How can this be?

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

pete
Posts: 2518
Joined: Tue Feb 10, 2009 6:04 am

Re: How can this be?

Post by pete »

Ok but the numbers are important. That part is .002" out, but concave or convex? And over how much distance? Obviously it's too much, but just remember those numbers because of the way you did that test are DOUBLE the actual misalignment. So your out enough it needs correction, but probably not quite as much as it seems right now as long as it's still facing concave. Your previous parallel test cut without tail stock support seems to prove the head stock is in alignment with the bed ways. Even with that are you 100% dead sure all your head stock bolts are in fact fully tight and secure. Any of this testing is a process of elimination, you have to verify everything is double checked first before moving to the next. I'd also do an indicator check on your spindle nose for both radial and axial movement, radial checking, then stack up enough wood scraps on the lathe bed under the spindle nose and then GENTLY pry up and down with any length of wood you can find as a lever. Maybe 20-30 lbs of pressure lifting the spindle nose. Axial checking, put a zeroed indicator on the spindle nose and push and pull back and forth as hard as you can. Let us know what those numbers are. Unless those spindle bearings are known to be still correct first, any test cuts are suspect for there numbers and just about worthless.

On that facing cut I'd also want to use a 10ths indicator and the exact distance between zero and the outside of the part you traveled on the cross slide needs to be known. Because if corrections are required, those numbers are going to be very important. And in my opinion your going to need to pull the cross slide assembly apart, spotlessly solvent clean, re-lube and then reassemble with a proper gib adjustment. If it's adjustable? Then that's also the time to readjust the cross slide feed screw nut for backlash since your in there anyway. But that gib adjustment is a lot harder to do correctly than most think. It's a very fine line between too tight and too loose. Your doing at least very minute movements on the gib adjustment screws. The best way I've learned to do it is without the cross feed screw in place and sliding the cross slide by hand back and forth. It's then easy to feel each adjustments effect. Once you think it's correct, then put an indicator on the side and at each end of the cross slide and check it's side to side movement by pushing and pulling on it by hand. You won't get zero because the parts do have to move and allow an oil clearance gap. Maybe .002"-.003" on a cross slide in good condition I'd guess. With any wear your still going to have to adjust for a best average where it's not too tight at the ends of travel for the cross slide. Then repeat that facing test.

A misalignment on the cross slide to the head stock has no way of adjusting it out without at a minimum re-grinding the fixed dovetail to reset the alignment to the head stock and regrinding the female dovetail to be parallel. It can be done by hand scraping, but I don't recall you mentioning you've done any and this sure isn't the place to start learning. But that's why the exact numbers on your head stock facing test are important. The parts need to be set up and reground to 10ths. Using the recommended numbers for a tool room class of lathe in Schlesinger's book, your looking at around .00016" - a maximum of maybe .0002" per inch of length on the fixed dove tail so the lathe then faces concave by that amount. But as I mentioned, I'd perform that facing cut test on your lathe face plate simply because it's almost the full diameter the lathe can swing so the indicated numbers will then have the greatest accuracy level of how that cross slide is actually aligned and / or how much wear it now has.

All these tests are a time consuming royal pita, but as each area is checked and verified it can then be ignored. Then you just have to correct where the actual problem is.
whateg0
Posts: 1114
Joined: Sat Mar 28, 2009 3:54 pm
Location: Wichita, KS

Re: How can this be?

Post by whateg0 »

Assuming that to be a 6" part based on is relative size to what is assumed to be a 2" face on the DI, that means roughly 1 thou in 3", or 3 tenths per inch. So it doesn't meet the tool room lathe requirements but I wouldn't take my tool room lathe apart to fix that for the parts I do. Ymmv
LIALLEGHENY
Posts: 363
Joined: Sat Jan 11, 2014 12:36 am
Location: Bohemia, NY

Re: How can this be?

Post by LIALLEGHENY »

My first question is what your using for a level? Only a good machinist level is going to be accurate enough to determine twist..... a thousands twist measured at the ways will be compounded the further your centerline is from the ways. The larger the swing of a lathe , the more critical leveling is.
One thing to test, put an indicator on your test bar at the tailstock end and make sure it turns true,, then remove your live center from the tailstock and reinsert it at 90 , 180, 270 degrees, or any position thereof....does it still turn true ? A little dirt, a chip, a burr on the taper of the center can throw everything off. I have an assortment of centers for my little Logan lathe......just changing the center or rotating it can change my tailstock offset. Another thing that changes the offset is how tight I clamp the tailstock quill.

Nyle
RSG
Posts: 1541
Joined: Fri Feb 04, 2011 9:59 am
Location: Ontario, Canada

Re: How can this be?

