Lathe alignment problem.

[mcostello]

Grab a cuppa of Joe or some popcorn as this is a bit long.

I have a 15" x 50" Clausing Colchester lathe that has a prblem that has Me stumped.
Since when I bought it used it has been out of alignment and I cannot figure out what the problem is.
It turns a taper of .002 in 2" and I cannot seem to adjust it out. It happens no matter how long the shaft is or where it gets turned. I have a Starrett precision level and the lathe levels up like it's supposed to. I have tried shimming up one corner to bias the lathe with no success.
All the gibs have been adjusted. The compound could use being scraped in as some wear is evident.
Locking the compound and cross slide down produces no change.
I can put a 2' bar in the chuck and with about 50 lbs. force no change is evident.
The lathe has a set of springs with 24 holes provided, and only 20 springs installed when I got it. I bought a new set of 24 and installed them. The taper was .012 in an inch and this bought it down to the present value of
.002 per 2 inches.
The only thing left was to adjust the headstock to be angled towards the bank. The taper in the part is such that the part is bigger towards the headstock. This is true weather turning or boring a hole.
I adjusted the headstock at 3 different times with no result. I have kept notes and move the headstock .002 at a time with a dial indicator ona ground shaft. The movement makes no change. I have moved it a total of .022" with no improvement with the result being the tailstock is no longer aligned and I could not drill a hole correctly.
I thought the chuckmight be in error as it was in bad shape so I bought a new 4 jaw. Since then I have added a collet set up.The 3 jaw and 4 jaw chucks and collet setup all make the same error.
When I use the tailstock with a live center the problem seems to go away and the error can be dialed out.
It seems like the problem is with the spindle bearings and I have removed the spindle and inspection revealed no apparent problems. No signs of damage apparent. Spindle is not bent, nor any signs of a wreck
visible.
Any suggestions or tips are gratefully appreciated.
Spindle bearings are around $3000 if purchased from the factory. I am trying to avoid that.

 

[goldstar31]

My guess is bed wear and possibly the mating saddle. Remember that wear of a 3 thous in the first shear near the spindle conceivably results in 6thous taper in a machined part.

I hope that this is constructive advice

Norman

 

[Anatol]

Norman's advice is from lifetime experience, mine isn't. But I have a similar issue but the runout is on the opposite direction (should yours be called run-in?). See "3jaw chuck alignment problem" thread, many wise suggestions.

Its difficult (for me) to imagine a mechanical situation that would result in the bar drawing a cone where the point of the cone is at the outer end and the 'base' off the cone is in the chuck! That implies the point, floating in space, is a fixed point.

"The 3 jaw and 4 jaw chucks and collet setup all make the same error."
So its not misaligned chuck jaws. Just like my problem. I believe I've narrowed my search to spindle nose. Have you tested there?

One nagging question I have is: is there a possibility of misalignment of spindle axis with the plane of the ways? You're measuring from the cross-slide. Various folks have said it doesn't matter where you indicate from. But if the spindle and ways are not parallel in both planes, surely this would introduce an error? (Not the one you have, probably.) If your lathe is like the pics I saw, it has a vertical flat face on the chuck side of the headstock. Can you/did you indicate from there?

 

[mcostello]

Spindle alignment seems to be the problem. If I move the spindle with a dial indicator, the dial shows it moves as wanted. Taking a cut shows the problem remains. Moving it again shows no improvement. Taper remains the same after biasing the spindle a total of .022. Then the tail stock does not line up at all. Had the spindle out with nothing amiss, but there has to be a piece of rubber in there somewhere.

 

[Hopper]

...
The lathe has a set of springs with 24 holes provided, and only 20 springs installed when I got it. I bought a new set of 24 and installed them. The taper was .012 in an inch and this bought it down to the present value of
.002 per 2 inches.

