Hispano-Suiza 8F (Part 8)

I’ve continued to push on with the two Hispano-Suiza engines, but the burgundy coloured one is going in a slightly different direction to the black.

After adding both cylinder banks to it was time to look at the induction manifold.

Previously it had been a single piece casting which worked quite well, especially in pewter. It was very awkward to remove the more delicate cold cast resin part from the mould though. The one installed on the black engine did suffer a little damage that had to be repaired. The arrangement also has no adjustment within it, so getting the cylinders exactly the same height and at 90° to each other becomes critical to a decent fit. I wanted something that was a little more tolerant of minor errors.

The obvious solution was to divide the casting up in to 5 parts, as it is on the real engine.  You quickly appreciate that the original designers must’ve done this to allow for manufacturing tolerances, amongst other reasons.

Hence a revised version was developed in CAD and sent off to be printed.

The revised parts fresh from the 3D print bureau with the print supports still attached.
The parts after a quick clean up. The manifold is symmetrical about the centre, so I only need one side printing to make the mould for both.

The link piece in the centre of the picture above is deliberately slightly over length so that it can be trimmed to a perfect fit on the engine, giving me the slight adjustment I wanted.

The cold cast components. The left hand part has yet to be polished.

The link pieces were glued to the central casting and the outer parts dry assembled on the engine using 1 mm bolts to locate them accurately. Then commenced the trial-fettle-trial sequence making sure I reduced each link piece equally in length. During the process I also filed off the locating pegs on the bases of the side pieces because they were more trouble than they were worth to be honest.

The result is shown below, and a it is definitely better fit than previously, though there’s no actual visual difference between the two arrangements.

The parts can’t be permanently fitted though until the carb has been attached as well as the plugs, ignition leads and their guide tubes in the V.

Speaking of ignition systems brings me on to the main change on this model. The black engine retains the magneto ignition system of the aircraft engine. The full Auto Engine Works marine conversion changed this to a coil/distributor version with dynamo. This makes more sense in a boat where swinging the prop is problematic to say the least, and a starter motor with charging system almost essential.

I wanted to model this arrangement on the burgundy engine, so it was back to CAD to develop…

Once again a set of .STL files were created and emailed to the 3D print bureau and 3 or 4 days later the printed parts arrived in the post.

STL = STereoLithographic which was the process used in the earliest 3D printers. The processes have changed, but the file type has stuck.

The parts with print support structure.
And cleaned up ready to make moulds from.

Moulds were taken and the parts cast in polyurethane resin. The dynamo and front face of the water pump were cold cast because they would have bare ‘metal’ areas I wanted to highlight. Incidentally,  exactly the same resin was used throughout. The only difference is that the cold cast items have 67%, by weight, aluminium powder added. The end housing was also cold cast simply because I had a little mixture left over after the other two, and it seemed a shame to waste it.

The parts cast in resin. The water pump has also been partially assembled in this picture and the bare metal part of the dynamo polished up.

The water pump faceplate came out quite well.

A quick bit of airbrushing later…

The black and silver components were sealed with a satin varnish.

The rest received a coat of burgundy gloss.

Once it had hardened I scraped back the faceplate of the water pump to highlight the company name.

The assembly also included a few photoetch components. The mask for them is shown on the right in the photo below. The large mask is for the exhaust system, including another drill template. I’ll come on to that saga in a future post.

The picture below shows it all assembled and mocked-up in-situ, but no plumbing or wiring has been installed yet. The pipe in the photo below is a trial fit, and I think it needs a smaller diameter one really. A bit inconvenient that, as the lugs are designed to take it. Guess which parts aren’t included in my CAD model…

The distributor assembly on the real engine includes a mechanical means of adjusting the timing advance. To replicate that it was back to CAD again. Then Photoshop to develop more etch masks. I cocked things up slightly on the latter. I forgot to add sprues to them and had to stop the etching process when it was 90% done and the components were still held together. Otherwise they would’ve fallen off the support and disappeared in to the bottom of the etch tank, never to be seen again. 

Two sets of advance and retard mechanism components. The top are fresh from the etch tank. The bottom ones are after clean up.

