Topic: Designing a new motor - wood models or CAD?

[lordmoonpie]

With Mike's billet blocks going down a storm, clearly there's folks out there with the ability to design items like this and turn them into reality, so there's some good experience to draw on I think.

I'm planning an engine build....from scratch. Although it's not an SOHC4 I figure this forum is where the right experience lives so am posting here rather than "other bikes" where it might get missed by all you experts like Mike and Mec (oh yeah and Sam ;-)). Hey what a great idea this forum is eh? 

My plan is a 60 degree vee twin, water cooled. I was planning to build a wood model and get it cast which gives me the question of how accurate the model needs to be, whether I create the cutouts for the crank bearings to sit in etc or leave them blank for later line boring or machining etc. Then I got to further thinking....rather than wooden models maybe I should invest in a home PC CAD package and try and create it that way. Most casting companies seem to work from CAD and it might also allow me to consider billet machining instead which I'd actually prefer.

So, in the experiences of you experts who have built things from scratch, what's the best thing for me to invest my time in - wooden modelling (and to what level of accuracy) or go down the CAD route (and if so can you recommend an easy package capable of modelling a V-twin?)

All help gratefully received - I'm so serious about this I'm selling the Factory Stage 3 Moriwaki VTR to fund it (Superhawk to you guys over the pond) and the only other one left of those is apparently in the Moriwaki museum!

[mlinder]

Well, CAD would be the way to start, in my opinion.
It is, after all, exact, and easily manipulated for changes.

What size were you going for? I'd try to design it to use some stuff that's already available, like, say, con rods from a gs750-800-1000, etc.

[Sam Green Racing]

Simon, when I started talking to Ron Philips (Fahron) about building a 175 RS to attack the British 175 records, I asked him what it would cost to cast a full size block (175cc) to fit the RS125 bottom end as oposed to a big bore (134cc) that I ended up using, Ron has cast many blocks over the years.
He said casting the block would be peanuts but making the plug/mold would cost thousands, go figure.

Sam.

[754]

Yes, foundries that charge out their pattern makers, charge a mint..
 but..
 if you luck out you find a retired one with lots of tools, much cheaper..
 but the good thing, as Burt showed us, it is POSSIBLE to do it yourself.. can be done with patience and not that many expensive tools.. + remember even if it eliminates a lot of the machining ( waste & time) you are ahead ..Thinking rough pattern.

 But pattern cost is usually amortized by running 100s of pieces or larger quantities.

 Barrel for a twin is easy to machine out of a lump of billet.

 Also if you got a modeling machine to carve a couple patterns out of foam you could use that for a pattern.. one time use.. it gets melted..

[mec]

i would go the CAD/CNC route. this is the future (or present).
why a 60° twin?
a 90° twin can save counterweightbalance shafts.

mec

[wannabridin]

i would work with Solidworks personally.  much more powerful than AutoCAD.  also, you have access to COSMOS, which is a finite element analysis tool that can help you visualize weak spots.  if you get your rough idea drawn out in Solidworks, then you can go to a wood mock up to check for clearances and accessories.  but the part about making a mold is correct.  for something like a cylinder block, getting the coolant passageways done for anything less than straight through is difficult with a CNC.  lost wax casting or investment casting is the way these are done, i did research in my undergrad's experiment foundry for awhile, and i have a friend who works for Toyota's cylinder block and cylinder head foundry and i've taken a cool tour there.  either way, if you make the motor simple enough with straight through water passageways, then you won't have to worry about foundry.  and you can control the quality a lot more with CNC, because you'll start off with a block, and you won't have to worry about dialing in the different cooling expansions and overfills and runners associated with founding...

hope this helps!

[simon#42]

With Mike's billet blocks going down a storm, clearly there's folks out there with the ability to design items like this and turn them into reality, so there's some good experience to draw on I think.

