What's new
What's new

Scale Merlin Crankshaft

dynamotive

Plastic
Joined
Jan 20, 2007
Location
San Diego
I joined this forum to ask questions about machining a crankshaft for my quarter-scale Merlin engine. You can see the engine at www.quarterscalemerlin.com. I first started the topics 'fast stock removal from steel', and 'forging' to simply ask questions about processes, and there's some good answers in those. Here's a picture of an investment-cast 4130 steel crankshaft blank, one of four I had done recently.
castcrank.jpg

It's 10.3" long. It looks nice, but it's unusable because there are voids in the casting, and it's just a bit too long. This batch was actually the third time I've tried this. In previous attempts, the casting places picked their favorite exotic alloy, and the results were not machinable or straightenable by me, plus they were incredibly warped. I added the extra bridges between the crank webs to reduce warpage, and it worked pretty well. But, of course, if the previous ones had been straightenable, it wouldn't have mattered. All part of the learning and flogging around process.

The casting place warned me that this shape might not cast well, and they were right. Years ago, I machined a crank for the prototype out of a 2.5" diameter 4140 rod, and had it ground at a precision grinding place. It came out fine, but ever since, I have been most anxious to find a way to simplify the process of producing more crankshafts. That's why the effort and expense on the cast steel crankshafts. Many people will undoubtedly suggest exotic ways of doing this, so let me state right up front that I do not want to invest large amounts of money in any part of this project. No large quantities, and no expensive processes or tooling. Ok?

Somebody suggested a cast iron crankshaft. That is not something that I have considered before, because I think of cast iron as being brittle and weak. I know that many cars use cast iron crankshafts, but racing cars rarely do, and the real Merlin certainly didn't. However, this model engine actually is not that highly stressed, at least the way I envision it. The real Merlin turned a maximum of about 3200 rpm, and in order to make the scale version as realistic-sounding as possible, it should be kept in that range. According to some supercharger experts I consulted, this engine should be capable of producing about 16 hp at 3600 rpm, which should be enough to power a 1/4-scale Mustang or Spitfire at scale speeds. An engine this size running at 3600 rom should be relatively unstressed, so maybe a cast iron crankshaft would work fine, but I still feel very reluctant about going that way. What are the chances that I could get a cast iron crankshaft blank that is reasonably sure to not have any dangerous defects in it?

Incidentally, I just got a book called 'Vees for Victory' which describes the Allison V-1710 in great detail. In the early versions of that engine, the crankshaft did not have any counterweights. It was designed to run at 2400 rpm, and at that speed, the bearing loads due to imbalance were acceptable. Weight is obviously very crucial for any aircraft engine, and leaving the counterweights off saves weight. When the design speed was increased, counterweights were added, but only to reduce the bearing loads to acceptable levels, never to try to achieve complete balance. Again, it's that weight thing that's hanging over them. The Merlin's counterweights do not aim for complete balance either.

Getting back to the investment-cast steel crank, I think that if I give up trying to cast a near-finish shape, and make it look more like what a forging would like, in other words very rounded, and only an approximation of the finish shape, and with no counterweights, that it would cast fine. The problem is finding places that are willing to try it for small quantities. I think the place that did the last ones doesn't want to hear from me any more.

The other option brought up was machining a flat blank and twisting it to get the correct throw orientation. The mains are pretty narrow - would that work? Would that be something I could do with a torch and some sort of jig?
 
Millions of lawnmowers soldier on year after year with their nodular (ductile) iron cranks at similar speeds and probably longer strokes.

Rear axle gears in even "heavy duty" light trucks are very commonly nodular iron.

This is all very jolly, because the limiting factor will be finding someone willing to make a few nodular iron castings and have them come out right. I do not expect you will have any greater luck with nodular iron than you have had at the investment casting places.

I'd get the steel material you want and wire edm off everything possible. I think for this engine counterweights are just for looks.

Many inline aircraft engines did not have counter weights. This would certainly include the 440 Ranger inverted six that was run continuously at around 2000 RPM with a stroke of 5 1/2". A known fact is that as surplus, these engines were run considerably faster in various racing cars. They probably broke cranks then.