After the residual flash had been removed the parts were stained and assembled using 0.7 mm brass pins as pivots.

I couldn’t resist mocking the whole thing up again on the rear of the model. It won’t be fitted permanently until I have sorted out the cooling pipe size issue and have installed the spark plugs etc. in the V of the engine, which will be the next job.

The advance and retard mechanism assembled and in place at the rear of the engine.

As an aside: I mentioned last month that I might put some mild weathering on one of the two engines,  but wasn’t sure about the idea. To get an idea of how well it might work I carried out a trial on the scrap gearbox casting left over from when I dropped it.

Post weathering trial.

I’m reasonably happy with the result,  but need to include a bit more gloss black in the mix to give a more oily appearance. Other than that I think it’s on.

It’s been decided that the black engine will be the standalone desk curio and the burgundy one will go in the Miss Severn boat model. Based on that it’s the black one which will get the weathering treatment.

Hispano-Suiza 8F (Part 4)

Since my last update I have been busy at yet more CAD, and have drawn a period correct gearbox for the engine model. It is based on “The Capitol” Hispano-Suiza marine conversion gearebox manufactured by the Auto Engine Works of St.Paul Minnesota.

The marine conversion also included a subframe to carry the engine and gearbox as a unit.

The conspicuous blank square on the top of the gearbox will eventually have an etched name plate installed on it. I’ve already drawn up the etch mask for that.

Marine HS 3H Name Plate Etch Mask

A similar mask has been created for the engine name plate too (remember this is a licence built version of Hispano’s 8F engine, and is actually a Wright-Martin H3).

Wright-Martin Nameplate

Following on from my previous post I’ve also completed printing out the parts of the engine on my little 3D printer (XYZ Da Vinci Mini with wifi if you are interested) and used them to build a rough prototype of the model to check the basic  arrangements.

A quick word on the Da Vinci printer: It’s an entry level machine (AKA budget base model) and is good as far as it goes. The software with it has however driven me to distraction and intemperate language at times. I’ve suffered features randomly just not working and requiring it to be reinstalled to resurrect them etc. I have yet to get the wifi feature functioning correctly. It gives every indication of working seamlessly, but seems to corrupt my files somewhere during the ether transfer ‘twixt PC and printer. The prints suffer from sudden dislocations and the printer makes some very untoward noises during printing, so I have abandoned that entirely.

Provided I use the USB link to download the print file, calibrate immediately before each print, cover the print areas with Pritt Stick glue to hold them to the bed, and then don’t even look harshly at it until it has finished, it works fine. The quality of the prints you can gauge for yourself from the pictures.

As a result of seeing the parts in real life I changed the profile of the front of the sump a little. It is better now but still requires some attention. However I can achieve that in 5 minutes with some wet and dry paper, or probably half a day plus in CAD. Hence, I’ve left it for now.

Slab sided nose which wasn’t as evident in the CAD rendered images, but is obvious in real life. This has been rounded off in the model now, but still isn’t perfect.

As I changed print material colour during the printing process the prototype was a mishmash of black and dark blue. In order to see the features better and give it a uniform colour I gave the model a quick coating of grey primer from a rattle can.

The white bar is just some square section plastic to hold the units in line while I work on the mounting frame.

At least the engine is going to fill the engine bay properly in the model. I was concerned at one point that it might look a bit lonely.

I’ll cover the actual installation in the model in update on the Miss Severn build when there is enough progress to warrant one.

Having checked the basics of the model in real life, and made sure it actually fitted in the boat, I finally committed to getting the parts 3D printed by the professionals. Not a cheap activity at £300 for a complete set of engine parts. I wouldn’t have to build many of these models to have spent enough money to buy an upmarket 3D printer of my own. If I’m going to continue down this route that is something I’ll have to consider.

That said, the nice people at GoPrint3D provided me with some useful feedback on where the CAD models might need tweaking to optimise the prints, and the parts arrived at my door 5 working days after I pressed go.

The actual quality of the prints was slightly better than I was expecting, but don’t have the sharp detail of a good injection moulding. I also asked the print people to leave the support structures on as I think the amount of care with which I’ll remove them would be uneconomic for them, and I have more time than money. I am also a tightwad.

