Lathe setup question(s)

All discussion about lathes including but not limited to: South Bend, Hardinge, Logan, Monarch, Clausing and other HSM lathes, including imports

Moderators: GlennW, Harold_V

curtis cutter
Posts: 559
Joined: Fri May 18, 2012 11:46 pm
Location: Curtis, WA

Lathe setup question(s)

Post by curtis cutter »

First off, my old Enco lathe isn't much but at this moment it gets what I need done. Specifically, that is how to understand the operation of a lathe, tooling and materials. Not proud but it works. It is a good machine to learn on I believe.

My first question is, when I put a piece of 1/2" drill rod in my chuck and gently tighten it I can see movement in the rod at about 12" out from the chuck. I assume it is due to wear withing the jaws of the chuck, no surprise there. Am I most likely correct in this assumption?

With the above in mind, I had been considering the process of grinding the chuck to regain a square surface. This brought me to my next question which is, if the chuck is not exactly square to the bed, I am simply expanding my problems?

Finally, am I correct that before ANYTHING is done with a chuck in the line of grinding, should I be certain that my lathe is set up correctly and can that be done with an old chuck that may or may not be accurate due to wear?

Should I just get a new chuck and then see if the lathe is set up correctly?
Gregg
Just let go of it, it will eventually unplug itself.
User avatar
NP317
Posts: 4557
Joined: Tue Jun 24, 2014 2:57 pm
Location: Northern Oregon, USA

Re: Lathe setup question(s)

Post by NP317 »

Adjusting a lathe set up is perhaps best done with a known ground rod mounted between centers.
This will tell you if the head and tailstocks are aligned to the bed. Only then would I start looking at the chuck alignment.
This helps eliminate basic variables first.
~RN
John Hasler
Posts: 1852
Joined: Tue Dec 06, 2016 4:05 pm
Location: Elmwood, Wisconsin

Re: Lathe setup question(s)

Post by John Hasler »

RN writes:
Adjusting a lathe set up is perhaps best done with a known ground rod mounted between centers.
This will tell you if the head and tailstocks are aligned to the bed.

That will tell you that a line from the tip of the center in the spindle to the tip of the center in the tailstock is parallel to the ways. It does not tell you that the axis of rotation of the spindle is parallel to the ways.
User avatar
tornitore45
Posts: 2077
Joined: Tue Apr 18, 2006 12:24 am
Location: USA Texas, Austin

Re: Lathe setup question(s)

Post by tornitore45 »

1/2" drill rod 12" out will move appreciably on a dial gauge with moderate finger pressure.
Now if you centered it near the jaws and then travel with the carriage and the needle move either the stock is not straight, the jaws are not parallel, the head-stock is not "looking" at the Tail-stock or all of the above.
Spin the chuck an see if the deviation is symmetrical about zero or not.

Adding to what John said, a ground rod held in a good collet will tell is the head and spindle are parallel to the ways.

If you are a beginner, my suggestion is: start making the parts you need, perhaps starting with the easy one and try to understand any error, the cause and what the correction might be. Refrain from trying to adjust anything until you have some good feeling and experience. Do not ask how I know.
Mauro Gaetano
in Austin TX
pete
Posts: 2518
Joined: Tue Feb 10, 2009 6:04 am

Re: Lathe setup question(s)

Post by pete »

Your question can't be answered without getting into some of the complexity's of proper machine tool alignment. And some of this will be outside of what your asking about but it's still OT information.

First Google Dr. Georg Schlesinger and there's a few online copys of his book Testing Machine Tools in PDF form. Print that off and spend some quality time reading through and more importantly fully UNDERSTANDING what it's trying to explain and how each test is done and exactly why it's done the way it is. His research was started in the early part of the 1900's. Even today the finest and most accurate cnc machine tools will have there basic factory alignments and test certificate results based in some way on his work.

