Multi-Cylinder Pressed Crankshafts

I haven't been able to find a lot of information on the internet about pressed crankshafts. I know they're fairly common in the 2-stroke world, seem to mostly be used in single throw engines, but have definitely also been used in multi throw engines and 4-strokes as well. Looking for 4-stroke, 4 cylinder versions seem few and far between. Off hand the only I'm familiar with is the Suzuki GS550.

In theory, this seems like a potentially more affordable method for producing custom crankshafts for specialty applications. It could possibly even be DIY. So I'm looking to pick up any tidbits I can from people here who might have some knowledge or background in working with these types of cranks. What would be the major shortcomings or pitfalls relating to this type of crankshaft on 4+ cylinder engines? Obviously alignment when assembling the crankshaft would be critical. And I know about the propensity for them to twist during operation if they don't have some means (often welding) to lock them in place. But what else? Would this really be cost prohibitive compared to a billet crank? Would the performance or longevity really be that inferior to a billet crank? Or is it more a matter of perception and a lack of resources/knowledge?

I think that rigidity is going to be a Big Deal, and I suspect that a pressed together crankshaft will always be badly inferior.

But that's my opinion, and a pretty poorly-informed one at that. Mostly based on the general impression that a crank can never be too rigid, and that a multi-part assembly is nearly impossible to get really unified...

The first person I think to turn to is Kevin Cameron, who's worked with pressed cranks, and has written extensively on every aspect of first-principles engine issues. I'm not sure where to find the core nugget of wisdom, though, I'm afraid... Might be worth adding his name into the googling... If I get a chance to re-skim TDC and TDC2 and find something pertinent, I'll try to toss up a link or a quip...

I suspect that anyone with the skills and equipment to build a DIY pressed crank will also be equally able to make a fully machined 1 piece crank. The reliability of a 1 piece crank will far out weigh the increase in cost and time to build it.

Just my $.02. I've never built, or even needed a custom crank so my point of view is limited to say the least.

Edit: I forgot to add that I used to be a motorcycle mechanic, so I do a few thing about pressed cranks.

. mechanic, so I do know a

The Auto Union V16 used Hirth Couplings.

I think torsion would be your worst enemy on anything over 2-3 cylinders. Then bearing support, especially 4-strokes w/ babbit brgs., 2-stroke roller brgs. are another story. Then balancing.

Inline, horizontal or V configuration?

I think the main justification for pressed together crankshafts is to enable roller bearings in place of journal bearings.
Otherwise, why bother?

HappyAndy wrote: I suspect that anyone with the skills and equipment to build a DIY pressed crank will also be equally able to make a fully machined 1 piece crank.

That's one thing I'm not really sure would hold true. Consider that shops with the machines required to turn and grind on a single centered axis only, as is possible with a pressed design, are far more readily available and at far more reasonable rates than for machines which can do billet crankshafts.

erohslc wrote: Otherwise, why bother?

In addition to cheaper machining rates, the raw material could be considerably cheaper, and with the ability to have substantially less material to remove it would be in those cheaper to run machines for less time too. Also economies of scale when multiple identical pieces are used in an assembly vs a quantity of one.

fasted58 wrote: I think torsion would be your worst enemy on anything over 2-3 cylinders. Then bearing support, especially 4-strokes w/ babbit brgs., 2-stroke roller brgs. are another story. Then balancing. Inline, horizontal or V configuration?

I agree on torsion being an issue as cylinder count rises. I guess I'm not seeing the concern with bearing support when using the same journal bearings as any other engine? Also what would the pressed crankshaft be any more difficult to balance than a billet crankshaft?

Configuration seems like it could apply pretty equivalently between inline and v configurations.

JamesMcD wrote: The Auto Union V16 used Hirth Couplings.

Thanks, I was not aware of that. It instigates some interesting thoughts.

Additive manufacturing will eventually render most of these arguments moot.

erohslc wrote: Additive manufacturing will eventually render most of these arguments moot.