I'm planning an engine build....from scratch. Although it's not an SOHC4 I figure this forum is where the right experience lives so am posting here rather than "other bikes" where it might get missed by all you experts like Mike and Mec (oh yeah and Sam ;-)). Hey what a great idea this forum is eh? 

My plan is a 60 degree vee twin, water cooled. I was planning to build a wood model and get it cast which gives me the question of how accurate the model needs to be, whether I create the cutouts for the crank bearings to sit in etc or leave them blank for later line boring or machining etc. Then I got to further thinking....rather than wooden models maybe I should invest in a home PC CAD package and try and create it that way. Most casting companies seem to work from CAD and it might also allow me to consider billet machining instead which I'd actually prefer.

So, in the experiences of you experts who have built things from scratch, what's the best thing for me to invest my time in - wooden modelling (and to what level of accuracy) or go down the CAD route (and if so can you recommend an easy package capable of modelling a V-twin?)

All help gratefully received - I'm so serious about this I'm selling the Factory Stage 3 Moriwaki VTR to fund it (Superhawk to you guys over the pond) and the only other one left of those is apparently in the Moriwaki museum!

you have found a new project then!

[mlinder]

I use Rhino for my farting around with 3d crap. Dunno if it has a COSMOS like plug in, haven't checked.

[lordmoonpie]

Good advice - thanks guys. The reason for a 60 degree twin (1000-1200cc I'm aiming for) is to keep the head angle narrow enough to be able to put large cam covers on to be able to then hang the seat and the front end directly off the engine. (Can you see where I'm going with this :-))

The cases will be horizontally split and use a GS1000 gearbox as the ratios are nice and the selection mechanism compact. The top case will have the barrels cast in one piece. I'd like to use a Ducati crank but not sure that a 90 degree twin crank will go straight in a 60 degree V engnine....but I can't see why not?

I'm going in for some major surgery on my spine in January so will be laid up for about three months. My plan is to spend my time doing as much technical drawing as I can in this time but I might just as well do all of that on a CAD package so will look into the cost of Solidworks (Thanks wannabridin!)

It may all end in nothing but sometimes you just got to have a go haven't you!?

Simon#42 - yup looks like my old plans are resurfacing, still got my XV750 and ducati cucciolo to finish though, you can never have too many projects IMHO

[Sam Green Racing]

Sounds like much much much to much time on your hands

Sam.

[wannabridin]

no problem!  i know i got my student copy a year ago for like $140 w/ a 2 year license and COSMOS.  i used to be somewhat proficient with it, but that was 4+ years ago...  i found a solidworks for dummies book at half price books that has helped a bit, but practice is what makes things there perfect...

solidworks: http://www.solidworks.com/sw/products/mechanical-engineering-cad-software.htm  it's not called cosmos anymore, it's "Design Validation" but it serves the same principle (FEA)

you can also look for a copy of ProE (pro engineer) as it does the same thing: http://www.ptc.com/products/proengineer/

here's some cool videos just so you can appreciate the software and assembly characteristics:

[simon#42]

the 90 degree crank will have the wrong balance factor simon , you could possibly get it altered but it might be worth looking at the aprilia v twin crank i think that is a 60
dont underestimate the cost of all this , i had some two stroke cylinders cast a couple of years ago and by the time i had the patterns made the casting done and the plating i was the wrong side of 30 grand ........ beware!!

[754]

Some considerations;
 A  2 stroke jug  versus a parallel twin air cooled jugs

 Roughspeak (not an expert).. on a scale of 10, 10 being most difficult.. 2 stroke is a 5 or 6 the twin jugs a 2


 I was involved in machining some  IRS parts to restore some German  low production roadrace cars from the 70,s.. mame starts with a J.. 3 letters..

 Pattern making cost in Germany 10 K.. 2 pces.. biggest size of a football .. cost of getting a retired patternmaker here to do it 3 or 3.5 K... IIRC, it may have even been less.

 Seriously a set of twin jugs, you could glue layers of plywood (one for each fin) together on a weekend, with a drillpress , jigsaw, some glue and a 4.5 angle grinder.. .. It dont have to ne perfect, it dont have to be nice, just got to eliminate a lot of machining.. its a prototype..
 

