Designing and Building a Custom Crankshaft

[Jerminator96]

So I have a question for all of you engine experts and automotive machinists.

How expensive is it really to have a single, custom crankshaft machined from something like a 4340 steel alloy? And what comes with that? Am I handing the machine shop a CAD drawing that is then coded and produced, or are there engine building specialty shops that can take a few parameters like throw, journal width, diameter, bob weight, etc. and produce the crank themselves?

The reason I ask is that though I can model a crankshaft in Solidworks, I don't know the first thing about crankshaft design.

The engine in question would have 8 cylinders in a 90 degree V with a ~55mm stroke and ~76mm bore. The crankshaft would be single plane and fully counterbalanced if possible. Bob weight is low as the rods and pistons are stock liter bike, most likely Honda though possibly Yamaha. I would want it to survive at 12000rpm.

I know some motorcycle crankshafts are "assembled", and I could see how this would possibly be easier to machine, but is the resulting crank too weak, too heavy, or too unreliable for service at 12k and ~350hp?

I have lots of questions, but my main question is whether or not it makes sense to have a custom crankshaft made, or if I should just save up for a Hartley H1.

[lada ok]

175 bhp / lt is a rather tall order ... your sneaking into super bike territory
can you not couple two engines end for end, or
couple the engines side by side driving the cranks with a chain or idler gear drive

Check out the Hirth crank construction , used by Porsche and others ... this system allows you to use ball and roller bearings, and one piece c / rods

[engineguyBill]

There is a guy in the Tampa, Florida area who specializes in custom crankshafts for motorcycle applications. His name escapes me at the moment, but maybe someone here knows who he is.

[Rizzle]

visit the very recent thread in advanced, Crankshaft design for 10,000+ rpm, theres probably some good info for you in there as well.

[saltfever]

Great question but really too much to type.

A lot depends on the manufacturer and your design. A high volume and popular manufacturer like Crower or Bryant will almost certainly have done all the popular engines and therefore already have the dimensions in their CNC database. However, they will still charge you $150-$300 for one-time non-recurring engineering. The degree that you want to deviate from stock dimensions, or your “customization”, will drive up the price. Lower volume, less well known manufacturers will most likely not have your dimensions in their data base because they probably do not have as diversified list of cranks manufactured in the past and therefore will need a fully dimensioned drawing. Also, and most importantly, they will be the ones to be most sensitive to consumer liability and will want fully dimensioned CAD or hand-drawn drawings simply to cover their butt. They will manufacturer a good part that meets drawing specs and it is a legally binding contract between both of you. However, the responsibility to get the drawing and design right is fully on you. Yes, they know all about single plane, dual plane or anything you can think about cranks. In some cases they could never share it with you because it is proprietary information or Intellectual Property they gleaned from some well healed NASCAR team or some other bazillionaire racer. Design engineering runs about $125-$150/hr. CAD runs $50-$125/hr.(These are fully burdened rates with overhead added in). When you put it all together how much can you afford up front just to get a reasonable design with good drawings even before manufacturing starts. Since you can drive Solidworks your CAD time is free. However, crank knowledge is what you are seeking and you will either have to develop it on your own, pay professionally for it, or a combination of both. If you can be happy with a highly popular model, stock in all dimensions except stroke, you will be way ahead.

No matter who you use, you can assume approximately $3,200 (plus $500, minus $200) for something fairly standard. Figure 20-26 weeks for delivery depending on the time of year (i.e., off season or racing season). The heat treating process (before and after machining) is one of the major contributors to the long lead time. Costs are based on the ubiquitous SBC. A motorcycle may have less material and you would think cost less. But they are as equally demanding in manufacturing and volume is less. YMMV. Good luck.

[Kevin Johnson]

There is a guy in the Tampa, Florida area who specializes in custom crankshafts for motorcycle applications. His name escapes me at the moment, but maybe someone here knows who he is.

http://www.faliconcranks.com/

[mk e]

There is a hyabusa based V8 available if you haven't seen it:

http://www.h1v8.com/page/page/1562069.htm

maybe some ideas there if nothing else.

