Author Topic: Atlas 10F mods and fixes  (Read 1192 times)

Bogstandard

  • Guest
Atlas 10F mods and fixes
« on: February 02, 2013, 01:31:42 PM »
Yet another transfer post, but this time from a fair while ago.

So you must realise that this was done on a machine that was still in my possession before I replaced it with the machine I have now.

I incorporated two posts into one on this topic, the first one showing some of the mods I did to my machine, and the second on how I fixed my 'droop snoot' tailstock spindle.

#########################################################################################

My Atlas 10F lathe is now 70 years young, but in the time it has been with me, I have resurrected it from a wreck, to one that will match anything size for size on the market today, with regards to keeping size and precision. All due to loving care and attention, and a few mods that the original manufacturers didn't deem necessary. OK it has a few paint chips and is a bit oily, but paint doesn't do the cutting, it is the ironwork that matters, look after that, the paint can be changed anytime.

Now a close up on the reverse selector. I actually fitted the bronze bar (the same width as the slot in the gear), and set it up on the mill, picking up on the half moon that was already in the top piece and bored the bottom piece, so I had the correct sized full support hole. But the way you suggest, having just a bar sitting in the gear groove and giving support to the gear is a definite step in the right direction, and should improve things no end.




A very fuzzy picture here, but it shows enough of what I did. I turned up a pulley with a 1/4" full depth round bottomed groove in it (1/4" deep). This was the same diameter and central bore as the actuator knob, and they were swapped over. Then a little bracket and a bent handle were knocked up, pivoting towards the top of the handle, the bend was to allow the bottom part to not hit the apron so that the top half had enough movement to operate the linkage with a piece of 1/4" unhardened silver steel used in the knob groove to move the operating linkage in and out. It never failed to work and the crossfeed became a joy to use rather than a PITA. For a quickie guesstimation of sizes and lengths, I think I made the handle out of 1/2" stainless square bar.




I had a little trouble, when heavy cutting on back gear, the back gear would jump out of mesh. So if you look at the little ball on the end of the red line, it is pointing to a collar that stops the shafts moving side to side. I drilled a blind hole into it, I think 3/16", and fitted a ball and spring. Then on the small casting the little ball is also sitting on, I drilled a small detent into the face. Once all assembled, the back gear was put into its operating position and the collar then moved around the shaft until the ball went into the detent, the collar was then tightened up in that position. So whenever the back gear was used, the detent used to engage and stop the gear from coming out of mesh. Again, another little mod that worked perfectly.




This was just a mod because I was forever losing the tightening spanner for this nut. I got a ratchet spanner, machined up the top of the nut to fit it, and basically made it a fixture to the nut. Cut the unwanted end off and put a bit of heatshrink on it to stop pain when in white knuckle mode. A quick flick of the lever and it was ready to undo or do up. Nuff said.




Now this is a collection of bits around the saddle and apron. Top to bottom.

Fitted a 100 sized QCTP. This required the bottom of the toolpost turning so that it could fit on the limited size offered by the topslide, I also had to remove a bit of the casting top, just so that I didn't take too much off the toolpost.
Next comes all singing all dancing resettable dials, courtesy of Messrs Myford and their ML7 lathe. BTW, the Atlas 10" is very similar to the ML7, just 50% larger. I was at a show, and bought two of the dials for 17 squid each, and by tapping out the centre to Atlas size, they were fairly easily fitted, but did require new index plates making and fitting. I also did the anti backlash thrust washer jobbie on the topslide and cross slide at the same time. Much better than the tiny little bits that are on there as standard.
Then comes the saddle clamp handle, I don't think I need to go into details, only to say that I got a few brand new parts from a dealer when I made up a few extras of the underneath clamp plates. They are unobtainable new in real life, and he was willing to do a fair exchange.
Next comes the major bit that you can't see. The half nuts and holder are made from Mazak, and under normal conditions, I would expect a life of about two years out of them. They were over 20 squid for a new set, and very badly cast. By the time you got them to fit, they wobbled all over the place. So I made a copy of them out of a piece of gauge plate and some lumps of brass. I also fitted a basic gib set to them, so that play could be eliminated as they wore. I checked them after about 3 years use, and nothing needed doing and showed very little signs of wear. When I sold the machine, I gave the chappie two spare blocks I had made with the screw thread cut thru them, all he had to do was cut them in half, mill a little, and he would be ready for another long trouble free machining spell.




I had a stage in my workshop where I was repairing all sorts of odd things with weird threads. So I marked up the top of my drop in indicator with 16 markings, this helped me a lot during my single point screwcutting phase.



One thing I need to mention was the fitting of a reversible 1/2 HP motor and control box. This is a must if you are to do any metric screwcutting on this lathe.

I think that fitting the smaller motor really helped keep the machine from being damaged. If it didn't like what I was doing, it would slow down or stall before any mechanical damage came to it.

Over the years, I got to know it inside out and back to front, so very rarely would I ask it to do more than I knew what it was capable of, but it did everything I required at that time.
Loads of lube everywhere, and a clean up every now and again was all it required, and it became one of my most loved and accurate machines I ever owned, up until I got my new one sorted.

#########################################################################################


This lathe is going to a new owner, and before it goes I want to repair the droop snoot on the tailstock. If you try to put a centre into a part while the ram is towards full extension, the chuck droops at the drill point by about 20 thou. This is caused by 70 years of wear as shown in the sketch. I could have done a quikfix and it wouldn't have been noticed for another few months, but that isn't my way. So as shown on the sketch I will be inserting a bronze bearing in the nose, replacing the worn area, and hopefully it won't need doing for another 70 years, so I don't think I can give a lifetime guarantee.




