[bmurph]

I need to make a precision flywheel with OD 1.9" ID 0.25" x 0.375"

I've made one, but my arbor broke as I took it apart. First using a 2" round I turned, faced, and drilled, bored the center hole. Then cut it off. Made an arbor to slip fit the center hole, made a 10-24 tapped hole in the end of the arbor. With a marker, I marked one of the chuck jaws and the arbor where they meet, this way I can reuse the arbor, and it will be alined within 0.5 thou at least, because the shank of the arbor is not precise. I put the workpiece faced side first in the arbor, and clamped it in place with a 10-24 screw, so I could face the rough side. The problem is I made the arbor out of aluminum, and the tapped hole had a very thin wall around, so when I took it apart that section of the arbor broke off. I'm wondering if there's a easier way, and if I should use 1018 or 4140 for the arbor, or would a smaller screw clamp hard enough? The flywheel material is aluminum. I've tried making a very slightly tapered arbor, but when I press fit to the workpiece, it bends.

[azmachining]

1018 should work much better, more rigidity and strength. Or you might be able to get away with a expanding mandrel and not have to make anything.

[Dylwad]

I have used super glue and green loctite before to hold the workpiece to the arbor, just ise some heat to soften the glue to break the parts free.

[basalt]

I would turn, face and bore the part then part off part way, come out and take a small clean up cut on the back face of the part then move over and finish parting off. That way you can turn the part around, grab it on the O.D. and then dial in the cleaned up face and face it to length. Use some shims between the chuck jaws and the O.D. so you don't damage the part. I hope the attachment comes thru. Hope this helps.
Colin

[LotsOfChips]

Do you have a "spider plate" for your lathe? Its just a round plate with at least 3 evenly spaced set screws, and some slots cut out of it in the same pattern as your chuck jaws so that it slips past the jaws. Obviously you have to leave material in the center so there is a limit to how small of a part the spider plate will allow your chuck to grip. The set screws should stop against that back of the chuck, and your part should rest against the spider plate; a light tap with a soft hammer help it sit flat. The set screws adjust for workpieces of different thinknesses by setting the plate's standoff distance from the back of the chuck. Just make sure all of the set screws are at the EXACT same height (ideally with a height gage and a surface plate but calipers are ok) and use a lock nut. Done correctly, even real thin parts will hold with minimal if any runout. And do use shims to protect the part, per advise from the previous post.

[bmurph]

thanks for the help. Now how would you make that round into a pump/turbine rotor, with a base of 0.070", as seen in the attachment.

[Juaner]

Well I was also looking for some kind of assistance for doing some experiment with the flywheel and you guys really make my work really handy.
Lets see how I do with the flywheel!!

[basalt]

That rotor is probably a casting, grab the OD, face and bore, mount on a mandrel or fixture, turn and face.

[bmurph]

I'm using a CNC mill to make the rotor.

[LotsOfChips]

Bmurph,

I would probably cut a .070" deep circular pocket into so some aluminum soft jaws in your vise (I'm assuming you have a 4-6" machine vise in your mill). Maximize support underneath your part so minimize the gap between your fixed and floating jaws when you cut the pocket. I think 1.5" is a common width for a soft jaw, so 2 of them will be sufficient. Obviously some air space is required so that the vise clamps your part before it closes, so I would clamp a sufficiently short .125" parallel between the jaws when you cut the pocket. Make sure you use a SHARP tool (any corner radii at the edges of the pocket is detrimental, especially with thin parts). Also, make the ID very close (within .001") to the OD of your part, so start small and use cutter comp to dial it in. This will assure that there is no localized pinching and that entire OD of your round is in even contact. 070" isn't a lot of material to grab onto but its very doable as long as you have a precise fit. I forgot to mention, do not overclamp your part.

[bmurph]

I raised a parallel between 2, 123 blocks, and with long 10-24 screw, bolted it through the center hole, with a piece of scrap under the parallel to use as a washer, because the parallel's hole is too big.

[LotsOfChips]

How did that work out, or have you run the part yet? That's an interesting fixturing setup. I'm assuming this is all located in a vise? I'm no expert, and I can't say its unfeasable but my visualization of this setup would concern me. First of all, the wall around that the center ID of the finished part looks thin. That 10-24 is going to need to be fairly snug to securely hold that part down and such torque may crush or deform that wall once excess material is gone. Also a single bolt in the center may not supply much clamp force at the outer areas of that part. Part rigidity is a big factor here as the ID wall and the .070" base approach finish size. Real potential here for chatter and edge "flapping", with the resulting poor finish and dimensional inconsistency. Worse case is the part lifting up at the edges and spinning in the fixture. You may be able torque it down for roughing and then loosen it some amount for finishing, but then you have to be careful about not losing the clocking of the part.

Second, assuming that the parallel laying flat and clamped in a vise, I'd have some doubts about that parallel being secure albeit I've never done this myself. A possibly remedy would be to just use a piece of aluminum at least 2x2 in size and at least 3/8" thick. Hold that in the vise normally, face the sides as required, and tap it (save this hole position to use for your part origin).

Again that's just my .02, and there's only one way to find out.

[bmurph]

I've made a few of them like this, they turned out ok, but I didn't do a finishing pass, used a 4 flute 1/4 inch. I used T-slot clamps to hold the parallel on top the 123's.

[LotsOfChips]

Ok now I have a clear picture, and that is pretty solid. And it sounds like the part had the rigidity so good work.