Head Scratcher...

I am running an Okuma LS-N 2 turret lathe.

Now what I don't understand is that the turret on the right side (see picture below) has a taper of .015 mm and the turret on the left side has a taper of .055 mm no matter what length I cut at.

I took a piece of scrap to reset the tools because it was down for months waiting for parts.

On the left turret I prorgamed it to take off a .03 mm finish pass at 50 mm in length, it gave me a .055 taper.

"Can you program it out?" - That is what I hear when I mention it.

So I do and then look at the print of the part I have to make so I take the taper out of the program and go half of the distance (25 mm in length) only to have the same exact amount of taper .055 mm.

I put out a long brass bar and proceed to take off a finish pass of .05 mm at a length of 165 mm.

The same exact thing happened - a .055 taper (0 at the beginning and .055 at the end) and I used a dial indicator to check the taper so it wouldn't be a misreading of a micrometer.

What would be the cause of this?

Some thoughts :

(1) Headstock not aligned with the bed (crash) ... put a ground test bar or similar in the spindle and run an indicator attached to the turret face along the length of the bar ... should be 0. The headstock is normally bolted and maybe some pins, you can move it around to zero it out.

(2) Take a cut on the face and the OD ... are they square to each other ... if so, see above.

(3) Does it happen with both turrets? ... if not, you isolated it to one turret ... if yes, see above.

Just some thoughts ... hope it helps. Please let us know the results.

Bluechip
"Real World Machine Shop Software at www.KentechInc.com"

That is only a couple of thousandths on the diameter, so the real error is half of that.

Could be a bit of wear in the chuck jaws, aggravated by a bent piece of stock?

Could also be the cutting action of the insert, if it is heavy honed, it could push hard if it is slightly above center.

Which way is the taper, big on the outer end?

If you are getting the same amount of taper regardless of the length of the turn then the taper is probably not uniform - or in a straight line. I would guess that if you took several measurements close together you will find that the taper steps out rather quickly and it is straight the rest of the way. Loose thrust bearings in the X axis ballscrew would let the screw shift as the load picked up and stay there until a change of direction. Just checking lost motion in the X axis may find that.
I am not familiar with that lathe but from the picture your turrets look like the front and back turret on a cross slide. It also looks like both turrets cut right side up so that you have to run the spindle in reverse for the back turret. Loose X gibs might let the slide cock differently when cutting front or back to explain the difference in taper.
A lathe should cut straight. If you can't make it cut straight by leveling or alignment it should be scrapped. Programming out taper is always bad practice because it will probably be different the next time you run the job.

Bluechip - The turret in question is the one on the left (in the picture) and it was indicated in both X & Z to read 0.

HuFlungDung - When the previous person that ran the machine before I started is able to do repair work that an outside company would normally do, then you know the machine was crashed on a regular basis. The story says that the boss crashed HARD it the first week he had it over 20 years ago.

A bunch of parts were replaced because of the problem I described and after months of downtime waiting for parts, it still does the same thing.

The test cuts were with a finishing tool that was on center so the problem isn't there.

John Michael - I used a dial indicator (.01 mm) to check the taper and it wasn't a giant step causing the taper.

When they pulled out the jewel indicator (.001 mm) you could actually see it jump around while going across the bar.

After seeing that, I thought it was the ball screw causing that but I wanted to be sure before approaching the boss and saying that was the problem.

Programming out taper is always bad practice because it will probably be different the next time you run the job.

I know this, but it seems like it is their band-aid to everything.

My reply was "I could program it away on 1 part but who knows what the next part is going to do." (This was from past experience from the same machine following the same advice).

If you have access to a cylindrical square (or any other accurately ground and parallel bar), mount it in the chuck and using a tailstock if you have one run your indicator from both turrets to establish if the taper is
a) Gradually increasing which suggests misalignment or
b) If the size difference is immediate it suggests perhaps some damage to the slideway.

To prove out b could you take a series of cuts over the maximun length possible?
Mike

Bluechip - The turret in question is the one on the left (in the picture) and it was indicated in both X & Z to read 0.

... Sooo the top turret cuts straight? If not, I would suggest the previous posts options ... I might think that the spindle is still out of alignment.

Some other thoughts :

(1) Mount the ground or precision bar in the chuck ... run the lower turret back and forth along the Z with an indicator mounted on the turret and touching the bar. Does it change as you move along the length of the bar? Does it move differently in both directions?