[lordmoonpie]

Thanks for all the links wannbridin - I looked for solidworks in the UK last night and they are all - "ring us for a quote" sites so I'm guessing very expensive but your $140 sounds good 
I did see the dummies book too - looks like a good place for me to start LOL!

Simon - yes I was thinking about the Aprilia twin crank, it is a 60 degree as you say. The other route would be to talk to someone like SEP and get a crank made but again, big money I suppose. As for casting, I guess we're quite lucky here in the UK as there are loads of tiny casting firms in places like Birmingham who are happy to do one off casts etc. If a bloke I know in Chester can cast replica CR750 hubs in batches of 6 and make money on them then there's got to be some hope for me if the model is a good one. I suspect the biggest cost will be in the post-cast machining for gearbox and crank bearings to sit square and in line. Another cost I haven't got my head round yet is that I want this to be belt driven cams rather than chain plus fuel injection. I've given myself a target of ten years to complete which is probably not that silly, if it ever completes...

754 - thanks for the input, I want to water cool this motor if possible but it will be a four stroke. I guess I could go air cooled route first and later enhance to water cooled but given that it's more setup, I thought I'd go straight for the water cooled and try and cut the costs a bit.

Thaks for all the views guys - one of the reasons this forum knocks the socks off others, there is sooo much knowledge and time served experience here it's a fantastic place to hang out! 

[754]

BUB Mfg is pretty far along on their V water cooled design, I think  they are into it for a million bucks...

 Mostly based on their LSR motor, they are trying to put a street version out there..

 I was probably thinking of sams LSR 175, regarding the twin jugs.

[von_Wanderlust]

Thaks for all the views guys - one of the reasons this forum knocks the socks off others, there is sooo much knowledge and time served experience here it's a fantastic place to hang out! 

And here we all thought you were dissing the hi-po forum a few weeks ago! 

[Sam Green Racing]

Thaks for all the views guys - one of the reasons this forum knocks the socks off others, there is sooo much knowledge and time served experience here it's a fantastic place to hang out! 

And here we all thought you were dissing the hi-po forum a few weeks ago! 

I agree there Syd, this one should have it's own forum, The Ultimate Motor Forum.
If you look at Simons leathers in his avatar, perhaps you'll get a clue about which road he is going down (a guy from the other island)

Sam.

[tinyrobot]

I learned a very valuable phrase when I went to school for Industrial Design: "Mock it up, before you Fock it up!"

I say create your engine in 3D software ei. Solidworks, Rhino, Pro E.  Make your own one to one out of foam, wood, cobble together real engine parts is need be.  Then find a place to rapid prototype.  This way you can have a one to one to mock up in frame.  Once you start having the molds made for the parts you are a little too far gone to turn back.

Good luck!

[Sam Green Racing]

Simon has been far gone for a long time, He's no chance of turning back good job we are mates

Sam.

[lordmoonpie]

Simon has been far gone for a long time, He's no chance of turning back good job we are mates

Sam.

Yeah good job we WERE mates 

Sam's right though - my avatar is where I'm heading. One day, just one day...and as for the hipo forum of course it's a good idea, it was my idea wasn't it? LOL!

Tinyrobot - great mantra for the project, I love it thank you!

So far solidworks seems the only option. Are there any other cgood cad packages that won't break the bank I can look at? What about autocad?

[mlinder]

Try rhino3d.

Tryout version at http://www.rhino3d.com/download.htm

[Ricky_Racer]

Yeah, Rhino! NURBS rules!     RR

[mlinder]

Yeah, rhino is fun. Just started using it again to design my industry-shattering race bike I'm out of practice with it, but used to use it for fun. Had some free time 

[Ricky_Racer]

Dude, you're good!      RR

[mlinder]

Thanks.

I tend to get a little carried away with modeling details. I have some stupid aversion to using texture maps :/