Eventually yes...But how long until that is a common and affordable method of manufacturing for steel? 10 years? 20 years? Longer? Until then the arguments will all stand.

erohslc wrote: Additive manufacturing will eventually render most of these arguments moot.

I'm waiting for a means of additive manufacturing that can make a crank shaft, and is affordable enough to put into my shop. But that kind of tech has ramifications that, if we got into them, would surely flounder this thread.

Additive manufacturing? Now I'm really confused, Does that mean 3D printing?!

Yes, additive manufacturing is geek speak for 3D printing.

erohslc wrote: I think the main justification for pressed together crankshafts is to enable roller bearings in place of journal bearings. Otherwise, why bother?

Right, the roller bearings were what caused the need for the pressed crankshaft.

Porsche found that roller brngs weren't necc. a long time ago.

Driven5 wrote:

erohslc wrote: Additive manufacturing will eventually render most of these arguments moot.

Eventually yes...But how long until that is a common and affordable method of manufacturing for steel? 10 years? 20 years? Longer? Until then the arguments will all stand.

How long until someone ponies up the time and money for conventional tooling to make new pressed multi-cylinder crankshafts?
(crickets)
The case for AM is compelling.
The technology for creating AM steel cranks ALREADY EXISTS, the sole issue for one offs is price.
F1 is already using it for some parts.
Time to get rid of those shares in buggy whip manufacturing companies.

JamesMcD wrote: I'm waiting for a means of additive manufacturing that can make a crank shaft,

You could probably do it today with a computer controlled MIG welder, if you wanted to prove a point.

If you want a STRONG steel part, that's basically what it's going to be. For something large like a crank, the energy outlay would most likely mean that it's still cheaper to start with a forging and machine that.

In reply to erohslc:

The "problem" with AM cranks is that they are cast. Forged cranks are quite a bit stronger than cast ones. And being cast, one still needs to machine it when you are done.

AM is a great way of making one part- which is why F1 uses it for body parts. It's not a great way of making forged metal parts.

Well, until someone comes up with a way to deposit metal material that is useful for cranks. I can't imagine how much that will cost....

It's going to be a while until AM can make things that are other than plastic and hand made carbon fiber (which is what the f1 parts are- the AM makes the molds, which is then used to lay the CF, cure it, and remove the AM stuff).

lets not too far ahead of ourselves, here.

Motorcycle cranks are, as noted, generally 'built up' or pressed together. They do have a tendency to twist under really hard use with big compression etc., for instance if building a rephased big bore XS650 the companies will disassemble, rephase, press it together then weld the pins in place. There's aftermarket companies (Falicon comes to mind, http://www.faliconcranks.com/crank_svc.html ) that specialize in fixing the multi cylinder cranks, generally they assemble, true, then weld it together.

erohslc wrote: How long until someone ponies up the time and money for conventional tooling to make new pressed multi-cylinder crankshafts? (crickets)

Exactly what conventional tooling to machine the components for new pressed multi-cylinder crankshafts doesn't already exist in countless machine shops in every major city in the country? The only thing missing right now seems to be a design. As far as I can tell, a pressed crankshaft could be broken down into pieces that are all able to be machined from near-net sized bar stock using standard simple turning, grinding, milling, and pressing operations. No special machines or tooling required at all. The most challenging fixtures would be those used during assembly for aligning and straightening the parts as they are pressed together.

Oddly enough, while I seem to be defending the idea of a pressed crankshaft, what I am really wanting is a logical and convincing argument as to why this would not work as described.

iceracer wrote:

erohslc wrote: I think the main justification for pressed together crankshafts is to enable roller bearings in place of journal bearings. Otherwise, why bother?

Right, the roller bearings were what caused the need for the pressed crankshaft. Porsche found that roller brngs weren't necc. a long time ago.

They still are on 2 strokes, because you don't have the same lubrication setup that is possible on a 4 stroke engine.

alfadriver wrote: Well, until someone comes up with a way to deposit metal material that is useful for cranks. I can't imagine how much that will cost....

It exists already - spray welding, or other forms of automated welding.

Plastic printing is just swirting molted globs of plastic in a specific pattern, nothing more fancy than that.