So, away we go.
I have been mulling this over for a couple of months now, you only get one real go at this, bugger it up and you are into a bigtime rescue operation.
I needed a datum point to work to, and eventually selected the top of the ram bore, I reckoned this would be the least worn point. By use of this datum and the screw under the nose I could get a precision bar locked in there to check if everything I am to do later will be correct.

So a bit of hex brass was chosen to make this test bar out of.




It was machined to exactly 1.125", the optimum size of the ram.



I now needed a method of checking the height of the top surface of the test bar, then be able to use the same setting to check when the bar was eventually mounted into tailstock.
I came up with this idea. A friend had given me some very high precision ball raced sliding ways a while back. I mounted my small mag base with a DTI onto the top of one. The same idea was used with my height gauge. These would bridge the gap over the bed and allow me to very smooooothly move them side to side.




They were duly put onto the cleaned down bed, and the DTI was zeroed on the top part of the bar while it was still in the chuck. This was then taken away very carefully, as it had the datum setting for the bar top, and put somewhere safe, hopefully it will retain its setting.



The same system was used to mark the centre of the bar, for later use on the mill.
After this was done, the bar was parted off, end cleaned up and mounted into the tailstock and locked into position against the top face of the bore by tightening up the screw underneath the nose.



Tailstock was locked down, then the DTI was very carefully reintroduced. The difference in height was 0.0005". I think near enough for what I want to do.



To do a double check, I took a swipe across with the height gauge. No detectable difference. So I now knew my datum was the correct choice. I could continue with renewed confidence. The tailstock was rebuilt to allow it to do its last job with a droop snoot.
A lump of bronze was mounted up in the chuck and the end was machined down to 1.375" diameter, this would give me a bush with a wall thickness of 1/8".



I had to do this shot. This is a 1" blacksmiths drill from a set that I bought at Harrogate show, and got rid of the metal as though it wasn't there. I love drilling big holes.




What a useless picture you might think, but this is one of the major bits of the repair, and I will explain why.
I very accurately bored the hole to fit the ram, job done. It was why it was done in that position that was the main part.
If you notice, all the work on the bush was done OUTSIDE the chuck jaws. If it was bored whilst inside, because of the thin wall thickness, the pressure of the jaws would minutely distort the inner surfaces whilst the boring is carried out. When it is taken out of the jaws, the metal springs back to a circular shape and you end up with a slightly triangular hole. Do all boring work like this OUTSIDE the chuck jaws if you can.
The bush was parted off and cleaned up very carefully, then put to one side.




The t/stock was stripped down and clamped to an angle plate, with the t/stock sitting on a pair of parallels between the t/stock and angle plate.
The ram was locked into place and used as a datum face to get the t/stock into perfect vertical position. It was then checked at 90 deg. to the first and it was spot on.



The ram was replaced with the test bar, and was roughly centred by using a point in the chuck and rough aligning it with centre marks done with the height gauge. Thank goodness for pre planning the whole job.
The DTI was then mounted up, and the test bar was centred up. Table locked up until all cutting was finished. Time to take the plunge, no backing out now.



Really sweating on these cuts now, if anything moved or was banged I would be in deep s**t. The hole was bored to 3/4" deep and to a very nice sliding fit for the bronze bush. Don't want interference fit on the bush, because it is machined to such fine tolerances, it just might make it slightly undersize if it is forced into the hole, again causing anal sphincter twitch.



Boring done and perfect. A fine coat of engineering adhesive (Loctite), and the ram and bush slipped into position. The little clamp bolt was tightened slightly.
The whole lot was left overnight, untouched, to allow the adhesive to reach max strength.



A new day, and the ram anti rotation screw hole was drilled and tapped to penetrate the bush, this also acts as a secondary retention of the bush, to give backup to the Loctite. The t/stock was then reassembled using light lubrication inside and out.
Whilst it was off the machine, I took the opportunity to give the t/stock gib and setover screws a good checkout.



Time to try it out.
Ram out close to max, the ram is slightly tighter that it was before, but that will bed in nicely. No up/down movement detectable.



Using the smallest slocombe (centre) drill I have, I fed it into the test piece.



Job done and dusted.




Did it work?

You betcha, better than I could ever have hoped for.

All the worrying, preplanning and doing now over. With a perfect end result.

John

Offline Don1966

  • Full Member
  • *****
  • Posts: 2962
  • Morgan City, LA (Along the Gulf Coast)
Re: Atlas 10F mods and fixes
« Reply #1 on: February 02, 2013, 04:24:56 PM »
Beautiful bit of planning and execution John. It was very educational for me. Thanks for reposting it.

Don

Offline Bill_S.

  • Full Member
  • ****
  • Posts: 27
Re: Atlas 10F mods and fixes
« Reply #2 on: February 02, 2013, 05:27:36 PM »
Bogs, thanks for explaining the WHY you did things in the manner you did.
As Don said:  VERY educational and beneficial.  Kudos!
Bill

Offline + or - Zero

  • Full Member
  • ****
  • Posts: 11
Re: Atlas 10F mods and fixes
« Reply #3 on: February 04, 2013, 11:20:54 AM »
Thank you. Another bit of valuable information saved.

Probably needless to say (and maybe annoying for you to hear --but it isn't intended to be);  More please, as you can.

Thanks again.