If so, the gib might be loose and the turret might be moving around on the way due to the loose gib.

You might check the taper at different areas of a long bar. If it doesn't happen at say 8" from the chuck face ... but does happen at 4" from the chuck face ... the ways might be worn. This is a common problem with older machines as most of the machining over the years occurs with tons of small parts beng machined close to the chuck and not much longer bar work.

Hope this helps ...
Bluechip
"Real World Machine Shop Software = www.KentechInc.com"

Longhair
I don't understand your use of a dial indicator to check taper. If you took a turning cut with the Z axis then ran an indicator down the part in place of the cutting tool by moving the Z axis it should say zero. My question was - do you have half the taper in half the length of the cut? If the headstock was out of line or the lathe not level you would. You originally said the taper or difference in diameters was the same regardless of the length of cut. That would seem to say that something moves or shifts after the start of the cut then stays there. If the cut is 2 inches long measure the diameter with a micrometer every quarter of inch of length and plot the diameters on a graph. Spindle out of alignment or lathe out of level taper will be a straignt line.
Have you checked the X axis for lost motion? Do you have any scars on the bed ways? A dent in the way from an impact will raise a bulge on all sides of it unless it has been stoned off.

I`ll go for play in the Z axis slideway or worn X axis ballscrew/bearings.I`m basing this on the fact that you are measuring the taper with a clock gauge from the position you turned the job in.As other people have mentioned clocking the job of the slide you turned it with should show zero taper even when it is tapered as you are following the tool path.The fact it shows a taper is probably because either the job is moving away from the tool or the tool is moving away from the job under cutting load which you don`t have with a clock.
Mark.

Hi Mark
I never heard a dial indicator called a clock gauge before. Is that common over their, or have I just been asleep?
Regards, John

John,clock gauge over here,unless you have just bought your first Myford and been reading the books.We "clock up" couplings,jobs etc and dial the telephone.
regards,Mark.

Here is an update on the situation...

The repair guy was in yesterday and checked the spindle to find out it was not square with the cross slide.

After a while he couldn't get it to cut straight still and decided to take an indicator to the ways only to find out that one had a .015mm taper and the other one had a .07mm taper right by the chuck.

After most of the day, he finally got it to have a .001 mm taper on 1 turret and .003 mm taper on the other turret while taking a .005 mm straight cut about 160 mm long.

That was the best it was going to get and I thought it was fixed.

Today I spent a couple of hours making sure all the tools were on center, then setting them off & turning down some piece of scrap material so I don't make any scrap when I started running parts again.

All was good until about the 4th part and then started the taper which was taking up the whole tolerance on the ID (the ID turning is done by the turret on the left in the above picture) which made it smaller in the back of the part.

I figured out it was the finish tool giving me the taper, so I did some changes and made a few test passes - 1 rough pass then finish, measure, change program, repeat - until I got it to have a taper of .005 mm along the length of the part (I had to put a taper in the program of .03 mm).

The 2nd part after that had a taper of .05 mm and that was with the taper in the program.

I get the supervisor, explain the situation and then he changes the taper in the program to compensate for the taper in the part - there is a .08 mm taper in the program when the first 4 had none.

He runs a part and there is a .01 mm taper in it so he changes the program again even though I said that we should run another one without changing the program to see what it does next.

Needless to say he told me to run a couple like that and see what happens.

It was time for me to leave so I turned the machine off and I will resume with my fabulous disaster tomorrow.

...

Would the machine be able to be fixed properly if it was sent out and rebuilt or is it scrap metal now?

I figure if they can rebuild it they should because they don't make machines like that anymore - even though it is 25 years old now, it will probably have out lasted the new machines on the market today.

ok check this out...

If the taper is changing, there is one thing you might want to check...

inside our older 1985 hitachi sieki lathe, there are these 2 pieces of metal that slide along eachother...

These are for taking up any slack in the machines movement... (sorry i dont know what they are called)

You need to find the special bolt on each side...

Tighten them slightly to take up the slack... but not too much, because that will make the machine bind...

Someoen will hopefully post, and give you the correct terminology....

We have dont this on our older cnc lathe and it helps.... We were having the exact same problem with taper that was always changing...

Good luck to you

They are called gibs Solar, just so you know.

Dave