[4]

Flat way repair

[3]
[10] Like what you see?
Click here to donate to this forum and upgrade your account!

jwmay

Active Member
Registered
Joined
Dec 3, 2017
Messages
405
This is a simple question that probably has a complicated answer. While examining a lathe bed that I'd earlier relegated to become scrap, I decided there may be some things I could learn from it. Also, I really like Atlas lathes, and I have the remnants of one. So, I brought the bed in and started looking it over. What I found was that the bed ways, near the headstock have been deformed. What I mean to say is that the ways have a lot of chipping and scarring on the 90 degree edges, and most of it resulted in raised areas on the inside of the ways. I started stoning them, after I was sure about what I was looking at. But then thought that if I continued, I may end up stoning it out of square. Most of the deformation is on the top inside, not the bottom inside. Then I had the idea that maybe I could just file a bevel, effectively removing the chipping, and most of the raised areas, and that would be ok, seeing as how I don't believe the inside vertical surface is a bearing point for the saddle on the Atlas lathe. Then I remembered that I don't actually have a clue what I'm doing, and probably better make mention of it here, before I do real damage, or wipe out a reference point that I'll be needing later. So what say ye?
 

NortonDommi

Active Member
Registered
Joined
Nov 15, 2016
Messages
467
As has been raised is discussion elsewhere this is an area of the ways that can be removed in a gap-bed lathe. Something has probably come loose at some time in its life and a loud ker-thunk or many little ones has occurred.
Sometimes on a non-gap lathe this area will be relived to enable a job to get done.
Wind the apron to the headstock and see if it will hit the affected area. If not file a nice bevel it will look nice and allow a fraction larger diameter to be turned on a faceplate. If the apron will run over it file the worst dings down and scrape to the same as the rest of the way.
If you can post some pics that would be great and you will get a lot of good advice especially from fellow owners.
 

Bob Korves

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Jul 2, 2014
Messages
6,626
Do not remove metal to make it look good, assuming you want to use the lathe rather than merely making it pretty. Remove metal only to take away the raised high spots. Leave the low spots alone for now. When I say high, I mean high relatively to the undamaged portions of the surface. Low spots (dings) cause no problems with the function of the lathe, they just don't look good. Portions that have been worn swayback from lots of use, or from abuse by not oiling it, or leaving a mix of oil and grit on the sliding surfaces, are just wear. Wear can be lived with, to more of an extent than many of us believe. The only real cure to wear is to move all the surfaces down to match the lowest surface, making them all parallel, but lower, to the original surfaces. That is a MAJOR project, or a trip to a shop that grinds lathes back to the original geometry ($$$), or deciding it is not worth it, and selling or scrapping the lathe bed. It works the same with all the other components. There is a real learning curve to being able to do this work yourself, without wasting your time and/or just making it worse. The first step is to quantify the existing wear and any other problems. You did the correct thing by stopping the stoning (first, cause no harm.) Get help and information before proceeding. This forum can and will help you with questions and learning. You will need to learn how to separate out the good advice from the lesser...
 

markba633csi

Gold
Registered
Joined
Apr 30, 2015
Messages
3,832
The top inside way corners (and the outside corners also) on a flat way machine don't really need to be knife edge sharp.
So yes, you could smooth the inside top corners of the ways there and it wouldn't affect the machine operation. My 12x36 Atlas has a few inside corner dings too, was thinking of doing the same thing to it.
Mark
 
Last edited:

jwmay

Active Member
Registered
Joined
Dec 3, 2017
Messages
405
Here’s some pictures. If these don’t seem to be informative, I’ll be glad to take different ones. To be clear, I never expected this bed to be functional again, and most of the parts to this machine are gone. I don’t think I’d ever be able to source the parts at what I sold them for. But the experience of trying to make it a whole and functional machine again scratches an itch I have trouble controlling. Thanks very much for all the advice so far! I’m loooking forward to more!

5E4CF383-94EE-48F9-AEEB-FDC2811B3875.jpeg

2F7F860E-A9D7-4861-A0B6-0F1A9125047C.jpeg

F8994C58-A179-462A-A260-EABBF05F1D79.jpeg

F317045D-411D-480A-B528-17AABC476631.jpeg
 

Bob Korves

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Jul 2, 2014
Messages
6,626
That is pretty rough, but not a deal breaker. The bigger issue is that you don't have the parts to put it back together. I would simply buy a complete project lathe in rebuildable condition, or one ready to use. That is no doubt less expensive and way less hassle than buying all the missing pieces to complete a lathe using the bed you have. Something to be proud of when it is finished. Still, your machine, you have to make the choices...
 

jwmay

Active Member
Registered
Joined
Dec 3, 2017
Messages
405
Well...maybe I could make a wood turning attachment for it. Then all I’d need is a tailstock. Never had much interest in wood turning, but I do have lots of files with no handles.
 

