cylinder engine, available as a kit of castings from Hemingway. Then, doing a little more research, I came across the excellent diary of building this engine by Charles Lamont :

**LINK**

This description of his workings, all to tenths of a thou, I found pretty terrifying, with so many problems in making the cams, for a man who is evidently highly experienced. He ended up building his own camshaft grinder. Mr. Lamont is to be applauded for setting out this diary and for his sheer persistence, but the project is clearly beyond my horizon .

Even if I completed the engine I would then have to produce the auxiliaries, water pump, radiator with fan, coupling or pulley for power takeoff, and some kind of clutch, plus something for the engine to drive. Put it in a boat? God forbid!

Paul

Try the Westbury Jig. It is lathe orientated.

Charles Lamont sets himself an impressively high bar!

I've also looked in depth at the Seagull and like you, discovered it is not at all straightforward due to the number of errors in the drawings and the marginal nature of the castings.(See also https://modelengineeringwebsite.com/Seagull_revisited.html ).

However,

As far as I know, there is no requirement to harden and grind the cams for this type of engine (although one may wish to for authenticity).

The end of this video shows how cams can be machined on a lathe:

You may not have the skills required now, but you will by the time you finish, possibly discarding a few pieces along the way. That's how it works for most of us, we try something ambitious and learn as we go.

I think the Westbury jig is a way of machining cams in the lathe, if someone has a reference I probably have the magazine articles in my shed.

John

Maybe start off with a 2 stroke model engine. A model diesel engine would give you a good feel for 'fit's''. It's not too hard to learn but takes patience.

The cams in a 4-stroke engine take a bit of a beating, as do the cam followers and the rockers so unless it's a display engine (demo runs only) then consider case hardening and grinding the relevant parts if you want it to last more than an hour.

* Danny*

I’m not a model maker but as I see it, the requirement might be to measure to tenths, but machining to the next whole thous above is ‘all’ that is needed.

As long as (male fitting) parts are not made too small, those last few tenths are a matter of grinding (abrading), polishing, lapping and honing.

Reducing the outside of a part by a few tenths is far easier than increasing an internal dimension, although cylinder lapping tools are good for preparing the inner surface to be ready for finishing. As long as, say two cylinders, are within a thous (or so?) on diameter it makes no difference as long as the pistons fit their relevant bore.

Start with a simple one and progress from there. With care, the Seal should be doable. It does not turn at a million rpm and parts - other than the castings - can be made again if you are not happy with the first (or second) attempt.

Confidence in your machinery is an important factor as well as your ability. If you are determined, you can do it. There is -plenty of help/advice available on the forum if you need it. Go for it, I say.

Also, perhaps have a look at Allen Millyard’s videos on youtube. Quite basic machining facilities but it does not stop him making a crankshaft for a high revving 6 cylinder engine (built up by hacksawing apart, and then welding together, a couple of 4 cylinder motorcycle engines). The crank build used very little from one of the original engines.

You could pick a design where each cam is made as a separate part, hardened and then Loctited onto a silver steel shaft. Avoids any distortion when hardening, less to remake if a cam goes wrong, simpler to machine individual cams. Nemet Bobcat springs to mind as a two cylinder and being barstock no undersize or expensive castings to worry about.

Might even be possible to use the single cam method on the seagull

Edited By JasonB on 14/02/2021 07:33:09

There's several discussions and approaches here that might be useful;

ModelEngineNews Cams for Small Fourstrokes

..including a description of Graham Meek's attack on the Seagull cams..

Totally agree with John Olsen. Push yourself with a project and you'll learn as you go. I have made this and that over the years and have a lot of general skills but since I went the route of a big 'Challenging' project (see Ongoing projects...) boy have I learned a massive amount - including how to 'make do' with tricky parts and not having the 'right' equipment or tooling. Scrap box got very big over the last few years and there have been times when I've felt overwhelmed by problems and at the point of calling it a day, but persevere and you get such pleasure from succeeding!

Best of luck whatever you go for. Keep the forum posted as you go along. Regards Mike

Is that still the case with something as slow revving as the Seagull? (Asking, not arguing! )

FWIW, the Seagull's cams are machined as pairs and 'loctited' onto the camshaft.

Yes you are quite capable. There will be a few 'trial' pieces as you learn but you will get a working engine. I would recommend an overhead valve four stroke as this separates the various problems. The cylinder on a two stroke has al the difficult parts in one piece. One mistake and you start again.

The key is the quality and fit of the bore and piston. This will probably require a couple of goes (at least). Cams are not a real problem for small engines unless you are going to race or fly them. Most model engines run a few hours per year. Cams and followers can be made from silver steel (drill rod) and hardened and polished. Mine are made in a series of steps on a mill but careful filling to a paper template should work ( a lot of experiments on full size engines were made in this way).

Give it a go, there will be plenty of help and advice available on the way.

This is my first cam milling setup using a vertical slide. The blank is indexed using a 60 tooth changewheel with a paper degree scale stuck on. The depth of each cut is set using a DTI at the back of the cross slide and the slides are 'locked' with a G clamp. I now do much the same on a milling machine. The maching marks are cleaned up with a swiss file before hardening.

Thanks to all of you for the encouraging and helpful messages with plenty of useful information. It's still very much a thought experiment for me.

I think I would be able to produce a camshaft using the turning jig method, without needing a rotary table, milling machine or a grinder. It is reassuring to see that a decent sized paper disc marked out in degrees is good enough to be practical, though I am surprised that the one inch disc with marker pen lines, shown in the video clip from Andy G. would be precise enough. Some kind of pointer has to be registered to the shaft, to operate this system. The complicated method used by Graham Meek in the link from Diogenes, or that used by Charles Lamont, would be too much too much trouble for me.

I understand Jason's point on making the cams in pairs, on the turning jig, which I could case harden, and polish, then mount on the unhardened shaft, though I would need to devise a method to set the angular positions of the pairs relative to each other and to the shaft keyway. With all four cams made together this would be taken care of. The pairs would have to be drilled through after making, then the excess shaft length cut off, before hardening.

I could use milling in the vertical slide as shown by Roger B, though this would need a reducing sleeve with two keyways to register the change wheel to the shaft, and 60 teeth would not enable the 130 degree setting for the exhaust cam flanks. But, in principle I could do it.

So each method has its pros and cons, with balances of time, effort and precision to be decided.

I understand the point about the fit of piston to bore, it has to be much closer than in a steam engine. This is one reason why a diesel engine would not be my choice because this fit has to be even closer than for a four stroke. And not only do I need a close fit, but it has to be dimensionally accurate, to use piston rings as provided in the kit.

My aim would be to produce an engine which could run on a table top setup, and, importantly for me, would drive a load which could be regulated. No ideas yet on what would be practical for that. Thinking is a lot easier and quicker than doing!

Paul

Sorry not kept up with the thread but the Westbury jig is in ME 12th July 1962 pages 56/7, if you need copies I will scan and send plus a photo of my jig. Note this is for cam flanks.

The method I use mills the cam using a series of tangents so various angles can be used. I set them out on a spread sheet with the angle and depth of cut, ticking each one off when complete. There are various programmes around for generating the steps from the cam data.

This engine was built on a minilathe using a vertical slide. I initially used an O ring to seal the piston but later made a cast iron piston ring.

**LINK**

Sorry - I thought you were referring to an engine design (My fault!). I'm guessing it is the same one as in the Seal construction article - Link and also the type used in the video above.