John
 
dynamotive, I can't see an image in either of your postings.

The method I described in other thread which was used for milling a Merlin crank - sounds like it could be done by a CNC mill with indexing head. I am not suggesting you buy this equipment, but you could try getting quotes ie sounds like you are trying to make a run of parts, not just doing it for fun?

On the same lines - guys like John Stevenson have made their own CNC indexing heads, maybe one of these on a manual turret mill would help?

BTW nodular iron used on many automotive cranks, Chev V8 etc. I know a guy who has full-size steam engine crankshafts cast from nodular iron, but not sure about your application.
 
You might give the folks at Standard Alloys in Port Arthur Texas a call...
I've dealt with them in the past on replacement parts for obsolete pump parts for oil refinery applications. I also got a first class tour of their entire plant, it was quite impressive...

They specialize in hard to get, hard to duplicate, one off jobs...in just about any material you can name & As I recall their prices weren't bad at all...

Might be worth a phone call...
Here is a link to their site...


http://www.standardalloys.com/


I'd back away from cold forming or even heating & bending any part that has to stand up to cyclic stress... Potential for cracking and intergranular stress is pretty high...

As far as Cast Iron cranks, have you thought about having them tested at a Metal Testing lab (Most are in inexpensive enough that they are used to certify welders) (Cost per piece might be a couple hundred or so) before any expensive machine work is done... This can be done using Imersion Ultrasonics or Radiography (X-Ray)
Magnetic Particle testing for finished parts...
These may add a little cost to your cranks, but you would be able to control the quality up front...& if you have a small run it would be well worth the cost spread out over several parts...


Hope this helps
Pat

Oh & BTW the Merlin is excellent...
 
Sorry about that, for some reason it didn't work the first time I tried to put the cast crankshaft blank image in.

castcrank.jpg


Yes, it uses fork-and-blade rods.
 
They are still not visible. You need to upload the pictures to a host first. Then paste the URL from that site.

Packard alloyed their own steel for the crankshafts. I'd be interrested in seeing the rods too!

Glenn
 
It looks to me that no matter what you do, you are going to have to do finish machining and grinding.

A CNC mill with 4th axis would probably let you do the machining with .004" or so left oversize for the grinding shop to take care of. Once you get the journals close to size the thing would seem like it could get a bit flexible. I don't know if many crank-grinding shops would want to try and deal with something so much smaller than what they normally work on, but who else is going to have a grinder set up for doing the eccentric pin journals? You might get around that by having some complicated fixture that allowed all the rod journals to be set up so that they are a fairly straight-forward cylindrical grinding operation done between centers.

Yes, this is going to be a low-power engine (I'd think that a "sporting" V12 of 400-500cc would be in the 80-100hp range) but it still has to be accurately made if it is going to live for any length of time.

Maybe you need to make friends with someone who has one of those Mazak Integrex machining centers. They could probably spit out something that size every 5-10 minutes.

cheers,
Michael
 
How did Dick Yeagley make the crank for his scale Merlin that was exhibited at Cabin Fever a few years ago? He seemed like a very personable guy. I'm sure he'd share his secrets. I recall that many of the operations to make the heads involded EDM.
 
Looks like rod journals 2 and 5 have the main journal diameter running right across them. I think making them all like that would help with material flow rather than using the smaller diameter bridge on the opposite side. Start with a solid full length shaft of main journal diameter and 'grow' the rod journals and counterweights from it.

You want a clear path all the way down the center of this so you can pour and fill from the bottom up. If 1 and 6 look like 3 and 4, the metal is turning a couple of corners before it can even get to the 2nd main journal.

Should also bridge the counterweights from almost the outer rim all the way to the journal. Basically minimize the need for material to turn a corner as it fills the cavity. Eliminate the cavities in the ends of the journals, as well as the holes in the end flanges.

Might be a lot harder to find, but vacuum casting would probably be a big benefit as well.

Hog out the bridging on the mill, straighten and stress relieve, and then finish turn and grind.
 
