Now that I had the engine drawn out it was time to start converting the drawings to ones that could be used as patterns. Starting with the crankcase I made a copy of the file shown in my post above and then suppressed all the features that I did not need such as bearing hole, holes in the flanges for the cylinders to locate in, etc. I then added machining allowances where needed as well as draft angle so that the patterns could be pulled from the sand.

I decided it would be easiest to start by cutting two matching "halves" and then add the endplate with bearing boss as a third part so did a simple "extrude – cut" which left me with half a crankcase. I find that sometimes Alibre can have problems adding fillets to some internal corners once you start having more complex shapes joining at odd angles so left most of the fillets off the pattern file but will make use of the shape of the cutter to include them on the actual pattern.

The file was then exported to Fusion 360 to do the CAM, I ended up with 3 separate strategies to produce the shape. Firstly an adaptive clearing one which is basically a roughing out pass leaving 0.5mm stock which was done with a long shank 4-flute carbide cutter at 5000rpm with a feed of 450mm/min taking most of the material off with a 5mm high by 1mm deep cut then just stepping up 1mm for the fine roughing which can better be seen on the next but one photo. The blue lines are the path that the centre of the end of the tool follows when cutting the yellow are where it is moving between cuts

Next using the same tool and speeds but with a 4mm stepover the flat face had the final 0.5mm removed. you can see the stepped finish that the adaptive cuts left better in this photo as the tool paths don't get in the way.

Finally a long shank 4-flute 6mm carbide ball nose cutter was used with a "steep and shallow" strategy that was set with a fine 0.33mm stepover or "scallop" between cuts again running at my max speed of 5000rpm but feeding a bit faster at 550mm/min

This image shows how F360's simulator will leave the finished part, the bright green areas are where it has cut to the drawn surface, the blue is material it has not removed which you can see is basically all the places where I wanted an internal fillet. This method works well on the easily cut wood but you do have to watch when doing it on metal as the percentage tool engagement can go up a lot and chatter is possible, in these cases it is better to try and draw in the fillets with a radius say 10% more than the radius left by the tool.

A couple of mouse clicks and the thousands of lines of G-code that F360 created were on a memory stick that could be transferred to the mill.

It was now time to cut some brown stuff, however I was wary that the flanges and particularly the feet could be broken off easily if the grain was not in the right direction. So rather than cut from a solid block of timber I laminated some thin boards of Sapele together which would have a similar effect to using plywood with it's grain that runs in two directions. I glued my blank to a piece of MDF with Titebond placing a sheet of copy paper in the joint which makes it easy to separate afterwards. Sorry No video of the crankcase being machined but this is after the final cuts with the ball nose tool and I also added a 6mm hole for later alignment.

Early on Graham had suggested that the crankcase could be cast without the need for a core which would save me time having to make a core box and also keep casting costs from the foundry lower. By having the external draft angles going each way from the split line but the internal angle being constant through both haves from the open side of the crankcase this is possible. So the next thing to do was glue the outer face of one half to another MDF clamping board and then glue the two mating faces of the pattern together both again with the sheet of paper.

The inner cavity could then be roughed out with a long shank 10mm cutter and finished with a long shank 10mm ball nose cutter working it's way around the inside in a steadily lowering spiral. I also used the same cutter to form the inside profile of the end plate.

After separating the case from the clamping board and shaping the outside of the end plate the two were glued together and a fillet of filler added and sanded smooth. The actual surface of the wood needed hardly any sanding at all and came out far smoother than any of the hobby 3D printers would have done.

Stages were slightlt different to the traditional way where the main body would likely have been turned then turned flanges and the feet added on before blending it all with fillets but it seems to have come out OK and the photos I have seen of the castings look good.

I'm not sure if Graham will be making them available, if so it will probably just be a small batch for those interested so get your name on the list. At the moment there are just two sets of castings, one each for Graham and myself plus a few extra flywheels as they are a useful size for my 24mm bore engines.

Ron as well as the sizes Graham has given the whole crankcase would take up a block 135 x 115 x 65mm, completed engine will stand approx 240mm high x 210mm wide at the top of the heads.

Thanks for the interest Michael, I was going to do a shorter post for the flywheel but have done it in a similar fashion to the crankcase plus a video.

Needless to say the flywheel was always going to have 5 spokes right from the start although it was chopped and changed quite a few times, originally I thought that it would need something about 7" dia due to there being two cylinders which meant it would extend below the crankcase but Graham felt a smaller one would do so we settled on a of about 120mm dia with a wide rim and tapering oval section spokes. Like the crankcase the 3D model was enlarged to allow for shrinkage, draft angles and machining allowance added and a model of the half pattern exported to F360

First operation was to face the stock taking it down to the finished height of the edge of the rim. I've put the all the cutter details on the video so won't repeat them again here.

The bulk of the waste material was removed with a pocketing path leaving 0.5mm for finish cuts. While the 6mm tool was in the collet a finish cut was taken to bring the hub to height. As well as the blue and yellow paths mentioned in the last post there are some red ones which are where the tool ramps down into the work in a spiral which is kinder to the cutter than just plunging straight down.

Changing to a ball nose cutter the inside was finished with a "steep and shallow" path

Finally the tapered outer edge was finished with a "ramp" cut which has the tool gradually dropping down as it follows the profile, in this case I had it drop 0.75mm over one rotation which was approx 400mm length of cut.

I'm pleased with how it turned out, there was a bit of splintering of the top edge that can be seen on the video as I had used up some offcuts of boards that had a bit of snipe at the ends so that left a slight void not helped by the top layer ending up about 0.5mm thick. Nothing a bit of Upol filler could not sort out. This is the finish straight off the mill which needs almost no sanding.

Here are the two halves together with the crankcase pattern.

For those that like their facts and figures, the total distance that all the cutters moved was 64.45m, time taken was 2hrs 5mins and I was left with 29.8mm of the original round disc. Luckily F360 spits out the G-code which I'm rather grateful for as the talk of writing the code had always put me off and I hate to think what it would have been like to write the 50,003 lines of code Fusion tells me it took to cut the part. Approx 50% of that was the internal finish and 40% the external ramp.

No problem Ron, if ever you fancy making one then I have put drawings for a smaller 24mm bore single cylinder CHUKY together with a build thread on here or Graham may have some castings available but you best get that Muncaster finished first