Apparently printing at an angle improves the print quality.

Removing the support structures and cleaning up does take quite a bit of time.

Stromberg NA-D6 carburettor. It took an age to draw as there are virtually no flat surfaces on it, and everything is blended together.

I’ve put markers in for drilling where any bolts need to go, but will use aftermarket bolts as they look so much more convincing.

The parts have all been cleaned up now and I am starting the process of making casts of them. My intention is to build 3 sets of everything, and two final models. One for the boat, and one standalone. The other parts will be spares in case of cock-ups.

Up to now I’ve only done white metal casting, but some of these parts will be in polyurethane resin. The actual resin I’m using is Xencast P2  which I obtained from Easy Composites. I’m making the moulds from a RTV silicone rubber from EA Silicones using their softest rubber so that I can achieve moulds with significant undercuts. The softness and elasticity of the rubber allows you to extract parts from it that you couldn’t in stiffer rubber.

This is all new to me though, so there might be a significant learning curve to go up. I spent half a day researching casting methods and techniques before actually committing to mixing any chemicals. Well, I actually spent the morning in bed drinking coffee and watching YouTube videos on casting, but that is research, isn’t it?

I’ve got some specialist kit on order to help the process in the future. Specifically a pressure pot to compress any bubbles in the castings to nothing while the resin cures, and a vacuum chamber to allow me to de-gas the silicone mould material before pouring and so limit the number of bubbles trapped in it. I’ve needed one for sometime anyway as entrapped bubbles in my RTV 101 moulds have been a problem in my white metal casting to be honest.

As a test piece for the polyurethane resin, and to get used to the materials, I chose the water pump for my first mould and casting. As usual a bund was made with Lego bricks. This works well with RTV101 but is marginal with the silicone. It seeped through the joints between the bricks far more and I’ll have to find another method for the future.

My first resin castings. The Lego bricks are only in place to hold the mould in register while the resin sets. That starts to harden in 2 or 3 mins, and the piece can be removed from the mould after about 20 mins.

I’ve managed 2 successful castings of the water pump, but the first 3 were scrap due to trapped air bubbles. A 40% yield isn’t good enough and the pressure pot is definitely required if I’m going to be doing this frequently.

As far as the rest of the casting goes, the intention is to attempt to do the block, sump, carburettor, and induction manifold in white metal. The rest of the engine will be in resin and painted. I say “intention” because some of the parts are quite complex with various thin sections, and I’m not 100% convinced that I can cast them successfully in metal using home equipment.

I have done a partial cast of the block to see how things might look, and also as a morale boster. It seems like I’ve done an awful lot of work these last few months and spent quite a bit of money without actually producing anything tangible. I cast the block as a solid lump because I ran out of material to make the mould for the hollow core. It’s also only a partial cast because I didn’t want to melt enough metal to cast the entire piece solid (it’d take a lot by my standards).

The bolts are after market items added to see how they look.

If I can achieve that sort of quality with a full hollow cast I’ll be happy.

While I wait for various bits of kit and materials to be delivered, I’m pressing on with getting parts ready to have moulds made.

Simple casting blocks have been added to the various parts due to be made in resin, and the more complex white metal parts readied to have the first halves of their moulds poured.

That’s all for this post. Next time I’ll either have some nice cast parts to show you, or a tale of woe to reveal.

Hispano-Suiza 8F (Part 2)

A very quick update this time as there is very little to describe or show whilst building up the CAD model of the engine.

Things are going well, though the carburettor gave me some problems as it is all curved and blended shapes. Only the magnetos and marine conversion gearbox need to be added. All the bolts, cables and wires will be added during the actual assembly of the model. There’s no point 3D printing them as they won’t look right in the end.

I may even get around to building things in real life soon.




Chris Craft 19ft 1940 Barrel Back – Dumas 1:8 (Part 6)

It’s past time for another post about the Chris Craft build. This activity was always intended as one to learn the new techniques required for marine models on. Because of that, it is effectively  pacing the Miss Severn build, and only getting attention when I need to practise or prototype something for that primary project. Continue reading Chris Craft 19ft 1940 Barrel Back – Dumas 1:8 (Part 6)