Driven5 wrote:

erohslc wrote: How long until someone ponies up the time and money for conventional tooling to make new pressed multi-cylinder crankshafts? (crickets)

Exactly what conventional tooling to machine the components for new pressed multi-cylinder crankshafts doesn't already exist in countless machine shops in every major city in the country? The only thing missing right now seems to be a design. As far as I can tell, a pressed crankshaft could be broken down into pieces that are all able to be machined from near-net sized bar stock using standard simple turning, grinding, milling, and pressing operations. No special machines or tooling required at all. The most challenging fixtures would be those used during assembly for aligning and straightening the parts as they are pressed together. Oddly enough, while I seem to be defending the idea of a pressed crankshaft, what I am really wanting is a logical and convincing argument as to why this would not work as described.

How much experience do you have working with machine tools? Not being a dick, just trying to assess what kind of experience you have to support your assertions.

Before the crankshaft parts can even be designed, the manufacturing methods and processes have to be determined. This includes the fastening methods, the tools, jigs, partholders, etc. as well as the allowable tolerances and finishes.

As part of the design and prototype process, those non crankshaft items have to be built, tested, validated.

Then the crankshaft parts themselves can be designed, then manufactured, perhaps with a few prototype iterations to work out issues.

Do you envision using Hirth couplings, or press fits, or keyways, or splines, or welding (conventional, spin, laser, diffusion bonding, brazing, etc.).
Maybe cryogenic methods?
Will the parts all be machined to spec, or hand fitted on final assembly?

Are you aware of the detailed stresses imposed on any crankshaft in normal use?
Are you aware of the special considerations required for a pressed crankshft to withstand those stresses and preform reliably?

Do you know what costs and capitol investments are involved in manufacturing standard 1 piece cranks from castings or forgings, vs same for pressed crankshafts?

Knurled wrote:

alfadriver wrote: Well, until someone comes up with a way to deposit metal material that is useful for cranks. I can't imagine how much that will cost....

It exists already - spray welding, or other forms of automated welding. Plastic printing is just swirting molted globs of plastic in a specific pattern, nothing more fancy than that.

And people are "printing" parts this way? Doesn't seem all that practical- compared to machining a part out of a billet, even.

alfadriver wrote: In reply to erohslc: The "problem" with AM cranks is that they are cast. Forged cranks are quite a bit stronger than cast ones. And being cast, one still needs to machine it when you are done. AM is a great way of making one part- which is why F1 uses it for body parts. It's not a great way of making forged metal parts. Well, until someone comes up with a way to deposit metal material that is useful for cranks. I can't imagine how much that will cost.... It's going to be a while until AM can make things that are other than plastic and hand made carbon fiber (which is what the f1 parts are- the AM makes the molds, which is then used to lay the CF, cure it, and remove the AM stuff). lets not too far ahead of ourselves, here.

Actually metal AM is already production reality.
(But bring the BIG wallet) Several technologies are available, here are a couple of links.

http://www.optomec.com/Additive-Manufacturing-Technology/Laser-Additive-Manufacturing http://www.renishaw.com/en/additive-manufacturing--15239

None of these are castings or forgings.
Most methods use a metal powder, that is melted or sintered by laser.

But there are some brute force machines that essentially are computer controlled wirefeed welders, creating one continuous bead of precisely deposited metal as a finished part.

An exciting capability is to use several different powders, to create different alloys having different characteristics as required in different portions of the same piece.
The laser also has the potential to create different heat treatments as needed.

Imagine a solid part, that has high strength structural webs or bands embedded, perhaps even a mixture of different metals for specific purposes.

Imagine a metal connecting rod that mimics bone tissue, containing thousands of hollow elliptical cells, individually designed and optimimally oriented in the direction of stress, enclosed in a solid outer layer of material, whose thickness can be precisely controlled for the stresses in that individual location.

Forging is a bulk method that allows us to introduce anisotropic characteristics to an otherwise homogenous metal part ('grain flow'). But how much control of that characteristic do we have with forging?

