Setting up a worn-out motor armature in a steady rest.

[Just a Sparky]

Got a semi-hypothetical learning question here.

Suppose you've got an old motor armature with a damaged output shaft. Chewed up so bad the key seat is barely recognizable but the motor [for whatever reason] is still valuable and worth keeping. So you want to turn down the shaft to the next standard size to get rid of all the booger.

Shaft journals are heavily worn and not evenly so. [Motor fitted with undersized replacement bearings.] Factory centers are no good.

How would you set up that job in a four jaw chuck and a steady rest?

The traditional method for aligning a steady rest involves tramming an indicator back and forth along a known cylindrical surface, referencing the lathe's carriage. (Or the non-wearing top & side surfaces of 'vee' ways via a jig in the case of a worn machine.) But the armature from an old motor with plain bearings has no reliably cylindrical surfaces to indicate. The journals are worn round but acylindrical and are therefore unreliable for this purpose. The laminations happen to be hand stacked, and not only that, were damaged by a prior accident and are thus also no good for indication.

How would you handle a job like this?

 

[52 Ford]

Got a hypothetical learning question here.

Suppose you've got an old motor armature with a damaged output shaft. Chewed up so bad the key seat is barely recognizable but the motor [for whatever reason] is still valuable and worth keeping. So you want to turn down the shaft to the next standard size to get rid of all the booger.

Shaft journals are heavily worn and not evenly so. [Motor fitted with undersized replacement bearings.] Factory centers are no good.

How would you set up that job in a four jaw chuck and a steady rest?

The traditional method for aligning a steady rest involves tramming an indicator back and forth along a known cylindrical surface, referencing the lathe's carriage. (Or the non-wearing top & side surfaces of 'vee' ways via a jig in the case of a worn machine.) But the armature from an old motor with plain bearings has no reliably cylindrical surfaces to indicate. The journals are worn round but acylindrical and are therefore unreliable for this purpose.

How would you handle a job like this?

That is an interesting problem. The first thing that comes to my mind is to put the armature on a surface plate, and set a height gage at 1/2 of the diameter, then use that to find the center point of the shaft, center punch it, then center drill it.

Does that make any sense?

Edit: I meant set thr height gage to half of the diameter of the part of the armature that's resting on the surface plate.

Sent using Morse code on - .- .--. .- - .- .-.. -.-

 

[markz528]

At that point the air gap is what is important so indicate off the iron. You say armature - so this is a DC motor? If AC, you can skim the rotor iron to get it all in the same plane. I would not skim a DC armature.

How big a motor are we talking?

 

[Just a Sparky]

@ 52 Ford, I think I see where you're going with that.

Trouble is, the same crash which damaged the lamination stack also bowed the shaft a bit, hence the no-go on the factory centers.



EDIT:

@ markx528: AC repulsion motor. Approximately 80 lb unit. Armature O.A.L. appx. 14-16" or so. 5-6" swing. Journals somewhere between .900 and 1.000 off the top of my head.

I'm concerned with the power output portion of the shaft for this question, which is overhung from the bearings. Main concern here is getting the steady rest in line with the spindle axis to keep the armature from walking itself out of the chuck.


The only solution I can immediately think of is to start by measuring the diameter of the spot where the steady is to run along the front journal.

Then drill an accurate center hole in a piece of slightly oversized stock, turn down the end of stock to the measured diameter and then extend the stock out to the spot where the steady is, support with a center and dial in the steady that way using the stock as an analogue for the motor shaft. Said stock could then be proven for alignment with an indicator, then substituted for the real motor shaft as long as the steady fingers don't get moved.

 

[jim rozen]

You probably want to turn this between centers. No centers in the shafts? That's your first job.

 

[52 Ford]

@ 52 Ford, I think I see where you're going with that.

Trouble is, the same crash which damaged the lamination stack also bowed the shaft a bit, hence the no-go on the factory centers.



EDIT:

@ markx528: AC repulsion motor. Approximately 80 lb unit. Armature O.A.L. appx. 14-16" or so. 5-6" swing. Journals somewhere between .900 and 1.000 off the top of my head.

I'm concerned with the power output portion of the shaft for this question, which is overhung from the bearings. Main concern here is getting the steady rest in line with the spindle axis to keep the armature from walking itself out of the chuck.