Which springs are you referring to here? The ones in the headstock bearing? If so, sounds like the source of your problem is bearing related. The new springs have brought your problem down from 12 thou per inch to 1 thou per inch. Pretty good improvement there. Could it be that the bearing components are binding and not moving that last little bit to eliminate the last thou of misalignment? Might pay to pull the bearings and clean and lube and make sure all parts are moving freely. You can inspect races for wear at the same time.

How you could possibly pivot the headstock by 22 thou and not see a change in the taper being turned is beyond me. Unless you are using a tailstock centre to hold the far end of the job? This of course should not be done when testing spindle to bed alignment. Job should be sticking out of the chuck with no tailstock support at all while taking test cut. Job should be at least 1" diameter and stick out no more than 4" to maybe 6".

Unless of course your adjustment is moving the headstock over by 22 thou, but parallel to the bed axis, which I think is impossible if the headstock pivot dowel pin is intact.

 

[mcostello]

The problem makes no sense to Me also. The adjustment nut is tight. The races and bearings look in excellent shape. I am not using the tailstock center when doing the test. I am wondering about the pivot pin but the headstock probably weighs in the neighborhood 0f 300-400 pounds. Would have to get the engine hoist in to move it. Does not run hot or anything else. If You read about these lathes on the Internet everyone says they are a great machine. I am thinking that the only thing doing this problem might be a mismatch in the bearing and it's race such as one having the wrong angle. Happened to a friends Jeep, it wobbled slightly going down the street, He thought it was the tires, Some one seen the tires wobble then He found the problem. I had the spindle out but did not think to check to see if the angles matched.

 

[deeferdog]

OK, I'll show my ignorance. What are lathe springs?

 

[petertha]

The only thing left was to adjust the headstock to be angled towards the bank. The taper in the part is such that the part is bigger towards the headstock. This is true weather turning or boring a hole. I adjusted the headstock at 3 different times with no result. I have kept notes and move the headstock .002 at a time with a dial indicator ona ground shaft. The movement makes no change.

Just to confirm, if the test cut taper results in larger diameter at the head stock end vs tail stock end, that would infer the head stock is pointing to the operator side of the lathe. So to correct it would have to be rotated to the opposite side of lathe. And you say you did that & no change? Where is the 0.002" referenced from? Are you putting a DTI on the end of the test bar to observe this deflection while adjusting & confirming its actually locking on this new position?

Do you have jacking type set screws that displace the head stock over once hold down bolts are loosened or how is the new position secured? Assume you are working of a manual but maybe there are more set screws that work opposite to one another (loosen one tighten the other type deal). I have jack screws & found them to be uber sensitive. I would think it would be easier to radically change the HS noticeably out of tolerance but you are saying the opposite - its muted. I wonder if for example your tightening bolts might be in contact with their through holes & you don't have sufficient rotation travel so to speak before the bolt pattern prevents it from going any more?

Another potential source is the head stock is pointing in an up or down inclination relative to bed ways. So maybe the HS was shimmed to compensate that orientation & that has been disturbed during the adjusting phase?

 

[mcostello]

Yes it seems like the head stock is pointing towards Me. I have referenced the .002 with a dial indicator, the dimensions have been observed and written down so the mistake is not with a faulty memory. Done this 3 times. It does have adjusting screws on the back that are very sensitive and locked down against each other when done. I can move the headstock .022 out of alignment and then it will not drill a hole on center because the drill makes a circle in the part. If the part was out of tram vertically I could not center drill a shaft that sticks out a foot or two.

 

[Hopper]

... I have referenced the .002 with a dial indicator, .

