NOVICE’S WORKSHOP – FIRST ENGINE BUILD
Matthew continues his series for novices describing his first model engineering project – an oscillating engine built when he was 12 years old. This month he makes the flywheel.
he flywheel is (probably) the most iconic part of any stationary steam engine and it is very important to get it to run true; achieving this demonstrates the engineer’s care and attention to detail in their model engineering.
I was given a magnificent piece of phosphor bronze (it needs a display case in its own right) to make this part from. This material was lovely to machine and provides a fine golden lustre when polished (I could write poetry about this stuff). Though brass would work just as well, phosphor bronze will result in a more aesthetically pleasing model.
Having sold grandma and obtained your phosphor bronze, it’s time to set to work on it. To start off you will need to find your hacksaw and get your muscles working, by cutting off a length of your chosen metal. It should ideally be quite a bit bigger (in both diameter and thickness) than the finished dimensions, as you will have to turn it down to size and it needs to be gripped in a lathe chuck.
You will also, inevitably, cut at a slight angle, when slicing the material to approximately the correct thickness and this fact needs to be taken into account when judging the cut. It is better to waste a little material (in machining off more thickness than might, ideally, have been the case) than to waste a large amount of material, by making a mistake with the initial cut.
When I did this, it took me about half an hour but if you wanted to put your feet up and not do any work you could use a power hacksaw or a band saw (we have just achieved this level of luxury in our home workshop). But be warned; these can cut at a slight angle (particularly band saws) and you would not want to ruin the lovely piece of metal that you sold grandma to obtain! If in doubt, cut a piece of scrap wood first and judge the error based on that.
Machining the flywheel
Once you have had a rest and a cup of tea (an essential part of any model engineering project) it is time to head over to your lathe. I would recommend using a three-jaw chuck and a mid-size or larger lathe – a mini lathe won’t cut it, quite literally. I used a Myford ML7 which I found was very capable for everything except the parting off (back to the hacksaw for this bit). The main issue (before I get letters) was the lack of availability of a suitably-long parting-tool blade at Hereford SME, at the time I needed it, and not the capability of the Myford.
The trick to machining phosphor bronze (I have found) is to have a mid-range spindle speed, a few hundred rpm, and to take light cuts. Note: do not forget to tighten down your tool or your work will be ruined; a sad story, but one spotted in time – this is easy to forget when adding shim in the toolholder to set the correct tool height (see the panel on the next page).
Another point is to remember that the speed at which the lathe is running is the speed at the centre of your work, not on the outside; check your surface feet per minute, as you may find that you need to slow your lathe down. There is an article on Wikipedia which explains this quite well (https://en.wikipedia.org/wiki/ Surface_feet_per_minute).
The first step it is to ‘clean’ the outside and end of the round bar by taking a series of light cuts using a standard left-hand turning tool
(Photo 6). A tool ground for brass is ideal for turning phosphor bronze.
If you are not familiar with tool-grinding (and, as a beginner, this is highly likely to be the case), then there are a range of ‘indexed’ tools available very inexpensively these days (Photo 7). These incorporate a replaceable insert which is already shaped with the correct cutting angles.
The two main tools you will need in this project, a left-hand turning tool and a parting-off tool, are shown in
“Be warned; these can cut at a slight angle and you would not want to ruin the lovely piece of metal that you sold grandma to
obtain...”
Photo 8, together with a ‘neutral’ tool, which may also prove handy – as we will see shortly.
The end of the bar needs to be ‘faced-off’, in other words a cut needs to be taken across the end of the bar (or, more likely, multiple cuts). The aim here is to achieve a smooth, even, finish across the whole of the end face of the bar.
For the final cut (the ‘finishing cut’), a little more care is needed in order to achieve a nice, smooth, finish. Firstly, ensure that this cut only removes a small amount of material. For example, if you need to remove 2mm in total, consider taking three 0.6mm cuts, followed by a 0.2mm finishing cut (most lathes should be capable of this severity of cuts).
If you have a beefier lathe, such as the Harrison M300 we have in our workshop, then fewer, deeper, cuts can be taken. The same principle applies, however: take a lighter cut as the finishing cut. In this case, since we are not yet at the stage of having to worry about precise dimensions, take cuts until the whole surface is shiny (indicating that all undulations, and any ‘skew’ in your hacksaw cut, have been removed). Finally, take the light ‘finishing cut’, as described below.
The second point needs a little practice – try to ensure that the tool does not stop in its lateral travel across the workpiece. This requires a little skill and dexterity (and two hands on the handwheel). The aim is for your second hand to start turning the handwheel before your first hand has run out of twisting ability, your first hand then taking over when your second hand is close to its rotational limit and so on. With a lifetime of practice, you will become quite good at this.