AM will allow us to precisely tailor the metal characteristics as the part is built, with sub mil accuracy.

AM will not replace conventional machining. Anyone who has watched an automatic screw machine turning out thousands of parts per hour, or a five axis NC machining center creating complex parts from solid blanks, and then compares the AM part build rates understands that AM is not quick.

AM allows us to do things that have never been possible or practical before. But not quickly.

Carter

erohslc wrote: How much experience do you have working with machine tools? Not being a dick, just trying to assess what kind of experience you have to support your assertions. Before the crankshaft parts can even be designed, the manufacturing methods and processes have to be determined. This includes the fastening methods, the tools, jigs, partholders, etc. as well as the allowable tolerances and finishes. As part of the design and prototype process, those non crankshaft items have to be built, tested, validated. Then the crankshaft parts themselves can be designed, then manufactured, perhaps with a few prototype iterations to work out issues. Do you envision using Hirth couplings, or press fits, or keyways, or splines, or welding (conventional, spin, laser, diffusion bonding, brazing, etc.). Maybe cryogenic methods? Will the parts all be machined to spec, or hand fitted on final assembly? Are you aware of the detailed stresses imposed on any crankshaft in normal use? Are you aware of the special considerations required for a pressed crankshft to withstand those stresses and preform reliably? Do you know what costs and capitol investments are involved in manufacturing standard 1 piece cranks from castings or forgings, vs same for pressed crankshafts?

Am I a machinist? No. Am I a mechanical engineer that has primarily worked in machine and assembly shop environments with duties including but not limited to design, quoting, planning, quality control, and manufacturing support? Yes. That's exactly what led me into this thought experiment in the first place. On the other hand, I have never designed a crankshaft in any of my previous lives. Nor have I assembled/used a pressed crankshaft. Otherwise, I would already have the answers to all of the questions above and would have had absolutely no reason for starting this thread in the first place. What I am looking for is people who can provide insights from their own relevant experiences with crankshafts into the critical questions (stresses, special considerations, cost, performance, durability, etc) and help me see what I'm missing.

That being said, the recurring thought revolves around the idea of a modular design. There are many possible ideas that have come and gone. One of the more common ones has revolved around turned and ground cylindrical pins with an additional oil passage operation for both the connecting rod pins and main pins. A shop with a live tool lathe might be able to do this a little more efficiently, but are generally in shops with higher overhead rates. The idea would is that that any mom-n-pop small scale shop with a solid lathe and 3-axis mill could crank these out. Meanwhile the counterweights which would be used to connect the bearing pins could be waterjet to near net shape, to require minimal milling to bring everything to size and within tolerance. The pins could be pressed into the counterweight plates, with some secondary method of securing alignment. Maybe something like multiple match drilled and pressed steel dowel pins, or simply welding as has been done in other applications. None of these operations should necessarily require a significant investment in fixturing. The hirth couplings are interesting for their capabilities, but would generally complicate the manufacturing of the individual pieces to at least some degree.

Let's say hypothetically that we wanted a crankshaft that would be used to build a flat-plane 3.0L V8 out of a 1UZFE that was mildly tuned enough to easily run pump gas. Maybe just one for me, or maybe one for me and a small handful of other similarly dim witted individuals. But I wouldn't figure on any more than that. My somewhat limited experience with castings and forgings is that they tend to be prohibitively expensive unless you're going into moderate to large scale production and have a quite substantial initial funding available. Cast iron obviously being the more affordable of the two, but is it actually any/much stronger than a pressed crankshaft made from high strength steel components? Billet cranks are also extremely expensive, but are certainly going to be significantly cheaper than a steel closed die forging and is at least quite a bit stronger than cast iron. Off hand I don't know if a handful of billet steel cranks would be more, less, or similarly expensive than the same number of cast iron cranks either. But by comparison, the tooling and jigs I've seen used for rebuilding and aligning pressed crankshafts seems rather straight forward and appears it should be far cheaper to build, test, and validate. But again, if I had more first hand experience with this I would have been able to complete this mental exercise on my own. So please tell me what I'm not seeing.