To mostly answer your question visualze your lathes headstock and bed ways. The bed and it's way and carriage surfaces IF it's been correctly leveled so there's no twist in them help guide that carriage along a straight line path both horizontaly and verticaly from the tailstock to the headstock. To simplify things we'll ignore any wear issues you might have with your lathe. IF and until you test, verify and know for sure let's say your headstock was very slightly out of alignment and it's headstock pointed very very slightly towards you or away from you, then regrinding the chuck jaws using the carriage to support and carry whatever grinding spindle you use will regrind the chuck jaw surfaces parallel to those way surfaces. If the headstock, spindle face, chuck back plate, the chucks inner back plate mounting surface isn't dead true and square to the way surfaces then the jaws will still be tapered and won't correctly and firmly hold a work piece. There is a bit of clearance even in new chucks between the jaws and the chucks guide surfaces to help keep the jaws mostly square to the chucks face. That does help a bit to allow the jaws to slightly tilt and compensate for slight misalignment issues.

But to do things properly you need to back up to the point of testing and if needed correcting any of the lathes basic alignments first. That starts with whatever floor type it's supported by, how rigid and movement free it's stand or bench it's sitting on is. Then move on to getting the bed in that twist free condition. That takes a good and accurate machinst level. Only then can you start on those Schlesinger tests to then verify everything else. But ignoring that bed leveling and machine testing and just moving the headstock without doing the initial basics is a very critical mistake since moving the headstock is the very last thing you do and only after proper tests do verify 100% it is in fact out. It's more than easy to make things even worse if you don't fully understand what your doing.

There's a few different way's of testing the headstock, most involve using what's called a two collar test bar that's made on the lathe your testing. But using small diameter stock and trying to do that test cheaply with even a slightly dull tool simply won't work. And you'd better understand exactly why and how this tests the headstock alignment to the ways before doing it and what the test results show. Without the basic bed leveling then doing anything else is pointless and will not properly solve any of the problems that might or might not be there. If and when everything checks out then yes regrinding your jaws will then remove any wear and put them back to being square to the chucks jaw guide surfaces, the chucks face and in line with the machines spindle once everything else is checked as being correct. But regrinding the jaws still isn't quite as simple as most think. Don the owner of Suburban Tools did an excellent Youtube video a few years ago about how to properly preload the chuck jaws in the correct orientaion while regrinding the jaws. I'd highly recommend watching that video at least a few times. There's at least a few other ways of doing it, but in my opinion his method is about the best I've seen.

While it's quite expensive at around $100 the Connelly book Machine Tool Reconditioning was the best $100 I've ever spent on anything for my shop. It's a tough book to get through and is also about rebuilding machine tools for the most part. But it further expands on Schlesinger's work and details the 3 dimensional machine tool alignments that are very tough to fully understand and properly visualize without it. It's also saved me far more than it's cost just in the way it's changed my previous lubrication and way surface cleanliness habits. I've now spent more than 30 years studying everything I could find about machine tool alignments. I also now think I actualy know very little about the subject since it's a whole lot more complex than most at the hobby level seem to understand. But with the little I think I now know I've also managed to impress myself more than a few times just by using and following some of the information Schlesinger and Connelly provided. Adjusting and aligning any machine tool to it's very best accuracy is imo a great deal more important at the HSM level than the machinist magazines, forums, Youtube etc seem to indicate. In industry they mostly use very heavy duty and massively rigid machines. Any shop that knows what there doing will spend a few days themselves or have a factory tech level and adjust any new machine tool so the alignments are correct to the factory test limits. Most entry level hobby guys will bolt a light weight flimsy machine down just about anywhere and put it to work with zero checks. There are some major misalignments or wear problems that would require a great deal of highly skilled work to correct. Without proper testing to find what misalignments or wear that might be there then it's impossible to possibly adjust and correct or ever hope to compensate for them in any way. Scraping and correcting major components back into there correct alignments is way outside this threads topic. Lot's of good Youtube videos about it if your interested though.

Also, drill rod is rarely round nor is it ever exactly straight unless you got very very lucky. The general recomendation while machining is to never have more than 3 times the length of the stock diameter outside the chuck jaws. Testing with light indicator pressures would allow a bit more. But a 1/2" rod sticking 12" outside the chuck jaws even if it was exactly round and straight will still move a great deal just due to that gravity problem. One jaw on your used chuck worn by just .0001" would show a great deal of movement 12" away from the chuck face. When I first started learning some of this I spent hrs fooling with my lathes tailstock height thinking it was low. Gravity effect on a short 3" long 3/8ths diameter heat treat rod holding a light weight dti was bending that rod by almost .007". That same indicator on your 12" rod would bend it at least 030"-050" or more.