Why take so much trouble? The answer is that if the tool is allowed to pause at a point on the workpiece, it will leave a tiny score mark at that position. You will notice this mark and it will annoy you, being very obvious next to the beautifullysmooth finishes on either side. You can, of course, abrade this away with emery paper, but it is much better (and more accurate) not to have to do so.
Next, do the same on the outer surface of the bar – again, the aim is simply to achieve a uniform, smooth, surface, at this stage. The same hand-wheel technique is needed here.
Having achieved a uniform ‘clean’
finish, measure the outside diameter. If you are using callipers make sure they are exactly perpendicular to the machined outer face, and that the faces of the calliper blades are scrupulously clean, as neglecting either will lead to an inaccurate measurement (I am talking from experience here).
You are then ready to turn this down to the correct diameter (Figure 3). It is, of course, not necessary to apply a ‘finishing cut’ before turning down to the specified diameter, but it is good practice in learning and perfecting this aspect of machining.
Enhancements
You could leave out this next step out, but I think that the part looks a lot better with it – an indent in the face of the flywheel adds aesthetic interest to the finished model and is further, useful, lathe practice.
For machining the indent, you will need to change your tool; a reground parting-off tool will do the trick – you want something almost finger-like: narrow, with not too much depth but sufficient to provide strength.
An alternative is to use a ‘neutral’ turning tool (see Photo 8) – this is a simple option for a beginner (and one not yet comfortable with tool grinding!), but will result in sloped sides to the recess. This is not a problem, just an aesthetic difference from the prototype.
Don’t forget to slow your lathe down, otherwise known as putting it in back gear, if you do not want chatter and the resulting poor finish. If you are still struggling, invert your
PHOTO 7: Indexed (on right) and ground (at centre and left) lathe tools.
PHOTO 8: Lathe tools of type used in making the oscillating engine – lefthand turning tool (front), parting-off tool (centre) and neutral tool (rear).
tool and move your toolholder to a rear tool-post, if your lathe supports such an option, then run your lathe in the opposite direction – the chatter should lessen or be eliminated.
The chatter lessens since when the lathe is running in the ‘normal’ direction the tool is trying to force the work and chuck up which results in chatter (if there is any play in the main bearing). If you run the lathe in the opposite direction and invert the tool in a rear tool-post (if you have this capability) the chuck and work will be forced down – a position they are already in, due to gravity. This improves the finish.
Finally, centre drill the flywheel, then drill it to 6.0mm diameter (Photo 9). Make sure your centre drill is quite a bit smaller than 6mm in the tip section, so you can use the chamfer on the larger diameter of the centre drill to help the larger twist-drill to locate. Finally, use a, preferably HSS (highspeed steel) tap to tap the hole to M7 for the flywheel shaft.
Don’t forget to back-wind the tap and use plenty of oil or cutting compound so the tap does not catch and break. A turn or two forward, followed by half a turn or so backwards until the tap is free – you will feel the swarf ‘break’ – keep doing this all the way through. The use of a live centre in the lathe’s tailstock is a handy way of ensuring that the tap is in line with the hole, if your tapwrench has an indent in the rear for such a purpose (Photo 10).
The flywheel now needs partingoff from the remainder of the bar from which it was made. An indexed parting-off tool is ideal for this, as it will usually have a very long and uniformly narrow blade – something quite difficult to achieve by grinding. I didn’t have this luxury at the time and so had to use a hacksaw.
Note that whilst the part can be held in the lathe chuck for this operation, resist the temptation to use a hacksaw as a ‘poor-man’s parting off tool’, in other words with the lathe under power. This is quite dangerous, as it is easy to catch your fingers on the jaws of the lathe chuck and receive a very nasty injury! It is also easy for the saw to get caught by these same jaws; not a happy prospect.
If you use a parting-off tool, the finish may well be good enough to not require further machining, perhaps just some work with emery paper. If the hacksaw technique is adopted, the part will need returning to the chuck and facing-off with a left-hand turning tool. Spacers may be needed to ensure that the part runs true and sticks out beyond the chuck jaws – these spacers should be removed before starting the lathe (unless they are truly captive) – unedifying holes in the ceiling or, worse, the lathe operator, could otherwise result...
If you do need to turn the piece around, in order to face-off the reverse side, add some copper or brass shim to the chuck jaws in order to stop them from marking your beautifullymachined surface (see the September 2019 issue of EIM for details on how to make something suitable).
An alternative option is to make/ use a mandrel – I have just used one in making the wheels for the tender on my current project, a 5-inch gauge Manor. However that is possibly a step too far for this beginner’s series. If at all possible, avoid the use of a hacksaw and invest in a decent parting tool!
