Dene Collett wrote: > I was unaware of any failures of the type IV crank at > all.
I talked to Steve Makish again last night, and he said he'd had one stocker break and a Scat 4340 break. Neither of them had a GPASC Force One hub. He went to the Force One and put 200 hours on it before going Subaru. > I was led to believe that it was a very strong crank which if done > properly would handle a lot of punishment. Even the failures mentioned > earlier only spoke of the welded hub failing. I think if you'll ask around, you'll find more. If you want to talk strong crank, check out the Corvair. Not one single crank failure in an airplane, and there have been more of these flying since the early sixties than you'd suspect. Nobody makes a "high performance" crank for the Corvair, because the stocker is so strong. Sand dragsters in CA routinely pump out 500 hp on stock case, crank, and rods. I know y'all never imported Corvairs to SA, so I understand why you wouldn't consider that option. > I have considdered driving the prop from the flywheel end but cannot come up > with a way of handling the thrust bearing issue without adding a bunch of > unwanted extra weight. What's wrong with the stock thrust bearing that you're already using? If you drive it from the rear end, it's only inches to the prop hub. I doubt that it makes much difference to the crank, but I don't see the problem with the stocker, assuming you use the "double thrust" bearings that almost everybody uses (use the thrust half from two sets of bearings, so it works from either direction. > As a matter of interest I am getting a scrap crank from a local VW repairer > and am going to conduct a test to see just how much torque the standard > taper will handle before failing. I am first going to use some grinding > paste to lap the taper surfaces for a perfect match, put it together with > locktite and preheat the hub for a slight shrink fit on the crank.For this > test I will be using a welded flange to the pulley hub with a long lever arm > bolted to it. But the largest forces are gyroscopic (which are trying to bend the crank repeatedly in opposite directions, leading to fatigue), not torsional. You'll be testing for something that will probably not be the primary mode of failure. > My guess is that for every broken crank out there, there are at least twenty > that have been working fine for many hours.There was probably a very good > reason for them breaking as well. I personally know of four KRs here that > have used that system and have flown many hours. I must concede that they > all use the longer taper as in the force 1 hub (unsupported). You think a one in twenty risk of eventually having a forced landing is acceptable? Not for me, and I consider myself to be borderline foolhardy. And I can't help but wonder why you wouldn't start with something proven, like the GPASC Force One setup (although price would be a good guess). But I get the impression that even GPASC would like for more folks to go rear drive than the old Force One way (although I could be wrong about that). > If you could find that info I would appreciate it. I looked at this for about two hours this morning. Machinery's Handbook deals mostly in low rpm "machinery" type stuff, so I didn't get very far there. I searched the web for some shaft design stuff, and got bogged down in everything but what I was looking for. Then it occured to me...the reason I ditched the Type4 VW and went with the Corvair was because I didn't want to have to design a drive shaft. I have a 3" thick notebook that's chock full of design data for shafts, couplings, bearings, housings, torsional vibration, etc., and I STILL wasn't sure exactly how I was going to do it. Note that I didn't get into keyed or tapered joints at all, or I'd share that with you. When the Corvair came along, it didn't take long to jump THAT ship! The bottom line is that designing something like this is not trivial, even if you ARE a mechanical engineer. My specialty is not power transmission, but "design" work, where I try to think of the best possible way to do something, while balancing cost, manufacturabilty, weight, and that sort of thing. If in doubt, I'll ask a stress, thermal, or vibration guy to fine tune my work during the design process. Because the company I work for specializes in space and defense work, we don't have any real power transmission whizzes that I can pester about this. I did bounce it off of the stress guy that analyzed my prop hub for me. It happens that he's an aeronautical engineer, and has designed several planes around the Type 4. He's the sharpest engineer of any kind that I know. He laughed out loud when I told him about the plan to run the engine off the pulley end, and mentioned that even Steve Wittman designed a simple rear drive system in the sixties for his Formula Vee racer, and he was only putting out 36 horsepower! He echoed my sentiment to "go with the proven" GPASC Force One system. Even the guys who specialize in this stuff know that it's not an exact science. Torsional vibration is something that's almost impossible to model on a computer when the system is as complicated as an engine, even for the guys that I work with. There are way too many variables and interactions to ever get it right. I know of one aircraft engine company that paid an engineering company to design a shaft drive system for them, and the first example stuck on a customer's Type 4 self destructed in minutes! Contact! magazine is full of stories of guys who tried to make their own PSRUs or drive systems for automotive engines, only to have faced failure several times before experimentally working things out sufficiently to call it sucessful. Many of these experiments involved deadstick landings. I guess what I'm saying is that this has the potential of being a long road, with lots of problems, not the least of which is suddenly flying with no propeller. I hope you'll decide to do something that's been proven already. I know that sounds funny coming from me, but you've got to draw the line somewhere. You are welcome to prove me wrong and crow about it, but I'm not going to encourage it... Mark Langford, Huntsville, AL N56ML "at" hiwaay.net see KR2S project at http://home.hiwaay.net/~langford