Far too many of us Hobbist's get hung up about finding a good 3 jaw with very low runouts. If the work is fully machined and parted off and never removed from the chuck until it's finished then the work piece will be as concentric and true as your machines condition, alignments and it's headstock bearings allow. Doing exactly the same with a brand new chuck will not change anything or make a better and more concentric part. So for now your current chuck even with worn jaws should allow all of the testing to be done, then maybe regrind it's jaws if needed, and only then would you have to make a decision about buying a new chuck. I hope some of this helped.
curtis cutter
Posts: 559
Joined: Fri May 18, 2012 11:46 pm
Location: Curtis, WA

Re: Lathe setup question(s)

Post by curtis cutter »

pete wrote: Sat Sep 15, 2018 7:58 pm Your question can't be answered without getting into some of the complexity's of proper machine tool alignment. And some of this will be outside of what your asking about but it's still OT information.

First Google Dr. Georg Schlesinger and there's a few online copys of his book Testing Machine Tools in PDF form. Print that off and spend some quality time reading through and more importantly fully UNDERSTANDING what it's trying to explain and how each test is done and exactly why it's done the way it is. His research was started in the early part of the 1900's. Even today the finest and most accurate cnc machine tools will have there basic factory alignments and test certificate results based in some way on his work.

To mostly answer your question visualze your lathes headstock and bed ways. The bed and it's way and carriage surfaces IF it's been correctly leveled so there's no twist in them help guide that carriage along a straight line path both horizontaly and verticaly from the tailstock to the headstock. To simplify things we'll ignore any wear issues you might have with your lathe. IF and until you test, verify and know for sure let's say your headstock was very slightly out of alignment and it's headstock pointed very very slightly towards you or away from you, then regrinding the chuck jaws using the carriage to support and carry whatever grinding spindle you use will regrind the chuck jaw surfaces parallel to those way surfaces. If the headstock, spindle face, chuck back plate, the chucks inner back plate mounting surface isn't dead true and square to the way surfaces then the jaws will still be tapered and won't correctly and firmly hold a work piece. There is a bit of clearance even in new chucks between the jaws and the chucks guide surfaces to help keep the jaws mostly square to the chucks face. That does help a bit to allow the jaws to slightly tilt and compensate for slight misalignment issues.

But to do things properly you need to back up to the point of testing and if needed correcting any of the lathes basic alignments first. That starts with whatever floor type it's supported by, how rigid and movement free it's stand or bench it's sitting on is. Then move on to getting the bed in that twist free condition. That takes a good and accurate machinst level. Only then can you start on those Schlesinger tests to then verify everything else. But ignoring that bed leveling and machine testing and just moving the headstock without doing the initial basics is a very critical mistake since moving the headstock is the very last thing you do and only after proper tests do verify 100% it is in fact out. It's more than easy to make things even worse if you don't fully understand what your doing.

There's a few different way's of testing the headstock, most involve using what's called a two collar test bar that's made on the lathe your testing. But using small diameter stock and trying to do that test cheaply with even a slightly dull tool simply won't work. And you'd better understand exactly why and how this tests the headstock alignment to the ways before doing it and what the test results show. Without the basic bed leveling then doing anything else is pointless and will not properly solve any of the problems that might or might not be there. If and when everything checks out then yes regrinding your jaws will then remove any wear and put them back to being square to the chucks jaw guide surfaces, the chucks face and in line with the machines spindle once everything else is checked as being correct. But regrinding the jaws still isn't quite as simple as most think. Don the owner of Suburban Tools did an excellent Youtube video a few years ago about how to properly preload the chuck jaws in the correct orientaion while regrinding the jaws. I'd highly recommend watching that video at least a few times. There's at least a few other ways of doing it, but in my opinion his method is about the best I've seen.