Again, I'm a little confused here. You should be measuring the .002" taper per 2" with a micrometer, not a dial indicator. How are you measuring the amount of taper on the test piece after turning it?

a) I can move the headstock .022 out of alignment and then it will not drill a hole on center because the drill makes a circle in the part.

b) If the part was out of tram vertically I could not center drill a shaft that sticks out a foot or two.

a) The circle confirms you have moved the headstock over, but if the lathe is still cutting a taper, it means you must have moved the whole headstock over in a roughly parallel manner instead of pivoting it at an angle. Put a dial indicator on both the left and right hand ends of the headstock and check how it moves when you play with your adjusting screws.

b) But you say you can't drill a centre hole, it makes a circle, so should check vertical alignment too. Easy way is to turn a 60 degree point on a short piece of bar in the chuck and then bring the tailstock centre up to meet it. Big errors you can see by eye between the two points. Smaller errors you put a thin steel rule between the two points. If correct, the rule should sit straight up and down and straight across.

 

[Hopper]

OK, I'll show my ignorance. What are lathe springs?

Colchester uses a circle of springs to set preload on the tapered roller headstock bearings. This allows for take up for wear and also thermal expansion of the spindle etc.

 

[deeferdog]

Thank you Hopper. I suppose everyone else knew that!

 

[john_reese]

I am not familiar with Colchester so regard my remarks accordingly. If it has jacking screws to adjust the headstock you are not moving the headstock enough. You need to know, at least approximately, the distance from the headstock pivot point. The amount to move the headstock, measured at the jacking point, is the error on the radius x the distance between the pivot and jacking point divided by the distance over which error was measures. Another way to look at it: consider the taper as an angle. The angular movement of the headstock has to be equal to 1/2 the angle of the taper.

 

[mcostello]

All I know for sure is if I move the 1' long bar in the headstock .022" the drill will not start on the center of the part and will make a circle. The tailstock marks are aligned on the center mark. If I turn a bar between centers then I can dial out the taper by the usual methods, or make it much less, it's been a while since I have done it so don't have exact numbers on that process. Also it faces stock flat with a scale across the face. Have not checked with feeler gauges but no gross error there.

 

[john_reese]

If you shift the headstock the tailstock will be out of line with it. You need to shift it as well.
Start by aligning the headstock to get rid of taper. Next, turn a piece between centers and adjust the tailstock to eliminate taper. I just got my Nardini running and it is time for me to do the same tests to make sure it is properly aligned.

 

[mcostello]

I don't want to deviate from the lines scribed at the factory as I would think They basically are right. It seems like there is rubber mounted in there somewhere and things are flexing, moving a measured .022 is a heap. Everything I can put an indicator on seems to say things are solid. I have been trouble shooting this for 3-4 years and have not seem to found it yet. Wondering if Some one replaced the front bearing and race and did not get the race in squarely. Going to bed now will check in tomorrow about 2:00 pm EST.

 

[Hopper]

I don't want to deviate from the lines scribed at the factory as I would think They basically are right.

Not necessarily, especially after all these years. Disregard the scribed lines and set the tailstock by turning a test piece between centres AFTER you are good and finished with setting the headstock spindle parallel to the bed ways.

Meanwhile, you need to set two dial indicators on the headstock, one at each end, so you can tell how the headstock is moving when you adjust those jacking screws, ie pivoting vs moving over but remaining parallel to the bed. Just measuring what the end of a 1-foot bar is doing is not telling you the full story.

If you want to set your headstock spindle parallel to the bed it is essential that you first know for 100 per cent sure how and where that headstock is moving when you adjust it.

Once you know how the jacking screws are in reality affecting the headstock position, you can then make a plan to use that adjustment to bring the spindle back in line with the bed.

 

[TonyM]

Have you checked to see if the bearings have been fitted correctly in relation to the max runout marks.

 

[mu38&Bg#]

If you are measuring taper with an indicator, you aren't actually measuring taper. Measuring taper is turning a length of bar and measuring diameter along it's length to determine cylindricity.

Correcting this step will allow you to continue with setup.

 

[mcostello]

I am measuring the taper with a micrometer on the shaft. I have been only using 1 dial indicator though. I have pulled the spindle out but did not think to see if the bearing race was fully set square in it's bore. I can dial the tailstock in and get something like .001 difference in 12". Maybe should not have called it taper as it is only a difference. I will get on it soon as possible but a friend needs some work done.