Repairing Atlas 10" lathe

[bogstandard]

Ross,

You need to preload the bearings before checking anything.

With all taper bearings, they need to be overtightened slightly, and when the machine is started after a non working period of say 1 hour or longer, it needs to be run without any cutting being done for about ten minutes.

During this time, the bearings will generate heat, and the lathe spindle will then heat extend by anything up to about 0.005". This then cancels out the preload and you end up with perfect running bearings with no end float. If you don't do this warm up period, and use the lathe straight away from cold, anything you machine, until the spindle has warmed up, will tend to have bad surface finishes and tolerances.

Then, only after this warm up period, can your check the bearings for brinelling. They should feel as smooth as silk, if not, they either need to be replaced or have the races reground.

I wouldn't charge for any work done, just you pay the postage charges. Unfortunately, I am in no fit condition to get into my shop at this time.

If you take the bearings to any bearing factor, they will recommend what is available to you.

For preloading, you should really get the friction figures for the bearing, and using a spring balance and an arm in the chuck, you check the 'pull off' pressure, that is the figure required to get the spindle to turn from a standstill position before spindle warmup, and adjust the end nut until the correct reading is obtained.
If the figures for the bearing are not available, I usually tighten up the spindle until you either start to feel friction against turning or there is no end float, then turn the nut between 1/16th to 1/12th further. This will give a general setting that will work in most cases.

Most the people who have tapered bearings don't do this preload setup, and start to complain when they get rough finishes, then they blame either the bearings or machine, rather than themselves for not setting the bearings up correctly in the first place.

If you are not going to be using the machine for say a fortnight or more, I would recommend you release the preload on the spindle, and preload again when you next come to use the machine. The bearings should then last a lifetime.

I can't work in my shop at this time, but I still go in once a week and spin the lathe up for quarter of an hour. This ensures that the bearings don't 'settle' in a fixed position and they get a bit of lubrication.


John

[Ross77]

OK thanks I'll get it preloaded before testing then. Atlas recommends 1/16" turn (equivalent to two spindle gear teeth). so i will try that.

Its interesting that you recommend to remove the preload if its not being used, being able to look after and set up the machine correctly seems to be more important than knowing how to use it!!!


Thanks for the offer to regrind for free, maybe when your well enough and have finished all your over projects that have been stacking up eh...

With regard to the brinelling is it the rollers that deform or the races? (or both?)

If you don't mind me pestering you (or anyone else in the know) I also have another problem that I'm not sure of the best way to resolve. on the Halifax lathe the apron drive cog and reducing bracket was broken and there is a small groove under the rack were the cog had been rubbing the bed. I have replaced it with the atlas one but when the gibbs are tightened the same thing happens and the drive goes tight and binds.

The easy solution that I can see is to put a thin piece of shim steel on the side of the way to push it over a bit,(not sure how to secure it tho). Or the better but harder option is to pocket the mounts and sink the assembly in a bit. I did think of milling the bracket down but as its made of ZAMAK I didn't want to weaken it.

Do any of those options sound feasible? or have a missed a blindingly simple alternative:heehee:

[bogstandard]

Usually Ross, it is both races that get socketted. Hence my suggestion to get a new inner, and regrind the old special profile outer.

I changed my Atlas from a plain bearing head to a taper bearing one, and the second hand head must have stood preloaded for a few years. It was like trying to turn over a ball bearing detent fixture, it jumped out of one impression and clicked into the next one, hence that is when I found out about grinding and costs of bearings for the head.
BTW, if you used a plain bronze bearing head, you would find that the surface finishes on components are far superior to a tapered bearing head. The taper bearings were introduced to up production rates as they could have a faster running spindle, what you gain in quantity, you lose in quality. I noticed a definite quality reduction when I swapped over.

Another thing worthy of note, the bearings should be oil fed only. If you try to pack a taper bearing with grease more than half full, you run the risk of 'hydraulicing' on the rollers which can lead to roller damage, little and often for oil lubrication, if necessary, fit drip feed oilers if you can.