While it's quite expensive at around $100 the Connelly book Machine Tool Reconditioning was the best $100 I've ever spent on anything for my shop. It's a tough book to get through and is also about rebuilding machine tools for the most part. But it further expands on Schlesinger's work and details the 3 dimensional machine tool alignments that are very tough to fully understand and properly visualize without it. It's also saved me far more than it's cost just in the way it's changed my previous lubrication and way surface cleanliness habits. I've now spent more than 30 years studying everything I could find about machine tool alignments. I also now think I actualy know very little about the subject since it's a whole lot more complex than most at the hobby level seem to understand. But with the little I think I now know I've also managed to impress myself more than a few times just by using and following some of the information Schlesinger and Connelly provided. Adjusting and aligning any machine tool to it's very best accuracy is imo a great deal more important at the HSM level than the machinist magazines, forums, Youtube etc seem to indicate. In industry they mostly use very heavy duty and massively rigid machines. Any shop that knows what there doing will spend a few days themselves or have a factory tech level and adjust any new machine tool so the alignments are correct to the factory test limits. Most entry level hobby guys will bolt a light weight flimsy machine down just about anywhere and put it to work with zero checks. There are some major misalignments or wear problems that would require a great deal of highly skilled work to correct. Without proper testing to find what misalignments or wear that might be there then it's impossible to possibly adjust and correct or ever hope to compensate for them in any way. Scraping and correcting major components back into there correct alignments is way outside this threads topic. Lot's of good Youtube videos about it if your interested though.

Also, drill rod is rarely round nor is it ever exactly straight unless you got very very lucky. The general recomendation while machining is to never have more than 3 times the length of the stock diameter outside the chuck jaws. Testing with light indicator pressures would allow a bit more. But a 1/2" rod sticking 12" outside the chuck jaws even if it was exactly round and straight will still move a great deal just due to that gravity problem. One jaw on your used chuck worn by just .0001" would show a great deal of movement 12" away from the chuck face. When I first started learning some of this I spent hrs fooling with my lathes tailstock height thinking it was low. Gravity effect on a short 3" long 3/8ths diameter heat treat rod holding a light weight dti was bending that rod by almost .007". That same indicator on your 12" rod would bend it at least 030"-050" or more.

Far too many of us Hobbist's get hung up about finding a good 3 jaw with very low runouts. If the work is fully machined and parted off and never removed from the chuck until it's finished then the work piece will be as concentric and true as your machines condition, alignments and it's headstock bearings allow. Doing exactly the same with a brand new chuck will not change anything or make a better and more concentric part. So for now your current chuck even with worn jaws should allow all of the testing to be done, then maybe regrind it's jaws if needed, and only then would you have to make a decision about buying a new chuck. I hope some of this helped.
Thank you for this extensive reply Pete, I appreciate it.

I know there is much to learn and that is why I come here to read posts, even those that are far above my level of comprehension. Most of what you have said above is to the level of my understanding and I appreciate that. I have come to realize that there is never a simple answer to even what I would assume to be a simple question. I will do the reading you prescribe and I will undoubtedly have a better understand of the variables I will encounter.

The last thing I want to do at this point is remove metal or change a thing without knowing what I am doing... I have turned stock and measured the diameter at each end of a 1" by 24" rod and find I am only a couple thousandths off in diameter from one end to the other so I assume I am close. At least close enough for me at this stage of learning.

Again, thanks.

Gregg
Gregg
Just let go of it, it will eventually unplug itself.
pete
Posts: 2518
Joined: Tue Feb 10, 2009 6:04 am

Re: Lathe setup question(s)

Post by pete »

I'm happy some of that helped. I think you've got the right idea. Just proceed cautiously and don't try for more than your understanding and current knowledge will allow. If you have any doubts at all then ask questions. If you don't know then you don't know, there's no such thing as a stupid question. I still remember all to well how tough it was to gather up or fully understand what little information I could find. That was before the internet was thought of so even finding out what I needed to know and where that information might be was an effort. I've also seen more than a few posts (not on this forum) and Youtube videos about people highly recommending lapping way surfaces etc with valve grinding paste. Yes it makes the machine seem smoother to operate and the gibs might be able to be adjusted slightly tighter. It can't and doesn't correct any of the alignments or localized wear since there's zero control of where and how much metal gets removed. It's much more than likely you'd make things worse. If the idea worked at all then industry would use it before the much more costly and time consuming surface grinding and scraping had to be done. I've also seen almost knee jerk recommendations about moving the headstock first before even mentioning doing the proper tests to be sure it's required. Well meaning or not I tend to ignore any futher advise from those people about any other topic.

