Cylindrical interference fit?

I would recommend that you check your measurements as carefully as you can because with a shrink fit you only get one shot and it needs to be perfect. When you have made measurements up to now, did they repeat to .0005? Did you measure at 90 degrees to the first measurement and were they the same? Try to control all the variables that you can and good luck -- Jack
 
Getting there. Slowly but surely;



Still to do; finish the bore to final size for the transmission input shaft, machine a groove on the large diameter (so I have a place to grab with a puller when I need to remove it), machine a recess in the large diameter for the transmission input shaft retaining bolt and broach a keyway in the trans input shaft bore.

Then its just shrink for the CBR600RR wheel hub to the center diameter and it’s good to go. Couple more days of just lazy paced work should have it done.

Then its make the offset sprocket for the engine, cut the chain to fit and move on to building the exhaust.

Once thats done, a little work to finish up some small bit, wire it all up and then a test run. Maybe in a week or two’s time.

:)
 
A 4 to 5 thou shrink fit is going to put a lot of stress on the thin aluminum casting once it cools. I'd be tempted to knurl the shaft, true it up and maybe go with 2 thou interference. The aluminum will bite into the knurled surface when it shrinks.

Greg
 
I think I'm going to make change to the drive adapter construction. The mechanic in me really is rubbed the wrong way with shrink fitting the hub to the adapter and there being no way to remove it without , essentially, destroying it. After all the time put in to building it, I sure would be "peeved" if something broke and there were no reasonable way to salvage the remaining good pieces.


So I think I'm going to change from a shrink fit to a press fit. This will mean less "holding power", but I'm going to fix that with another change. I'm going to machine a 3/8" plate to fit on the large diameter of the adapter and then grub screws through the aluminum hub to the 3/8" plate. The press fit will keep it running true and hold torque inputs, the grub screws will aid in keeping the hub from rotating and help handle the torque. Where the screws will be going is the ticker part of the piece, approx 0.250". It also open the option to use 5 to ten screws if I want (there's 5 compartments in the cush drive). One row in close to the hub, one row further out by the edges. It would pretty much have to blow apart for that to fail under load...


I also decided to machine a groove in the large diameter. This will give me a place to attach a puller if/when I need to get the entire assemble off the transmission input shaft at sometime in the future. An example of needing to get it off in the future might be changing the driven sprocket to change the final drive ratio:





The adapter is a slip fit on the shaft, but a little corrosion and we all know how easy it is for these type of things to "rust weld" themselves together over time. The ability to use the puller makes it feasible to actually get it back apart in that case! LOL!
 
Last edited:
f350ca said:
I'd be tempted to knurl the shaft, true it up and maybe go with 2 thou interference. The aluminum will bite into the knurled surface when it shrinks.
Click to expand...


An Arctic Cat snowmobile drive clutch is done this way.
I don't know what clearances they use, but if you cut the fixed face from the center shaft the knurles are visible as well as some kind of adhesive.
 
Actually, come to think of it, it might make more sense to just use a socket cap head screw and instead of going from the inside out, come from the back of the steel plate into the aluminum body.

That would give me a full head to grab when installing and removing the components. For that matter, I guess I could use a standard head bolt if I wanted since I would no longer have to worry about the bolt interfering with the internal cush drive dampers....
 
Last edited:
Test fit complete! Hub goes on the adapter slick as snot!

I took a few thou off the adapter since last posting.

I used the same technique I use to drop #2 engine bearings on compressors (one of the things I do/did at work is build jet engines from the compressor disc up).

I placed the adapter on a nice thick, flat slab of granite (I have a couple pieces that I use for surface plates when building racks of carbs and such) for a couple hours to get it temp stable with the room. Then I wrapped the hub in tin foil and in to the oven it goes (500c) for 20 mins.

I pulled it out of the oven, move quickly out to the garage, unwrap the hub and slide it down on the adapter. Goes right down to the shoulder with no problems. Well, a little drag, but that's more about me keeping it square than it not fitting.

Since I need to still cut the keyway, I pulled it back up off the adapter before it had a chance to contract.

I didn't want the hub locked to the adapter in case I mess up the keyway. I had planned to put it together and then cut the keyway, but common sense won out and said to fit the hub after cutting the keyway. It would suck to mess up the keyway and have to throw out the adapter but I'd have a major meltdown if the hub was locked on to the adapter and I flubbed the Keyway. If I broke the hub trying to get it off, I'd need another CBR600RR wheel to turn the hub out of! No thank you, I'll wait to make sure the keyway is right first!

I worked it down to a measured 0.003" interference fit with maybe a +/- 0.001" error because of the measuring tools I have to use here. The way the hub slid on and off, it could very possibly be more like 0.002" or 0.001". Definitely won't go on when cold, so it's an interference fit for sure. If it is closer to 0.001, the back up plate and bolts I plan to add later will help lock it together to help deal with the drive torque it will see.

So, tomorrow; build myself the broach and if time, cut the keyway. After that is done, drop the hub on the adapter and on to making the offset sprocket for the engine output shaft.

Coming along nicely. Can't wait to take it for a proper rip across the field!

:)
 
There's so many structural variables to determining necessary interference #'s. I generally work with steel / stainless sleeves pressed into aluminum and I am usually working with diameters around 6" and I typically see a .003" interference. Here's a neat calculator I used when I put a Kawasaki KX500 engine in a shifter kart project :

https://www.amesweb.info/InterferenceFit/InterferenceFit.aspx
 
Well it took a fair bit of "grunt work" work and patience to get there, but the input adapter was finally ready for the hub:



I knurled the hub shoulder in order to give it a little more "bite" once the hub was shrunk on there. I had .0005 - .001 before the knurling, .0025 after knurling. I reduced that down to 0.002 by giving it a little polish with some 600 grit, which probably just knocked off some of the extraneous sharp bits from the knurling process that was messing with me being able to get an accurate measurement.

Then the hub went into the oven to soak at 500f for 20 mins:



I wrap all my heated parts in tin foil for the heat soak. Occupational habit. When I'm building jet engines and fitting something like a #2 engine bearing, we always wrap it in foil to try and keep it as hot as possible on the trip from the oven to the compressor stand.

One last check of the input adapter to make sure it's microscopic clean (IE: nothing to hang or cock the hub when it goes on) and then the foot race from the oven to drop the hub down on the adapter:



Nice. Slams right home on the shoulder of the adapter. You can't get a .001 feeler gauge under it:



I'd normally do what we call "drop measurements" to make sure it's fully home, but it's more than good enough as it is for this application. No light under the adapter works for me!

Once cooled, that sucker is locked on there and no cracks in the cast aluminum hub.

:aok:

Then I assembled the rest of the sprocket drive adapter:







and slid it on the transmission input shaft:


Perfect!

Now it's just make a spacer to take up the remaining shaft you see in the video to the inside shoulder. Mostly for a finished appearance as much as anything else. It will also be used to help retain the sprocket carrier in the cush drive, a wider shoulder at the sprocket carrier end of the spacer will handle that chore. Easy stuff!

Gotta say, I'm pretty pleased with how it all turned out. Took me quite a while to get it done, but that's what happens when you are designing and prototyping at the same time I guess. I shudder to think how much it would have cost to have a shop build this for me! It's not finished yet, but the hard/critical work is now done!

:beer mugs:
 
Last edited:
Really nice design and fabrication work.

You mentioned "build myself the broach". For the admiring newbies, can you explain how you get key slots in what appears to be a blind hole? I thought broaches were normally set up to pass thru a hub with a slotted guide bushing. Can you explain how you made the key slots?
 
You must log in or register to reply here.
Back
Top