Dabbler

Administrator
Staff member
H-M Supporter - Gold Member ($25)
Joined
Oct 11, 2016
Messages
935
This is a lathe that can be returned to use, if you need a lathe. It would make an OK beginner lathe for someone who needs a lathe to start somewhere.
 

jwmay

Active Member
Registered
Joined
Dec 3, 2017
Messages
405
That’s great! That’s exactly what it was. I messed around with it for a few months, trying to find parts, fix things, etc. I used to tell my wife that I needed a lathe to fix this lathe. Lol.

Anyhow if you find that new guy that wants it, he can have it, so long as he promises the next owner will be a scrapyard. I’m not selling any more pieces to resellers with a wire wheel, a rattle can, and an EBay account. Haha
 

jwmay

Active Member
Registered
Joined
Dec 3, 2017
Messages
405
In all seriousness though, assuming a person was going to undertake this effort, what would you all suggest?
 

jwmay

Active Member
Registered
Joined
Dec 3, 2017
Messages
405
So I found this sort of rig on YouTube, and Gabe it a try. I’m making a lot of assumptions here in considering these numbers. I don’t have a tenths indicator because I dropped and broke it. Anyways I’m assuming the bed is as level as possible, and the thousandths indicator would be sufficient. After all, this bed is trashed right? I was thinking I’d get .008-.015 readings. Well assuming again that my setups are rigid, the worst wear I could find was .004” drop from the headstock area, and about 12” forward of that. This isn’t the whole bed ways area, but just the visible wear from the saddle. Parallelism I’m having trouble quantifying...like say from back of rear way to front of front way. But that doesn’t seem insurmountable. Once again, I don’t know what I’m talking about, so that statement is probably not useful.

2B2CDB12-7527-48ED-A608-5D60E822CD73.jpeg

6F491C2C-2915-4C9C-91E5-DD34F72A028D.jpeg
 

Bob Korves

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Jul 2, 2014
Messages
6,626
Your test setup is not ideal, it follows the low spots, tending to average them out. Still, that is not too bad of a reading. The real question is, what are you going to do with it? If you have to buy major parts to complete the lathe, it will be cheaper and much easier to just buy a lathe in decent condition.
 

jwmay

Active Member
Registered
Joined
Dec 3, 2017
Messages
405
Well I thought there must be something wrong with my test set up. Maybe setting up and carefully leveling the bed to a surface plate, and then sliding the indicator base on the granite with the indicator tip on the ways?

What will I do with it. Well I don’t know. Probably nothing, other than using it as a learning tool. I’ll keep my eye out for another basket case. They’re out there. This is the 54” bed, which would be handy maybe. But mainly this is about learning something. I’m open to whatever lessons come out of it. I’ve got a pretty tight grip on my wallet, so there won’t be any rush to buy all the parts back. Here’s a novel idea: I buy the castings, which sell for peanuts, and start working on making all the innards. I feel like I remember reading you can get part drawings from Clausing. Anyways that’d keep me busy for awhile. There’s a lot going on in that apron. I’m pretty sure there’s nothing in there that’s precision ground. How’s that for an exercise in futility? Lol
 

Bob Korves

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Jul 2, 2014
Messages
6,626
Think about the "sled" with the indicator measuring out in front of it. On any evenly curved surface, curved upward or downward, it will show zero deviation. On any dead straight and parallel surface, it will also show no deviation. On an unevenly worn surface, the needle will move up and down, but what exactly is it telling you? Which is going relatively up or down? The indicator or the sled? How do you keep track of the cumulative errors? A nice, perfectly flat cast iron straightedge can be compared with the ways to find the high and low spots. The right tool for the right job...
 

WCraig

Active Member
Registered
Joined
Oct 10, 2018
Messages
186
Well I thought there must be something wrong with my test set up. Maybe setting up and carefully leveling the bed to a surface plate, and then sliding the indicator base on the granite with the indicator tip on the ways?
...
Use a straightedge and feeler gauges to identify low spots. Lee Valley offers some reasonably priced straight edges in various lengths that will give you a fairly accurate test.

http://www.leevalley.com/en/wood/page.aspx?p=56676&cat=1,240,45313,56676

BTW, Lee Valley also sells Starrett straightedges. These are more spendy but made to higher tolerance.

http://www.leevalley.com/en/wood/page.aspx?p=71658&cat=1,43513,51657

Craig
 

jwmay

Active Member
Registered
Joined
Dec 3, 2017
Messages
405
So with a 123 block, and feeler gauges, there’s no spot in the entirety of the ways that will accept a .0015” feeler. I can imagine that using a straight edge spanning both ways could show a different result. So I will see about borrowing a precision level from work. Thank you gentlemen.
 

Dabbler

Administrator
Staff member
H-M Supporter - Gold Member ($25)
Joined
Oct 11, 2016
Messages
935
You can also check using a precision parallel or straight edge that is 12" long or longer and using a light behind. If an incandescent light is seen between the way and the straight edge, and is orange, you have a thou or more at that point. As the light becomes bluer, you have a smaller gap, and when it winks out entirely, you have less than 3 tenths. When it is orange, you can measure with a feeler gauge...
 