Now this part is complete, have a moment of quite reflection; you are one step closer to a steam engine!
Flywheel shaft
Silver steel rod was used for this part; this is the ideal material for the job as it is easier to machine than stainless, but it still has a good degree of rust resistance. If you go on to larger things in model engineering, then you will use this for axles and the like, so it is a good material to be familiar with.
As noted above, I used ‘scrap’ materials which were to hand and hence designed the engine to use a 7mm diameter silver-steel shaft which was lying around, thus saving me having to machine to this diameter – silver steel is usually supplied pre-ground and smoothed, to a good degree of precision. The diameter itself is not critical – anything between about 5 and 8mm could probably be made to work, with suitable modifications to the holes/screw threads in the adjoining components.
The first step is to set the shaft up in your lathe, again in a three-jaw chuck, but with only a small amount of ‘stick-out’ from the chuck. The end can then be faced-off using a normal left-hand turning tool, not forgetting to add a small chamfer to the edges/ corner, as this helps the die to get seated on the work when it comes to cutting the M7 thread.
The shaft can then be threaded using an M7 die; you can use this either in a tailstock die holder or if you do not have this luxury, mount it in a normal die holder, push the tailstock up to the back of the die holder, then apply (gentle) force using the tail-stock
“Whilst the part can be held in the lathe chuck
for this operation, resist the temptation
to use a hacksaw as a ‘poor-man’s parting-off
tool’...”
handwheel whilst rotating the chuck – using the chuck key is usually the easiest way of doing this (Photo 11). Finally, part off your work to the correct length (Figure 4) and then chamfer this end slightly (Photo 12).
The flywheel should now be assembled onto its shaft, to check that the threaded end of the shaft doesn’t protrude from, or recess too far into, the flywheel boss. In the event of the former, the complete assembly can be put back into the lathe and the excess shaft trimmed with a conventional left-hand lathe tool. In the event of too much of a recess, then the cure is to thread the shaft a little further down its length. All being well, you should end up with something that looks like Photo 12 and Photo 13.
Thread concerns
There is a slight problem with the above method, which arises due to the shape of the die. In order to begin cutting a thread, a die must have a small taper to ensure that it sits centrally onto the shaft it is intended to thread and then only gently start to cut the thread. This end should obviously be used for the initial threading operation.
Once the thread has been cut to its end-point, the die can then be turned around and the opposite face now used, to ensure that the thread is cut to its full depth all the way to its end. Whilst this works well, in most cases, it is not perfect (although it was the method I used in the construction of my oscillating engine).
The problem is that the end of the thread is not very well defined and is often just visible where it enters the internally-threaded part (flywheel in this case) to which it must attach. This is not really obvious in Photo 12, but may be in some applications.
A better solution is to apply an undercut to the end of the thread, as shown in Figure 5, thereby giving the thread a clearly-defined endpoint, before the visible join between the flywheel centre-hole and the flywheel shaft. The flywheel will still sit at the same point on the thread/shaft, as it will come up against an unthreaded portion of the shaft, while the removal of 2mm of thread out of a total of 12.7mm will not impact the strength of the thread in any meaningful way, so there is no real downside to doing this.
This type of undercut is easy to achieve with a thin (2mm) parting-off tool, plunged to a just-sufficient depth to remove the thread entirely (a little extra depth is not a big issue). An even better approach is to arrange for the shaft to be thicker at the point where the shaft and flywheel abut, thereby creating a solid face against which to tighten the flywheel and a very well-defined end-point at which it can sit. This could simply be achieved by using an M6 thread, say, with the original 7mm diameter shaft. As I said, I didn’t do this on the prototype and it still looks pretty good, this part being difficult to see anyway as it faces and lies close to the upright support. I don’t think the judges noticed...
One point to note is that this approach can have a downside if it is applied to a wheel on an axle, say, which will be subject to large lateral forces (such as a train full of well-fed passengers). The diameter of the shaft has been reduced, thereby reducing its strength, and the sharp corner created by the parting-off tool concentrates the stresses in that corner, potentially leading to cracking and failure at this point. Neither of these is a concern in a small oscillating engine, however it is worth noting them for when you move on to more ambitious projects.
“The sharp corner created by the partingoff tool concentrates the stresses in that corner, potentially leading to cracking and failure...”
n Next month Matthew makes the crank and other components. Part 1 of this feature was published in last month’s issue of EIM – you can download a digital back issue or order printed copies from www.world-of-railways.co.uk/store/backissues/engineering-in-miniature or by calling 01778 392484.