[Jerminator96]

175 bhp / lt is a rather tall order ... your sneaking into super bike territory
can you not couple two engines end for end, or
couple the engines side by side driving the cranks with a chain or idler gear drive

Check out the Hirth crank construction , used by Porsche and others ... this system allows you to use ball and roller bearings, and one piece c / rods

Well the idea is to build something "Hartley-esque" using the cylinders and heads from a couple of liter bike engines, so hopefully it will be into superbike territory for power.

I want a small package motor, so end to end is not very appealing. I did consider a side-by-side arrangement with a common block to house the two crankshafts. This would ease my crank design troubles, but would likely create more problems than it solves. Not to mention it adds size and weight.

I will check out the Hirth construction method, thanks!

visit the very recent thread in advanced, Crankshaft design for 10,000+ rpm, theres probably some good info for you in there as well.

Thanks! You know I didn't even look in the advanced section, I guess I figured this was not an advanced topic, though I can certainly see how his crank would be!

There is a hyabusa based V8 available if you haven't seen it:

http://www.h1v8.com/page/page/1562069.htm

maybe some ideas there if nothing else.

Yes, that would be my back up if I find it is not feasible to build my own. Though at ~$30,000+ I hope I can do something similar for less.

[mk e]

Well the idea is to build something "Hartley-esque" using the cylinders and heads from a couple of liter bike engines, so hopefully it will be into superbike territory for power.

I want a small package motor, so end to end is not very appealing. I did consider a side-by-side arrangement with a common block to house the two crankshafts. This would ease my crank design troubles, but would likely create more problems than it solves. Not to mention it adds size and weight.

I will check out the Hirth construction method, thanks!

I had a thought years ago that it would be pretty easy to take 2 bike engines and combine them by sawing off the trans sections and bolting them to a common housing that acts as a bellhousing to connect to whatever. These engines are all gear drive clutch so keep that feature wiht both driving a central clutch and you have it. with only a marginal weight increase over a full custom block/crank...I never actually built one of course but I'm pretty sure it's possible.

[Jerminator96]

Great question but really too much to type.

A lot depends on the manufacturer and your design. A high volume and popular manufacturer like Crower or Bryant will almost certainly have done all the popular engines and therefore already have the dimensions in their CNC database. However, they will still charge you $150-$300 for one-time non-recurring engineering. The degree that you want to deviate from stock dimensions, or your “customization”, will drive up the price. Lower volume, less well known manufacturers will most likely not have your dimensions in their data base because they probably do not have as diversified list of cranks manufactured in the past and therefore will need a fully dimensioned drawing. Also, and most importantly, they will be the ones to be most sensitive to consumer liability and will want fully dimensioned CAD or hand-drawn drawings simply to cover their butt. They will manufacturer a good part that meets drawing specs and it is a legally binding contract between both of you. However, the responsibility to get the drawing and design right is fully on you. Yes, they know all about single plane, dual plane or anything you can think about cranks. In some cases they could never share it with you because it is proprietary information or Intellectual Property they gleaned from some well healed NASCAR team or some other bazillionaire racer. Design engineering runs about $125-$150/hr. CAD runs $50-$125/hr.(These are fully burdened rates with overhead added in). When you put it all together how much can you afford up front just to get a reasonable design with good drawings even before manufacturing starts. Since you can drive Solidworks your CAD time is free. However, crank knowledge is what you are seeking and you will either have to develop it on your own, pay professionally for it, or a combination of both. If you can be happy with a highly popular model, stock in all dimensions except stroke, you will be way ahead.

No matter who you use, you can assume approximately $3,200 (plus $500, minus $200) for something fairly standard. Figure 20-26 weeks for delivery depending on the time of year (i.e., off season or racing season). The heat treating process (before and after machining) is one of the major contributors to the long lead time. Costs are based on the ubiquitous SBC. A motorcycle may have less material and you would think cost less. But they are as equally demanding in manufacturing and volume is less. YMMV. Good luck.

