How to assemble a shaft precisely square to a plate

Hi,

I have an aluminum plate which has been ground flat, and I would like to insert a few precision ground shafts into this plate and have them rigidly/permanently affixed in place, precisely square and perpendicular to the plate. The shafts diameters are .75" and the plate is 1" thick. The shafts will protrude 6" out and I would like it to deviate from squareness by less than .001" over the 6". The straightness of the shafts are rated at .001" per foot.

My initial idea is to bore out the holes a couple thou oversized and then loctite the shafts into place, holding them in place by clamping them to a precision ground v-block w/ .0001" squareness tolerances. Or I could beuild a fixture to adjust the tilt and and measure the squareness with a squareness gage, but I'd have to buy one.

Does this sound like a good idea to anyone? I have a Fadal 4020 VMC and I suppose I could square the plate to the quill and bore for a press fit, but I'm not sure if I could ensure that the shaft would go in straight. If it didnt then I'd be upset, and we wouldn't want that.

Thanks for any advice.

Aaron Keit

Reply to
ADkeit
Loading thread data ...

If you're concerned that the hole won't interpolate round, you could always use a boring head, or improvise a boring bar. But it seems to me you are more concerned about the axis of the hole being square to the plate, which I would think is determined by the setup of the plate in relationship to the spindle centerline. You should be able to prove this by sweeping the plate with an indicator attached to a toolholder mounted in the spindle.

Reply to
Jon

Reply to
JR North

get yourself some smooth walled steel tube. chuck it in the lathe and turn across the end with a facing cut. make the tube about 6 inches long. put face cuts across both ends. make 2 or 3 of them.

stand the tubes on the plate beside your shaft. align the shaft with the tubes and you should have a perfectly square shaft. hth Stealth Pilot

Reply to
Stealth Pilot

Sporty numbers!

Loctite is too fast for that, if you meant thread locking loctite. 741 (?) starts to react in about 10 seconds. I would think about using a two-component epoxy and glue it in with that. That procedure is often used for aligning the guiding columns of punching tools.

Nick

Reply to
Nick Mueller

I understand the squareness condition. How about placement? Do you have a thousanth's "play" available in the actual locations of the shafts?

If so, I'd recommend pressing the shafts into proper-sized holes, then controlled peening around each hole to "move" the shafts into precise vertical position. Of course, this would disturb the flatness of the plate around each shaft -- the peening can be done from either side, if the shafts are pressed all the way to the opposite surface.

LLoyd

Reply to
Lloyd E. Sponenburgh

Can you drill and tap holes in the plate, and turn down and tap matching threads on the end of the shafts? Then you just have to have the shoulder on the shaft square.

You don't mention what the loading is on this shaft, but gluing the shaft into an oversize hole with epoxy or CA should work, but you'd need several squareness gages, which at 0.001 precision you might turn yourself on a good lathe, to jig it in place during the cementing.

Reply to
Richard J Kinch

That's a very tall order. As long as the Loctite will withstand the required forces without yielding, that may the best way.

If you bore the plate and press the rods, a rod that's bowed .001/12" would be out .001 from square at 6" above the plate, even if the bore was *perfectly* perpendicular and the pin followed the hole without deviation in the pressing operation.

Ned Simmons

Reply to
Ned Simmons

Thanks for the replies so far guys. The shafts will have such a slight load it hardly matters...like one pound. They are used to hold some mirrors that slide up and down the shaft but maintain their angle. I do have wiggle room in the exact placement of the shaft, so having an oversize bore is not a problem. I could turn a shoulder on the shafts and bolt them to the plate from underneath, but I doubt that would be square enough for my specs. Any other suggestions? Will loctite try to "self-center" the shaft and perhaps mess with the alignment while curing?

Thanks,

Aaron

Reply to
ADkeit

It self-centers, but not on the angle. You will be off!

If your plate is flat, you can have a thread on one rod's end with a smaller diameter than the rod itself. It will get very straight, even if the thread is not that straight. It gets even better, if you can have a bigger diameter on the rod's base.

Nick

Reply to
Nick Mueller

Hmmm... I can secure the shaft to the alignment rig very rigidly, and secure the alignment rig to the plate. Won't the Loctite takr the path of least resistance as it cures and just fill the space that is available? Plus, if I take the time to align the plate with the spindle before I bore the bore should be pretty well aligned anyway. No? I don't know...

Reply to
ADkeit

Bore the holes .0005 undersize square to the plate. Heat the plate in an oven about 200*F . Drop pins into holes. Air cool. Your welcome.

Fred

Reply to
ff

Yes it will. But you will have to apply the Loctite to the shaft before you insert it. And then, as soon as you have put the shaft into the bore, you won't have much time to align it.

But you intended to make it oversize. If you press it in, you will wipe off most of the Loctite.

You'll report how you failed. :-))

Nick

Reply to
Nick Mueller

Doable, but---why? Change the design to accommodate more reasonable tolerances. Anytime we run into a design that calls for this kind of situation, we redesign and come out with a better mousetrap.

Reply to
Tom Gardner

HMmmm, that gives me an idea. Bore the hole to the desired class of press-fit, then, without moving the machine, swap in a 3/4" toolholder and use it to press in the shafts. That should get the shafts straight into the just-bored hole.

Jon

Reply to
Jon Elson

I've made a number of gauges and fixtures where the pins needed to be very square to the plate. I've always just checked the work afterward with a 1-2-3 block afterwards to insure there was no gap, so I haven't actually indicated the work. With that said, the earlier suggestion about putting it in on the table with a collet is similar to how I do it. I put it in a precision drill chuck with the pin bottomed and press it into place with the mill in the same setup the hole was made. It develops something like 2000 lbs of force in the Z axis which will install a fairly loose press fit pin. You're guarantied the bottom and the top of the pin are in near perfect alignment.

Please let us know what you do and how it works out.

Reply to
Polymer Man

I cannot change the tolerances, this is for an optical assembly with multiple lenses and mirrors. The mirrors need to move up and down (Z- axis) without changing their alignment or X/Y position. I can change the method of manufacture to but not the tolerances.

I'm still talking about an oversized bore, just one that is accurately aligned with my desired shaft angle.

Does anyone know how they do it when installing collums in surface plates or squareness guages? I've seen products that claim to have shafts aligned square and parallel to within .0001" over 12"... so there must be a way. Not that its past anyone to lie...

Reply to
ADkeit

Why do you assume that the shoulder would not be square enough? Or for that matter, a counterbored pocket in the plate and grind the end of the shafting square? With as little load as there is to be on the shafts, I'd go with setting the shafts in a slighly press fit counterbore that is over half way into the plate, the deeper, the better, then use a fairly small cap screw from the other side to hold it in place.

Mike

Reply to
The Davenport's

plates or squareness guages? I've seen products that claim to have

They do it with two things you don't have...longer/deeper holes so that there is more engagement and fixtures that will hold said shaft/post/attachment where it needs to be then glue them in with mildly expanding glue.

Or so I was informed by a guy that I know that works at the Starrett facility in Mt. Airy, NC.

Mike

Reply to
The Davenport's

I'd make up a block with a bored hole that is the size of the pin. I would put make sure that the bored hole is perpendicular to the bottom of the block by doing the whole operation in a lathe. I would then use the block as a guide to press the pin into the plate. You could also heat the plate and drop the pin into the hole and use the guide to hold it straight.

John

Reply to
John

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.