Rephrasing my last comment

If you set a height gage to half of the diameter of the motor rotor and set the rotor on a surface plate, you can use the gage to mark the center line on the end of the shaft. Mark it. Roll the rotor to a different spot, mark it again, and so on till you get a good idea for where the center is. Center punch it and then center drill it.

I'd use a 4 jaw chuck and a put a center in the lathe and indicate of the laminations when you're tightening the chuck.

Sent using Morse code on - .- .--. .- - .- .-.. -.-

 

[52 Ford]

You probably want to turn this between centers. No centers in the shafts? That's your first job.

Putting the centers in the shaft is what was/am trying to explain.

Sent using Morse code on - .- .--. .- - .- .-.. -.-

 

[Just a Sparky]

I see where you're going now. Yeah, that makes sense. Could get it within a few thou that way in a steady rest if you've got good eyes or a loupe and a really sharp tailstock center.

Hey, I like that idea even more if one were to modify it slightly by putting the journals up on vee blocks instead of rolling it on the laminations. With a worn front journal the shaft would incline downwards slightly, but you'd be able to find the center of the rotating axis that way very accurately, since the journals are the constraining surfaces when the armature is installed in the motor. Wouldn't matter at that point if the shaft is bowed, since you'll be emulating the exact same rotation it will experience when installed.

 

[52 Ford]

I see where you're going now. Yeah, that makes sense. Could get it within a few thou that way in a steady rest if you've got good eyes or a loupe and a really sharp tailstock center.

Hey, I like that idea even more if one were to modify it slightly by putting the journals up on vee blocks instead of rolling it on the laminations. With a worn front journal the shaft would incline downwards slightly, but you'd be able to find the center of the rotating axis that way very accurately, since the journals are the constraining surfaces when the armature is installed in the motor.

So is this still hypothetical?

Sent using Morse code on - .- .--. .- - .- .-.. -.-

 

[Just a Sparky]

Based on a past "could have, should have" moment, yes. Lol.

I feel smarter now.

 

[magneticanomaly]

Here is how I would do it. I assume that all you need to work on is the output shaft, the part you would mount a sheave or coupling on,

Grab the journal opposite the end you need to work on in 4-jaw chuck, padding with copper or al if needed to protect journal surface. Do not swallow the whole length of the journal in the chuck...you will need to indicate it later.

Adjust chuck so that the journal at the bad end runs true. Does not matter how far out the end in the chuck is.

Mount steady, run fingers up to journal surface gently and lock.

Now readjust chuck so that the journal it is holding runs true.

If you want to drill a center-hole in the shaft extension you will be turning down, now is the time. It will allow you to take heavier cuts, but you will probably slightly increase your TIR. The smaller the hole you drill, the sharper the drill, and the gentler you drill, the truer everything will stay.

You could also weld up the shaft and turn it back to original size

 

[jwmelvin]

Here is how I would do it.

This method makes so much sense; thank you for sharing it and I’ll look forward to next time I set up my steady.

Edit: indicating along a cylindrical surface may be faster when possible but learning this additional method makes me very happy.

 

[eKretz]

There are a couple other ways too. Setting the part up and dialing it in while it's hanging free and bringing the rollers up afterwards is an easy one but you can still get considerable sag that way. You can mount an indicator to the workpiece and sweep the tailstock center as another method or also to fine tune the "bring the rollers up" method.

 

[Just a Sparky]

I like the idea of sweeping a center.

What sort of indicator holders would work well for something like that? Maybe just use some hose clamps to fasten a ball-jointed test indicator to the work?

 

[HappyWyo]

Here is how I would do it. I assume that all you need to work on is the output shaft, the part you would mount a sheave or coupling on,

Grab the journal opposite the end you need to work on in 4-jaw chuck, padding with copper or al if needed to protect journal surface. Do not swallow the whole length of the journal in the chuck...you will need to indicate it later.

Adjust chuck so that the journal at the bad end runs true. Does not matter how far out the end in the chuck is.

Mount steady, run fingers up to journal surface gently and lock.

Now readjust chuck so that the journal it is holding runs true.

If you want to drill a center-hole in the shaft extension you will be turning down, now is the time. It will allow you to take heavier cuts, but you will probably slightly increase your TIR. The smaller the hole you drill, the sharper the drill, and the gentler you drill, the truer everything will stay.