That is good information, thank you!

Unfortunately the crank I need is anything but standard. The bore spacing is extremely narrow, and the closest automotive application I know of is the Audi V8 with 90mm bore spacing. It really is a short motor (around 19" front to back), but I would guess the spacing is still more than can be compensated for.

If I'm not mistaken, crower makes the Hartley crank. That would naturally be closer with the Hayabusa 88mm bore spacing, but there are other considerations in his design that would be significantly different. Maybe I could get permission from him to use his crank as a basis for my design, since I have no plans of starting a business. That's probably a long shot though.

I guess in the end the number I had in my head was ~$5000, but without designing it myself I'm guessing that's a stretch. The good news is that I'll be spending the next few years in Guatemala, so I have plenty of time to read up on crankshaft design.

[saltfever]

When the Ford mod motor came out I needed a special crank. I couldn’t get very much information or drawings of any kind partly due to its newness and partly because I was a nobody. Unless you are “connected” some doors are hard to open. Finally, I just sent the existing crank to Crower. I think they had a Coordinate Measuring Machine (CMM) and just pulled all the requisite dimensions of the crank I supplied. There were still drawing charges and set-up charges before they even chucked up the bar stock. Nowadays everybody has embraced the CNC world so you will be money ahead if you commit right from the start to providing a digital dwg. Everybody will accept a Solidworks or DXF file.

You may want to consider another approach. All of the quality companies have a big overhead burden to keep the doors open. This cost has to be passed along to the customer so the more you can provide them the better it is for you. And . . . it is also less risky (and more desirable) for them to see a customer that knows what he wants and the “what if” phone calls will be kept to a minimum. I’m leading up to something here. You might call and ask them if they will accept a partially turned or “roughed out” billet. All you want them to do is finish grind and heat treat. If so, put your American dollars to work for you in Guatemala. There should be some high quality machine shops who would love to whittle out a piece of 4130 for you down there.

[Sir Yun]

have a look here as well.

http://www.rpeonline.co.uk/categories/RPE-RP-V8/

[Jerminator96]

When the Ford mod motor came out I needed a special crank. I couldn’t get very much information or drawings of any kind partly due to its newness and partly because I was a nobody. Unless you are “connected” some doors are hard to open. Finally, I just sent the existing crank to Crower. I think they had a Coordinate Measuring Machine (CMM) and just pulled all the requisite dimensions of the crank I supplied. There were still drawing charges and set-up charges before they even chucked up the bar stock. Nowadays everybody has embraced the CNC world so you will be money ahead if you commit right from the start to providing a digital dwg. Everybody will accept a Solidworks or DXF file.

You may want to consider another approach. All of the quality companies have a big overhead burden to keep the doors open. This cost has to be passed along to the customer so the more you can provide them the better it is for you. And . . . it is also less risky (and more desirable) for them to see a customer that knows what he wants and the “what if” phone calls will be kept to a minimum. I’m leading up to something here. You might call and ask them if they will accept a partially turned or “roughed out” billet. All you want them to do is finish grind and heat treat. If so, put your American dollars to work for you in Guatemala. There should be some high quality machine shops who would love to whittle out a piece of 4130 for you down there.

Not a bad point. The USD is pretty strong right now. I'll have to do some research to see what is down there. I know a few of the OEMs are manufacturing engines is Mexico. There is probably someone there if I want to risk my life to visit.

[Jerminator96]

have a look here as well.

http://www.rpeonline.co.uk/categories/RPE-RP-V8/

Yeah, pretty much the same deal as the Hartley, though I have always had good dealings with the Brits back when the Rover V8 was my engine of choice, so maybe I'd have good luck negotiating a crank design out of them?

[SchmidtMotorWorks]

Start by figuring out what bearings you will run and the thickness of the rod throws that result.
That parts sounds easy until you try to do it.
Making a cranks is relatively easy once you get that figured out.