Hispano Suiza V8

[Roger B]

An excellent start to a great project       

[RReid]

Thank you, Roger.

[bent]

Wow, this is going to be another neat build to follow!  Watching with interest...

 

[RReid]

I've been continuing to develop the design, and working through the decisions involved in the “to scale” vs. “practical at this size” and “doable with my equipment” debate.

One good example is the bore and stroke figures. My initial plan was use a 0.75” bore, and the rest of the design is scaled from that. The matching scaled stroke from the specs for the Hispano Suiza 8A would be 0.813”. I rounded this down to 0.8” for simplicity and because it also happens to be the stroke of my Curtiss Model K, so those offset turning blocks can be re-used. In fact, both crankshafts are very similar, with the H-S only having a narrower spacing between throws and wider crankpins to accommodate the knife/fork conrods.

When I looked more closely at the cylinder design with that bore though, I realized that if the cylinder OD was to scale, that bore would result in a very thin cylinder wall of only 0.033”, which concerned me. It also meant that I would not be able to seal the cylinder to the monobloc using 0-rings as I hoped to do. My solution is to increase the diameter of the cylinder inside the monobloc slightly, while leaving the lower, visible but short portion at the scale diameter. Meanwhile I reduced the cylinder bore from 0.75” to 0.688”, enough to enable the desired o-ring sealing. This bore also happens to be the same as the Curtiss. So now the bore that was the original basis for scaling has changed, but the adopted scale has not!

I have also opted for Mike's suggestion to run the cylinders all the way to the top of the combustion chambers, which is also consistent with Hispano Suiza's practice. However, they will be open at the top rather than closed, which is not, but avoids the need to form the valve seats in the top of the steel cylinders. Instead they will be in the aluminum of the monobloc, and so will be similar to that of my other engines.


Using the 3D printer in conjunction with the 3D modeling is not only fun, but helpful also. Makes it much easier to spot the little errors especially.

[crueby]

Fascinating project!  The CAD work is great, excellent way to find out those clearance/thickness issues. Looking at the 3D print, is it intentional to only have one bearing support in the middle of the crankshaft, and not at the other two centers? In the drawings back in post 23 it looks like each gets a bearing on the crankshaft?
 

[RReid]

Looking at the 3D print, is it intentional to only have one bearing support in the middle of the crankshaft, and not at the other two centers?

No, I've told myself I need to draw in the other two and just haven't gotten around to it yet. Good reminder, thanks!

[Vixen]

Hello Ron,

I am impressed with your use of the 3D printed models to verify and adjust your 3D design model and check clearances and fits etc. I wish those techniques had been available when I started on my Bristol Mercury and Mercedes Benz W165 engines. Back then, it was 2D drawing, cut the metal to see what it looked like. Now you can print the engine parts overnight and assemble everything next day and make and then prove any small changes before you commit to metal.

Two questions: how arel you planning gas seal the the spark plugs in the cylinder/ monobloc joint? And how are you proposing the attach the four cylinders into the monobloc? Have you an image showing the proposed arrangement at the bottom of the monobloc?

The Hispano Suiza was a world leader when it was designed. Your 3D printed model clearly illustrates how the two monoblocs provided most of the structural rigidity for the engine. The crankcase and crankshaft look frail in comparison. The designers got it right and produced one of the lightest and most reliable water cooled engines of it's day.

Cheers   

Mike

[RReid]

I wish those techniques had been available when I started on my Bristol Mercury and Mercedes Benz W165 engines. Back then, it was 2D drawing, cut the metal to see what it looked like.

Hello Mike,
I've done just enough of that to know how hopelessly (and shamelessly) spoiled I am.
On the other hand, I'm still machining by CCC (Crank Count Control), which is why that crankcase will likely be a pieced up brass fabrication.

Two questions: how will you planning gas seal the the spark plugs in the cylinder/ monobloc joint? And how are you proposing the attach the four cylinders into thee monobloc? Have you an image showing the proposed arrangement at the bottom of the monobloc?

I plan to make the cylinders a good close fit at that upper end. They will be tapped for the spark plugs in place, along with the monobloc, so the plug threads will extend into the cylinder wall. There will be only one through hole per cylinder (although the drawings show two, that is just for my convenience). Will that prevent all leakage? Maybe not, but I hope it will be minimal enough to not be problematic. If I have to I can resort to a bit of goop at that interface. Any suggestions?