You could also weld up the shaft and turn it back to original size

You can't weld it up and bring it back to specs until you have found center.
If the journals are bad you can't use them for your steady.
We would have two choices, on a heavy motor ( >100 lbs ) which is 95% of our rotor rebuilds.

The easiest is press out the old shaft and press in a new shaft.

The other option is to use two four jaw chucks, one on each end so it can be adjusted to perfect using a dial indicator, ( finding a good surface for your dial indicator on a rotor or armature that has been damaged may be a challenge ) then turn a journal for your steady to ride on. Set up your steady, remove your tailstock chuck, drill the end of your shaft, you may have to use a cutting tool to cut the hole true if a existing center hole won't allow you to drill a perfect hole, switch to your center, and remove your steady.
Now you can turn your shaft and journal down until it is true, weld it up, and turn it back down to specs.

A rotor that is swinging off center, or out of balance will cause vibration and destroy components.

I made my tailstock chuck adapter for a South Bend Chuck. I can just screw it off my small South Bend lathe and screw it onto my tailstock adapter when needed.

PS having a live tailstock makes it easier to attach a chuck but it can be done on a conventional also. It just takes a little more work to make the chuck live.

I'll have to post my adapter online when I get a chance, so you can see how simple they can be.

 

[Clive603]

Thinking laterally is there any mileage in setting it up on a vertical mill to get the centres drilled?

Four jaw chuck on a rotary table to get it centralised. Very light support at the top from "something" in the spindle so it doesn't ry to fall over. Extra straps when aligned so it safe to drill.

Flip to do the other end.

I've done similar on something rather smaller that need to be grabbed by the middle, wood holder bored in situ did the deed. Would have needed a much larger chuck than I had to do that on the lathe.

Clive

PS As its the air gap that matters maybe making a snug fit carrier to go round the laminations in situ with pre cut slits so a suitable bolt or three can nip it up to hold might work. No chuck or rotary table needed.

 

[Just a Sparky]

Air gap isn't a concern for this question.

It's a very old repulsion motor with hand-stacked laminations and a very generous air gap which predates the practice of rotor balancing. Can't replace the shaft because the lamination stack isn't welded or poured. All of the laminations are held in place by a key on the rotor shaft and compressed by a big nut. Once removed, I'd never get a new shaft back in and the whole thing would be junk.

The motor is fitted with extra-large, 3" long bronze bearings which are held in place by lots of iron and are more than adequate to handle the vibration caused by the bow in the shaft. These motors are known for having bad vibrations anyways because they were never balanced. That technology didn't exist when they were produced. Even with the bow in the shaft, the motor runs just fine. As they say, "If it ain't broke..."

The only part of this rotor I would be machining for the purpose of this question is the power output shaft since the undersized bearings have taken up the journal wear just fine. This output shaft needs to be concentric to the axis of rotation when installed in the motor, therefore the shaft journals are the correct surfaces to reference when setting up, not the lamination stack.


On another note, is there any reason a guy couldn't stick weld the power output shaft to build it up before re-machining? Or are submerged arc/spray welding processes necessary for this? I know those are frequently used on bearing journals to minimize distortion, but for an overhung shaft that's already shit-shot to begin with...
I guess my only concern would be hardening/embrittling it accidentally.

 

[Tony Quiring]

Here is a "I do not have fancy tools" suggestion.

Is the very end of the shaft still round ar can be made dound with a file ( to remove a dent)?

You should be able to make a simple drill guide bushing that slips onto the end of the shaft using the very end as a reference and maybe have a very small hole as a drill guide.

This allows for same as center punch but actual spot drill.

Now you can center drill.

Sent from my SM-G781V using Tapatalk

 

[magneticanomaly]

Yes, Sparky, you can stick-weld the shaft. that, and oxyacet is all I do. Preheat minimizes risk of embritlement, but has to be limited if the rotor is wound lest you cook the varnish.

If the actual problem is only a wallowed keyway, you can make a special key, whose top fits the good keyway in the sheave and the bottom is fitted, by filing and scraping, to fit the wallowed keyway

 

[Just a Sparky]

So how does one achieve a good welding ground on a shaft which is to be spiral-welded? Would be a real shame to blast pits into the journals...