Unfortunately there's no fast or easy way of learning as much as possible about machining or anything related to it. The internet, these forums and Youtube as good as they are don't in my opinion replace the need for good technical reference books. All help and work togeather to learn and expand your knowledge. Again it's not exactly cheap at about $50, but I can't recommend it highly enough. George H. Thomas's book The Model Engineers Workshop Manual. Far cheaper to order it direct from Tee Publishing in the U.K. than what the Ebay bandits want for it. That one book has taught me more about good machining practices and why you'd chose one method over another than everything else I have combined. There's another much cheaper book also available from Tee Publishing titled Work Holding in the Lathe that will also teach a great deal of the basics. It's another topic that seems simple but isn't.

A few thou taper over 24" is real close. But a 1" diameter bar over that length can easily deflect more than enough to give false results and especialy so the further you get from either the head or tail stock. Even spending a great deal of time getting my lathe as level and twist free as possible still showed some minor variations. Lathe leveling to be honest is still only a static test and cutting conditions showed some extremely small adjustments were still needed. That fact is almost never brought up when the subject of leveling is mentioned. Now I get my lathe quite close with a machinist level, then jump to the cutting tests, it's a lot quicker to get the exact same result that way. I should mention even the full series of Schlesinger tests are still only static tests. But they can and do show or at least indicate where there is or might be issues. There still a very good starting point to getting what you want. The one thing that took me years to learn was if you want accuracy then you never blindly trust anything unless you've personaly checked and verified each part of the machine is correct. I see many checking a chucks O.D. surface especialy on the U.K. forums. Other than getting a bit of vibration from a poorly centered chuck that surface means almost nothing and can mostly be ignored. How and where the chuck gets mounted to the spindle, the spindles internal taper, it's chuck locating surface, the back plates locating surface and size to the spindles locating surface, and the chucks face itself are all far more important areas to check.

If you do end up regrinding your chuck jaws? If you've got a semi decent chuck it's master pinion should be marked with a 0 mark. That's the pinion that was used to tighten and preload the jaws as they were ground concentric at the factory. It should also be used whenever regrinding the jaws. I'd also highly recommend pulling apart and spotlessly cleaning your chucks internals before regrinding the jaws. Then repeat that after the regrinding. Once they are reground then any time the chuck is taken apart for regular cleaning then keeping track of and replacing each pinion back it the same exact location each time is important to maintain the chucks jaw concentricity. It does make a difference and Emco made a point of mentioning that for a couple of the expensive chucks I bought from them.
User avatar
Harold_V
Posts: 20231
Joined: Fri Dec 20, 2002 11:02 pm
Location: Onalaska, WA USA

Re: Lathe setup question(s)

Post by Harold_V »

I read the description of the issue at hand and drew the conclusion that the point one is to understand is that the chuck may be allowing unwanted movement at the end of a long piece of material. Before drawing any conclusions, it's important to determine the source of the movement. It can be the result of a crash, whereby one of the jaws has sprung, often the result of a slight misalignment of the jaw slides, or even a slight bending of the jaw, assuming it's not heat treated properly. That's not likely to be the source, but it shouldn't be dismissed as one of the possibilities. Light duty chucks are also easily sprung by gripping at the outer tips only, then by over tightening. That, too, distorts the jaw/slide interface, causing the jaws to grip only at the very rear. leaving a gap between the jaw and piece at the outer edge. That condition often results in a three sided turn, which most likely would still mic round, although it's not round. You can determine if it's doing that by making a cut, then running an indicator on the machined surface. If it's round, it won't yield any movement of the indicator. If there is movement, you must then determine if the source is eccentricity, or out of round. The eccentricity will yield a constant pattern, with the high spot 180° opposed to the low spot. Out of round, the condition will very, usually three times, around the diameter of the piece. Hope that makes sense. Don't really know how to describe it.

Do keep in mind, this type of issue has nothing to do with spindle alignment, it's strictly a work holding problem. If that was not the case, the piece, when rotated, would not align with the tailstock, assuming it is in near alignment when the tailstock is brought near the headstock. When the spindle is in alignment, it will scribe a circle around the tailstock, but the centerline of the circle will be on tailstock center. So then, if the movement you describe matches the description, the issue at hand is one of two things. The material you're using isn't dead straight (that it is drill rod does NOT ensure straightness), all of the jaws are sprung, or one of the jaws is sprung. You must determine which of those conditions exists.

For your tests, I strongly advise against the use of drill rod. I do suggest the use of a drill blank, or a dowel pin, both of which will be centerless ground and should be dead straight. While drill rod is generally centerless ground, or at least centerless belt sanded, it often is not round, and it may or may not be straight.