Bob Korves

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Jul 2, 2014
Messages
6,626
A 123 block is not long enough to test anything with feeler gages, If you find problems with it, you can probably see the problems without it.
 

WCraig

Active Member
Registered
Joined
Oct 10, 2018
Messages
186
So with a 123 block, and feeler gauges, there’s no spot in the entirety of the ways that will accept a .0015” feeler. I can imagine that using a straight edge spanning both ways could show a different result. So I will see about borrowing a precision level from work. Thank you gentlemen.
A precision level and a straightedge are not the same thing. Ideally, you want a quality straightedge that is as long (or longer) than the bed. Remove the tailstock, carriage and headstock. Normally, there will be virtually no wear under the headstock and very little at the far end of the ways where the tailstock normally sits. Thus the straightedge from one end to the other will help you find the areas of maximum wear. Normally a short distance in from of the chuck is the lowest but it depends how your machine was used. The outside edges of the bed guide the carriage and could also have wear. Same for the inside edges of the bed as they guide the tailstock. Even the undersides of the bed could have wear although you may want to just check for variations in thickness with a micrometer.

Craig
 

jwmay

Active Member
Registered
Joined
Dec 3, 2017
Messages
405
Well thanks guys. But we've already reached the point of financial sensitivity. If I need a precision straight edge longer than 54" to properly evaluate the machine, then I'm done evaluating. I'll do the micrometer checks to see what sort of variations in thickness the bearing points have, just for curiosity's sake, and put the headstock back on, find a nice lighted spot in the barn for it, and use it for making file handles. I'll save myself a hundred dollars if I make a hundred of them.

Honestly, I'm not sure I'm giving up yet. But I'm not buying a one time use item to find out if something I decided was junk a long time ago, actually is junk. lol
 

Dabbler

Administrator
Staff member
H-M Supporter - Gold Member ($25)
Joined
Oct 11, 2016
Messages
935
I need to correct a mistake here.

You only need a 12" straight edge to measure a hobby lathe. they can be had for under 30 bucks. I don't know where Craig is coming from but it is simply not true. On a lathe you will find that the only discernible wear will be within 10-18" of the headstock. Even on 100 year old lathes, the way seem to not be worn right against the heaststock and can serve as part of the reference surface. I chose 12" to be safe, but I personally use a 10" reference surface with very good effect.

This is something you will eventually need for good quality control on your machining, if you care to do close tolerance work, eventually. For a long time I used a 6" parallel and borrowed a straight edge for the 3 or so times a year I needed it. I then graduated to a 10" Moore and Wright engineers' level, used as a straight edge (It was ground to be within 2 tenths, but the reality was closer to 4 tenths).

I now have a 10" straight edge from Starret that I use in place of the level. If I had been smarter, I'd have bought a 12" which is why I'm recommending it to you. You can measure an entire lathe bed - even a 60" one in 12" increments and learn a lot. All you need is a straight edge and a bright flashlight.
 

WCraig

Active Member
Registered
Joined
Oct 10, 2018
Messages
186
I did say "ideally". The Veritas 36" straightedge is USD $92.50 and guaranteed to be within 1.5 thous over its length. They also offer an aluminum straightedge that is half the price but only guaranteed to be within 3 thous. For an old lathe bed that was on the junk heap, a 3 thou measurement error is probably not be a big deal. If the bed only showed, say max 6 thous out, I would be done!

A 12" straightedge seems a bit short to me. OTOH, if the OP can beg or borrow any straightedge, that would be a start. If there is a serious valley (like 50 thous) in front of the chuck, the cost of correcting that likely exceeds any reasonable value for the machine. OTOH, if no light shows...

Craig
 

jwmay

Active Member
Registered
Joined
Dec 3, 2017
Messages
405
If a smaller straight edge can be used, then I'm back in the game. I did spend an hour going over the ways with a micrometer. I'm now wondering about the accuracy of my micrometer. I was unable to come up with a variance of greater than .0005" from one end to the other...anywhere. I did not plot these measurements, as I should have. But I can always do it again later. I'm unable to get into spots that are obviously lower, due to the rack being in the way. It's been pinned in at some point, and I haven't thought much on how to remove it. I'm always thinking it's a taper pin, and remembering horror stories of driving a taper pin from the wrong end in cast iron.

While using a 6" parallel and a light, I could see blue light or no light, in all the areas of obvious wear. I tried my feeler gauge anywhere it looked like "more" light, but it still wouldn't fit. The areas of wear are not as wide as a feeler gauge evidently. I'd say less than 8 inches from the headstock, there's no more light. I did just realize I was using the wrong light source, so I'll make the test again.

Thank you gentlemen.
 

Dabbler

Administrator
Staff member
H-M Supporter - Gold Member ($25)
Joined
Oct 11, 2016
Messages
935
A 12" straight edge will give you a better reading... a 24" better still, but with diminished returns, the longer the better, but you'll find 12" is more than sufficient.
 
[6]
[5] [7]
Top