I can't really understand about what you are saying about the saddle gear and rack, but there is an adjustment cage on the back of the apron that allows for slacker or tighter engagement of the gear. There is an adjustment bolt for it just to the rear of the handwheel. that allows you to move the cage up and down. Also, you will find that if you slacken the two top apron screws, you can shim in the joint to 'kick' the apron in whatever direction is required, but you must ensure you realign the leadscrew afterwards, so that it doesn't deflect when the half nuts are engaged. That is done at both ends of the leadscrew, by shimming up and moving about the reversing box where it bolts to the bed, and the same again at the other end at the break away sacrificial support for the leadscrew.
BTW, don't ever be tempted to replace that break away bearing with anything else more solid, that is a safety feature to protect your very expensive gear castings and apron area plus leadscrew in case you ever have a massive jam up, where the saddle stops and the head carries on turning. That piece is designed to break under such circumstances, when the leadscrew tries to carry on turning, and screws itself backwards thru the half nuts. You should see that it is cast with weakening cut out 'wedges', to control how and when it is to break.
I have seen these lathes where people have made massive bearing block replacements or had a broken one welded up. A very bad idea.

John

[Ross77]

thanks John, concise as ever, so should i upgrade to babbit bearing then?????

Sorry im not very good at explaining things. I meant the adjustment cage you talked about that mounts on the apron and drives the rack but it is rubing on the side of the main bed below the rack.(see pic)



So when I adust the gibbs on the saddle it pulls the saddle over even more.hope that helps.

heres some pics of the lathe as it stands, not pretty but hopfully functional

[bogstandard]

Hi Ross,

If it was feasible, having a swap over babbit head would be the ideal solution, then you could use whichever one was perfect for the job.

On your first pic, the one with the drill, if you go to the top right, you will see the vertical and horizontal joint lines between the apron and saddle. If you slacken off the apron bolts on the top, you should be able to insert a shim at either end of the vertical joint. I would use a couple of coke can shims, and by doing that, the apron should be moved away from the rack.

If that didn't work, as far as I can remember, the troublesome gear is just held on the shaft by rough peening over. By the marks shown, it looks like the gear isn't square on to the shaft. There would be no harm done if you just took a file to the gear and reduced the high spot until you got no fouling. Maybe try this before shimming.

not pretty but hopfully functional

Ross, paint doesn't cut metal, so just being a bit scruffy means nothing, you can slap a bit of paint on anytime.

John

[Ross77]

Thanks John

my descriptive skills seem to be lacking.......it isn't the drive cog and the rack that are binding but the face of the cog and the bed.

If it is the rough peening then how much can I remove before it becomes weak?

thanks for the encouragement for less than beautiful machine. personally I agree with you but there is a trend for perfect shiny machinery.......

[irving2008]

What John was suggesting was shimming the vertical split between the outer and inner parts of the apron (see arrow in pic). this would have the effect of moving the gear out away from the bed.

However, judging by the position of the marks on the bed the binding isnt the gear itself but the end of the shaft on which it is located. I'd have thought taking a light grind off the end of the shaft/gear face (a few thou) would resolve that without compromising the peening/fitting...

[Ross77]

Thanks irving and John.

Yes the bind is the end of the shaft but the apron is non movable

[bogstandard]

Ross,

I'm sorry, I should have said that I realised it wasn't a gear/rack problem.

Irving has said it all.

but the apron is non movable

Almost everything on that lathe is moveable, you just need to know how to move it.

The peening, as far as I can remember was a real rough job, and I think I ground down mine to be level with the face of the gear, then centre popped around the edges of the shaft joint to keep the gear in place.

Things like this really are a 'suck it and see' exercise, and you just do what is required to get the machine running correctly. The clones were never meant to be a super high accuracy machine, but with a little work, they can easily be turned into one.

John

[bogstandard]

I suppose I had better qualify my statement in the last post where I stated that everything is moveable.

My Atlas didn't have a dowel pinned apron, but from your comment, I suspect yours is.

If that is the case, you shim up in a slightly different way to achieve the very tiny movement that is required.

Slacken off the top apron bolts to give you a tiny gap on the horizontal joint. Then you insert a narrow shim into either end of the joint, but very close to the vertical join. When you retighten the joint, you will find the apron has kicked out slightly, maybe with just one coke can shim, enough to cure your interferrance problem.
This will move the apron out by a tiny amount, but I wouldn't go more than a couple of shims as you will be putting a lot of stress on the dowel pins and castings. If you can't get the required clearance by shimming, remove the shims, then resort to grinding or filing a little off the peened area.

John