The bottom of the monobloc will have a separate plate that the middle flange of the cylinders will screw into via flush fitting SHCS. There is a second o-ring groove in the cylinder just above that flange to provide sealing for the lower end of the water space. This plate fits up into a recess machined into the bottom of the monbloc, so that it too is flush fitting. It will then pinned into place with aluminum pins which will be filed & sanded smooth and painted over, making that joint semi-permanent. This is my way of preserving the look of the monobloc while providing machining access to the inside and a means of sealing the cylinders in place. I hope the images below well help to clarify this. The row of 5 holes along the lower edge of the monobloc are where those fastening pins will go.

[Dave Otto]

Interesting project and nice work so far, I'm looking forward to see it progress.

Dave

[Vixen]

Hello Ron,

I can see you have put a lot of thought into the monobloc construction. You were probably doing that long before I sent you the HS handbooks.

Here are some ideas for you to ponder. They are not intended as recommendations, just ideas you may already have considered.

The pinned base plate of the monobloc could be sealed with one of the high strength one-component polyurethane windscreen adhesive-sealer such as 'MANNOL 9910 windscreen sealant' or 'Loctite TEROSON PU 8597 HMLC'. This material could also be considered for the cylinder to head seal above the o ring. This polyurethane likes a bit of bond thickness, so you may need to machine some extra clearance or shallow groves to accommodate the goop. The beauty of a watercooled engine is nothing (except the gasses) ever gets very hot, so wont stress the sealant.

The spark plug threads will need to be in the monobloc wall, rather than in the thin cylinder wall.

You could consider a thin PTFE head gasket, compressed by the force of the four screws in the middle flange of each cylinder.

As an alternative to the pinned baseplate, could you machine the monobloc from a single piece. You could bore out the water cavity as four intersecting cylinders by entering a suitable wooduff cutter or boring head through each of the cylinder holes in the base and machining outwards. I did this on one of my earlier V8 engines. It worked well and the water cavities completely surround the cylinders and intersected each other.



There are many ways to skin a bear, use the way that suits YOU the best

Cheers

Mike

[RReid]

Hi Mike,
Thank you for offering up your thoughts and suggestions. Indeed, some do parallel my own list of ideas, but some are fresh. I especially like your suggested method for making a one piece monobloc. That is an approach I hadn't thought of, and is worth considering in detail. Even if I don't use it here it will be added to my toolbox!

You are right that I have been thinking about the HS previously. The idea of doing a SOHC cam engine to accompany my DOHC Halfa engine appealed to me, and the Hispano Suiza 8 series appealed in particular for its historical significance, its elegance, and the contrast between the more evolutionary technology of the Curtiss engines with the brilliantly fresh ideas of Marc Birkigt.

I did consider some other SOHC engines as project ideas, the only one of which progressed very far as a 3D model is this one. I'll leave you or anyone else to guess at its prototype:

[RReid]

More time has been spent of late weaving electrons on the computer than cutting chips, but a start has been made on fabricating the core of the crankcase. At this point that's not much what it looks like, but after some silver soldering and machining it should, I hope, start to emerge. It's supposed to be stinkin' hot today (again) though, so I think I'll put off the soldering and find something else to do.




One design element I've spent quite a bit of time on is the ignition distributor and its arrangement on the engine. I wanted to have at least a working plan before I get too far along. The full size engines normally had a pair of 8 cylinder magnetos mounted at the rear behind the cam drives, sometimes mounted to a horizontal platform, sometimes on a 45  inline with the cylinders. My first hope was to make two 4 cylinder distributors, each with four dummy posts to simulate an 8 cylinder version, and mounted as the originals, maybe even disguised somewhat to look magneto-ish.

I have a good article by Dwight Giles and Mike Rehmus on the subject of model distributors in an issue of Model Builder Magazine, which also includes dimensioned drawings. They recommend maintaining a 1/2” spacing between posts in the distributor cap to avoid arcing, which then sets the basic dimensions of the cap, resulting in a distributor for a V8 with a nearly 2” diameter. Even my proposed 4 cylinder version, while smaller, is still not small enough to really look right at my scale, especially since two are needed. So my current plan is to just use a single V8 distributor, but mounted low and away behind the engine to minimize its visual impact.




Mixed in amongst this was a mini-side project. My 3D printer was shipped with a random “gift” included, which in my case was a small, round, LED light fixture, for which you can download any number of housing files suitable for printing to mount it into. Instead though, I chose to draw up a mount that would afix to my mill by sliding over the motor mounting post, giving me a rather nice little work light.

[Kim]

Looks like a great start, Ron! 

And that light looks mighty nice too.  You can NEVER have too much light! 

Kim

[RReid]

Thanks Kim!
Yeah, the extra light is really nice, especially on that side where the motor tends to throw a shadow.