Correct me if I'm wrong, but it seems as though you are trying to find the most economical process to manufacture a small batch of crankshafts. Since you don't intend to turn 10k rpm, wouldn't a cnc turned crank be more economical? Fiddling with patterns and foundries to produce a blank which still needs to be roughed, finished, drilled etc. seems a waste. As Michael Moore pointed out, the Mazak Integrex could turn out what you need and probably not need grinding. After you analyze the total cost you migh be better off finding a shop with the proper hardware who can finish machine a crank from pre-hardened 4340.
YouTube link to the Integrex making crankshafts:

http://www.youtube.com/watch?v=E1Fj1j8Sg1g

You are undertaking a most impressive project, but why re-invent the wheel? Several others have built scale Merlins, why not look them up and ask for some advice? Barry Hares in England made a gorgeous example:
http://www.enginehistory.org/merlin_xx.htm
 
Why not ask the RAF Museum at Hendon to make their Air Pubs on the Merlins and Spits available?

Oddly, I was in charge of the Technical Library at RAF Hendon between 1949-1950 when the last of 601 and 604's Spitfires took off for RAF North.

There is certainly one of 'our' Spits- a LF16E is in the US but SL721 based at San Diego had a Griffon 65 and not a Merlin.

Trying to get the old memory churning about the US, I now recall Packard made Merlins under licence.

Does this help?

Norm
 
Nice to think that the Merlin powered P-51 Mustang changed the course of the War.
The Spits and Hurries were the short range heroes but the Mustang was the long range hero right into the heart of Germany.

Thanks, John, for 'The Memory'

Norm
 
would there be any advantage to machine this in sections then pin and tunnel braze it together. I would think dowel pins would position everthing nicely and a face to face brazing will be very strong. for what it is worth.
 
Michael: Thanks for putting the picture of the cast crankshaft blank up. I put in the link as indicated, and it did show up once, but that was it.

Chris: Thanks for the link to the Barry Hares Merlin. I actually visited him some years ago, and he very kindly showed me his Merlin and some other things. His method of producing 'castings' fascinated me. He would rough out an aluminum block, and then chisel it and pick away at it to get the finish shape. Not something I'd care to try. I asked him how long it took him to build the thing - as I recall, he said 12 years. I'm surprised he was able to do it that fast. He obviously is an absolute perfectionist, something I'm not. I jokingly say that he must be an alien sent here to make the rest of us look bad. I have asked him a couple of questions about certain things, like for instance about his functional magnetos, but I feel funny about asking him things like 'how did you machine your crankshaft?', because after all, I successfully machined a crankshaft, and do you think he's going to have some magic way to make it much easier? I doubt it. But then again, he was a machine shop foreman at Rolls-Royce, so he may have been able to use more heavy-duty machines.

Also thanks for the link to the crankshaft machining video. Great stuff! That would obviously be the best way to have my crankshafts machined. However, I suspect that the setup charge alone would be substantial. I have spent far more time than money on this project, and I don't plan on changing that. The market for these castings is very limited, as will be finished engines, if I ever manage to build some.

As to the question about accuracy: When I started this project, my goal was to produce an engine which looks like a Merlin, sounds as much as possible like one, and is capable of flying a scale Mustang, etc. However, it also should be as easy to build as possible, so that the chances of that happening were greater. A 1/4-scale Mustang is cool, even if it's powered by a two-stroke. It would be even more cool if it were powered by a V-12, especially if it looked something like a Merlin. If it were an exact scale model of a Merlin, that would be even better, but a lot more more difficult. However, once I finished the prototype, and started working on the 'real' version, I started trying to make it more scale. I did get the drawings, but I've never completely abandoned the idea of simplifying things. When designing a wax mold, sometimes putting in some details can increase the complexity of the mold considerably. The best example is the wheelcase. I lucked out and got a Merlin wheelcase cheap, and I spent a lot of time working on a wax mold to produce a scale version of it. It was extremely complicated, with lots of complex water-soluble cores, and I finally decided that finishing it was just going to take too much time, so I designed a simplified version that didn't require any wax cores. Nobody's complained about it. Undoubtedly there are people who don't want to buy my castings because they're not scale enough, but there seem to be plenty of people who like them.
 








 
Back
Top