Insert the pin and tighten the chuck enough for a good grip. Using a feeler gauge, thin as possible, see if you find any clearance at the outer edge of the gripping portion between the jaws and the pin/blank. If you find there is not full contact of the jaws with the pin/blank, the slide(s) are distorted. The chuck , at this point, will grip only at the very rear of the jaws, and not have the ability to move the pin so it's parallel to the spindle centerline. In that case, regrinding will improve the jaw alignment tremendously, but they will still not run dead true, although they will now hold the stock parallel, which will eliminate the condition you reported. Error in the scroll, and random selection of the hole in which you place the chuck wrench will generally yield a different closing of the jaws, to say nothing of the error that may exist in the scroll. Using the same hole and gripping the exact same size should provide consistent results, however. It pays to use the same socket each time if you deal with this level of precision.

If you are familiar with the procedure of preparing soft jaws, you'll know that you must properly restrain the jaws before grinding. If you are not familiar with that type of operation, I'd like to point you to the Resource Library, where there is a thread on the use of soft jaws. In it you will find what is pertinent, and a procedure to follow in preparing your chuck for grinding.

A reasonably rigid spindle should be used in grinding, so that any surface irregularity in the jaws will be removed, which is important if you hope to improve the condition of the jaws. A high speed grinder with sloppy bearings will have the ability to simply follow the existing surface, and may or may not improve the existing condition.

H
Wise people talk because they have something to say. Fools talk because they have to say something.
User avatar
NP317
Posts: 4557
Joined: Tue Jun 24, 2014 2:57 pm
Location: Northern Oregon, USA

Re: Lathe setup question(s)

Post by NP317 »

John Hasler wrote: Sat Sep 15, 2018 10:59 am RN writes:
Adjusting a lathe set up is perhaps best done with a known ground rod mounted between centers.
This will tell you if the head and tailstocks are aligned to the bed.

That will tell you that a line from the tip of the center in the spindle to the tip of the center in the tailstock is parallel to the ways. It does not tell you that the axis of rotation of the spindle is parallel to the ways.
Agreed. That would be the next test I would do.
RN
User avatar
Harold_V
Posts: 20231
Joined: Fri Dec 20, 2002 11:02 pm
Location: Onalaska, WA USA

Re: Lathe setup question(s)

Post by Harold_V »

Headstock/way alignment would be determined by chucking a long shaft, known to be straight, and dialing it true at the chuck and at the end. An indicator would then be traversed along the shaft, end to end, both on top or bottom, and on either side. That would reveal the true orientation of the headstock as it relates to way travel. Once ascertained, a shaft held between centers could then be indicated similarly to determine tailstock alignment. The tailstock alignment wouldn't guarantee a straight cut, however, although it should be close. The only truly reliable method of eliminating taper is to take identical cuts on each end of a long shaft. Even then, the slightest change of any feature, including how tightly the centers are set, will most likely make a difference in the outcome.

Do keep in mind---these machines (lathes) are just that---lathes. They are not precision grinders and most likely will have great difficulty achieving cuts within less than a thou.

H
Wise people talk because they have something to say. Fools talk because they have to say something.
John Hasler
Posts: 1852
Joined: Tue Dec 06, 2016 4:05 pm
Location: Elmwood, Wisconsin

Re: Lathe setup question(s)

Post by John Hasler »

Note that horizontal alignment error is much more important than vertical. If the far end of your bar is .001" closer to the toolpost than it is near the chuck then you are going to get a diameter .002" smaller out there. If the far end of the bar is .001" higher than it is up by the chuck then a tool that is on center near the chuck will be .001" below center at the far end. This will result in a diameter error on the order of .000002" on a 1" bar.
curtis cutter
Posts: 559
Joined: Fri May 18, 2012 11:46 pm
Location: Curtis, WA

Re: Lathe setup question(s)

Post by curtis cutter »

I am finding this to be a very good read on the subject:

http://totallyscrewedmachineshop.com/do ... ger%29.pdf

I suspect in this process, I am going to have to invest in a level. Does this seem like a reasonable level to educate myself with:

https://www.amazon.com/Master-Precision ... evel&psc=1
Gregg
Just let go of it, it will eventually unplug itself.
Post Reply