Porsche Inspired Flat 6

[SailplaneDriver]

I've been unable to find plans for a Volkswagen or Porsche boxer engine so I am going to attempt to design my own. This is likely going to be a long journey if I can even get it done. I have access to some torn down engines through a local repair shop. I did some preliminary measurements and lots of photos. I have also been researching what info there is on the web. I have drawn up a scaled down version of the crankshaft. Right now it is a direct 1/4 scale of what I measured. I did not include the gearing on the rear of the crank for the cam or distributor nor did I include the bolt holes for the flywheel. I'll add them later.

This is my first attempt at a design and I need some help from you guys with all the experience. Things start to get real small when you scale them down and I expect it is not appropriate to keep everything exactly to scale for structural reasons. The crank webs wind up .090 thick which seems pretty flimsy to me but there is a moderate amount of overlap between the main and rod journals. Do you think this design would work and be constructable? Should I increase the web thickness and decrease the journals? Any input would be appreciated.

 

[johnmcc69]

Very nice, I look forward to following & seeing more.

John

 

[LSEW]

Sailplane, it's good to see someone set out on a serious design project. You have a long row to hoe, but it will be very rewarding.
You have about what I had when I did a scale design of the Forest 1888 marine engine, except I didn't have the original or any parts.
Your question about the thickness of the crank webs is interesting. Since there are so many crank bearings, I tend to think the .090 thickness may be ok, but since this is likely to be a high RPM engine, the thinness of the webs scares me a bit. Can you do a force analysis? I would suggest you use tool steel for the crank (drill Rod).
I suggest that from the start you do scale fastners and try the fastner models on your design. I have experience with too many designs (some of mine included) where the bolts just won't fit in the allotted space.
It might be a good idea for you to consider making your model a bit larger. This will help with the mechanics, and allow you to include more authentic detail.
Good luck and keep posting
maury
Lone Star Engine Works Retired

 

[SailplaneDriver]

Thanks for the input Maury. I did the crank design in Fusion 360 so I should be able to do a force analysis - but - I am an electrical engineer. Much of this mechanical stuff is outside of my education and background. I assume I would need to estimate the mass of the piston assembly as static load and the force of combustion as dynamic load. The question then is how much dynamic force? I can guestimate the static load. Then next thing is to figure out how to insert the appropriate forces into F360.

 

[Brian Rupnow]

There is a guy over on Model Engine Maker forum who has just redesigned my flat 2 cylinder as a flat four cylinder and it looks really good.
http://www.modelenginemaker.com/index.php?topic=9156.msg202975;topicseen#new

 

[SailplaneDriver]

Thanks Brian. I'm really looking to make a Porsche clone if possible. Likely I have bitten off more than I can chew but there is only one way to find out. The overhead cam arrangement on this design will likely defeat me. The rest looks doable.

 

[SailplaneDriver]

This simulation can get quite complex. The F360 simulation only does static forces. What would be the worst case, top dead center, 90 degree crank angle or something in between? From what I can find on the web, a 1000 psi conbustion pressure is probably reasonable. For the displacement in this particular design the combustion force per cylinder would be a bit under 500 pounds. I would expect that one end of the crankshaft would need to be fixed to model the torsion within the length of the crank and I am modeling all the main bearings as fixed as well. I don't really want to model the compound forces of the piston pushing the rod pusing the crank if I can get away with it.

 

[mnay]

Model Engineer Magazine did plans for a VW engine.
I will have to get back to my work computer to find the issues.

 

[SailplaneDriver]

Here is a stress analysis from Fusion 360. I ran several cases and the results are similar. The big round things are the main bearing caps. The rod is omitted to allow a section view of the crank.

The results show that the stress is concentrated where the rod and main journals overlap. F360 shows a safety margin of 3.8 so the crank dimensions should not pose a problem. I think I will increase the web thickness to an even .100 and decrease the width of the rod and main journals accordingly.

 

[LSEW]

Good Job! I agree with your assessment of beefing up the journals to .100. Even .010 should help.
One assumption I didn't mention was that you would be making a 1 piece crank, and not building it up. Using a good grade of steel should keep you in the safe range.
maury

 

[johnmcc69]

Interesting analysis...
What kind of power & RPM are you expecting to get from your model?

John