Post by RSG »

pete wrote: Sun Nov 14, 2021 5:51 pm That part is .002" out, but concave or convex? And over how much distance?
I'll look at that today and report back. For now unless the top slide comes off from the front I won't be able to pull it as my lathe is in a spot that I can't completely slide it off the back like I did once when I first bought it. Not sure it can be taken off by the front without a pile more work. I'll look today.
Vision is not seeing things as they are, but as they will be.
RSG
Posts: 1541
Joined: Fri Feb 04, 2011 9:59 am
Location: Ontario, Canada

Re: How can this be?

Post by RSG »

LIALLEGHENY wrote: Sun Nov 14, 2021 9:34 pm My first question is what your using for a level? Only a good machinist level is going to be accurate enough to determine twist..... a thousands twist measured at the ways will be compounded the further your centerline is from the ways.
I have a machinists level and monitor the bed every 6 months or so. I have it set based on the carriage being close to the chuck were most of the work happens.
LIALLEGHENY wrote: Sun Nov 14, 2021 9:34 pm One thing to test, put an indicator on your test bar at the tailstock end and make sure it turns true,, then remove your live center from the tailstock and reinsert it at 90 , 180, 270 degrees, or any position thereof....does it still turn true ? A little dirt, a chip, a burr on the taper of the center can throw everything off.
I backed it out and extended the quill a half inch and did a test cut and the numbers came it very close, but I did not pull the live center completely, I can try that. One thing I did do was to clean up the centers on the test bar before performing the tests.
Vision is not seeing things as they are, but as they will be.
RSG
Posts: 1541
Joined: Fri Feb 04, 2011 9:59 am
Location: Ontario, Canada

Re: How can this be?

Post by RSG »

pete wrote: Sun Nov 14, 2021 5:51 pm I'd also do an indicator check on your spindle nose for both radial and axial movement, radial checking, then stack up enough wood scraps on the lathe bed under the spindle nose and then GENTLY pry up and down with any length of wood you can find as a lever. Maybe 20-30 lbs of pressure lifting the spindle nose. Axial checking, put a zeroed indicator on the spindle nose and push and pull back and forth as hard as you can. Let us know what those numbers are.
Hey Pete,

I took the chuck off and used an aluminium flat bar to gently try to lift the spindle as you requested with my 10ths DTI on the top and it never budged! Same with the face. The spindle bearings are still loaded properly as far as I can tell. I then put the DTI inside the spindle lip and rotated the spindle and the dial never moved so it's nice and concentric as well.

I'll be performing journal tests again shortly before I ramp up for winter reel building as well now that the tailstock has been corrected. I'll let you know the outcome there too.
Vision is not seeing things as they are, but as they will be.
pete
Posts: 2518
Joined: Tue Feb 10, 2009 6:04 am

Re: How can this be?

Post by pete »

Ok, that's at least one item you can cross off the list as being ok Ron. And with zero movement that's really good, I would have expected to see at least a few 10ths. Just to be 100% sure, I'd repeat that same test after you've run the lathe for a few hours when you know it's then up to it's average operating temperature.

I also agree with LIALLEGHENY and it's possible your live centers Morse Taper shank being not concentric to it's live center point. However you can easily check that by center drilling a piece of scrap in the head stock, install a dead center in the tail stock, then set your live center between those with it's point in the head stocks female center, spin your live center by hand while indicating the Morse Taper shank at both the front of the shank just behind the bearings and at the rear just before that MT shank ends. That will tell you for sure if there's a concentricity issue between it's MT shank and the actual live center point. If it turns out there is? Then I'd mark the shank just behind where the bearings are on the MT shank at it's high spot and always install it with the mark facing directly up in the tail stock for the easiest way to get the same consistency each time it's used. As with anything else, if you can measure and identify where any inaccuracy's are, most times there's a way to compensate or make counter corrections for them.
Last edited by pete on Wed Nov 17, 2021 12:55 pm, edited 1 time in total.
RSG
Posts: 1541
Joined: Fri Feb 04, 2011 9:59 am
Location: Ontario, Canada

Re: How can this be?

Post by RSG »

pete wrote: Tue Nov 16, 2021 4:23 pm I'd repeat that same test after you've run the lathe for a few hours when you know it's then up to it's average operating temperature.
Good idea! I will....
pete wrote: Tue Nov 16, 2021 4:23 pm However you can easily check that by center drilling a piece of scrap in the head stock, install a dead center in the tail stock, then set your live center between those with it's point in the head stocks female center, spin your live center by hand while indicating the Morse Taper shank at both the front of the shank just behind the bearings and at the rear just before that MT shank ends.
And thanks for that! I will definitely do that today.....
Vision is not seeing things as they are, but as they will be.
Post Reply