Proxxon lathe PD 400

My proxxon lathe is 14 years old ... it has been very extensively used for a lot of work ... now it needs the headstock lining back up with the bed and tailstock ... the headstock assembly is clamped to the bed with 4 off long 8mm bolts ... so far I have been unable to extract any of them ... they are buried in the middle of the headstock casting ... the only way to reach them is via an Allen key ... please has anyone come across this before ... I would like to find a non-destructive way to remove them ... another thing that could be causing the taper problem are the 2 taper roller bearings ... they look clean, shiny and smooth ... can they wear on one side of the outer cup so the spindle is effectively pushed over sideways ?

Anybody any offers of help please ?

martin

Martin,

I've looked for an exploded view of the PD 400, but failed.

Some photos of your machine might be helpful.

If you are sure that you need to remove the headstock, I would certainly recommend using a 'socket-set' drive, for controllability.

MichaelG.

.

P.S. ... I was a little surprised to discover: Headstock: Of die-cast aluminium. Oversized main spindle with two adjustable taper roller bearings and MK 3 bore on chuck side. Spindle bore 20.5mm. Run-out without chuck 1/100mm.

https://www.proxxon.com/en/micromot/24400.php

Edited By Michael Gilligan on 15/05/2018 22:41:10

Page 86 onward **LINK** Oddly, the four holes in the bed are shown but the screws themselves don't appear to be listed.

Would it be practical to cut the "L" off of a long allen key and turn it with a socket?

Brian

Thanks, Brian

MichaelG.

Just had a look at mine 'in the flesh' as it were, and the headstock screws align with the four holes for fixing the headstock end of the bed to the bench - so should be doable in the way Brian suggests, or with T-handle Allen keys.. Personally I'd be looking at the bearings/spindle housing though unless you've got good reason to think the headstock has been somehow knocked off true.

Robin

Before you go moving the headstock around and upsetting the factory alignment, a few things should be clarified and/or checked:

What exactly is "the taper problem"? How much over what distance? With the tailstock supporting the work, or without? (There are two distinctly different alignment issues between the two).

1. Have you correctly set the headstock bearings to the correct pre-load and ensured there is no spindle movement up and down or side to side?

2. Have you checked the lathe bed for "level" (which is really twist but is sometimes determined using a level)? If not, the Myford ML7 users manual available free all over the net contains the best description of how to level your lathe bed and how to add a small amount of twist to it if necessary to make it turn parallel. Consists of putting a small amount of shim under one foot or other on the tailstock end of the bed.

3: Regarding lining spindle with "bed and tailstock", aligning the tailstock is a separate operation, done last, by adjusting the tailstock body on its base to line it up with the spindle. Nowt to do with headstock mountings etc.

All these should be thoroughly checked and eliminated before you disturb the headstock mounting.

Re wear on the taper bearing races, yes they can wear. They should not be too expensive to buy (buy races and bearings, no point in replacing one without the other) and after 14 years could possibly be a little worn. I regard roller bearings as "consumables" like clutch plates and brake pads and like to replace them when I have something apart.

Edited By Hopper on 16/05/2018 07:17:19

You shouldn't need to do anything with the headstock, it shouldn't need removing unless you are doing a full bed regrind or sumfink

... now it needs the headstock lining back up with the bed and tailstock...

I'd start with a bearing replacement and take it from there, especially since its been worked hard for 14 years

lathe boring taper, not parallel.

How is it turning ? Are O/Ds not parallel by a similar amount in the same direction ?

I would expect a bearing issue to show up as surface finish / vibration rather than taper on the workpiece. Clean, smooth, shiny taper roller bearings sound like new ones ! I would regrease & re-install them, then set to zero play / lift (or slightly preload) and try your cutting test again, both boring and turning, before diving in further.

Have you tried checking the headstock alignment ? The usual way is with a Morse taper test bar in the spindle socket, with a dial gauge mounted on the saddle. "Normal" alignment is for the test bar to point slightly upwards & towards the toolpost (slightly being around 0.001" per foot). The test bar set between centres can also be used to set the tailstock for parallel turning, though a test bar of the tailstock socket taper is required to check the actual tailstock alignment. At a pinch, you could go for a 2MT test bar to fit the tailstock & use a 2-3 sleeve to check the headstock, though adding the sleeve also adds a potential source of error.

If you want to start throwing money about, a test bar would be a better place to start than changing out the bearings initially IMO - you could recoup much of the cost by selling it on later if you felt you no longer needed it after use.

Nigel B

Wear or lack of pre-load can cause taper boring/cutting - a long(ish) workpiece would deflect further from the true centerline at the end of the workpiece than at the chuck end.

Somehow doubt it is wear though - perhaps you could give us an idea of the taper error over length? If is is wear, which may have resulted in loss of preload, then try putting on preload, even a bit to much, and do a test bar - if that is still taper, then you may have an alignment issue.

Joe

Wow ... I am always amazed at the depth of knowledge of the Model Engineer readership ... 6 replies and all aiming straight at possible causes ... thank you all.

MICHAEL ... I have the original handbook for the lathe ... the headstock assembly is not shown ... everything else is !

BRIAN ... the long Allen key is a very good idea, but I have already tried this ... I have a 3/4 inch socket set as well as the usual 1/2inch set ... the hexagon head in the bolts is 6 mm ... I have tried using 3/4 drive right down to 6 mm with 2 small adaptors ... I am worried about something shearing !! the bolts are steel, tightened into an aluminium casting ... did someone spit on the bolts before assembly ? or use Loctite ? they certainly will not budge. Maybe this is a clue from PROXXON that they do not expect it to be moved ?

HOPPER ... the error is 0. 003 ( or 3 thou. ) diameter for about 1 inch length ... not very easy to get a piston to fit a cylinder bore !!

MGNBUK ... I have NOW used a test bar and it is pointed nicely at a centre held in the tailstock ... the error only shows up when cutting ...hmm ... hmm ... that might point the finger at the roller bearings... I have found the exact bearings in the Arc Euro Trade catalogue ... they are £ 16. 99 each.

JOSEPH ... maybe the error is mine ... maybe I have not put enough preload onto the bearings ? ... difficult, as I have no idea what it should be ... some of my old cars, like an Austin 1800, needed 300 ft/ pounds as a preload on the steering swivel nuts ... surely it will not need that much ??

Can I ask a general question please ... if the bearings look o.k. but do need replacing, can anyone suggest a non-destructive method of removing the bearing nearest the rear of the chuck backplate ... the opposite one is a light interference fit, but this is very tight ... so far it has not moved ... which is the safest method, oxy-acetylene for a bit of local heating or a grizzly disc please ? Obviously I want to avoid damage to the headstock spindle.

martin

.

Martin,

Knowing very little about the Proxxon lathes, I thought it best to step back ... BUT given your update, I would like to offer my [now a little better informed] further comment:

It is clear from p88 of the document that Brian linked, that the 'die cast aluminium' headstock is potentially a fragile thing ... I have no proof of this, but my "reverse engineering" of what I can see would be that the bed is probably machined, then the headstock attached [permanently] and the bearing housing bored in-situ [using the bed as its reference]. Consequently, I would be very wary about detaching it.

MichaelG.

.

Edited By Michael Gilligan on 16/05/2018 13:30:35

As the headstock is aligned using the v-bed, it's unlikely to have shifted.

My worry is that the alloy headstock casting, which is very lightly built compared to most lathes, may have distorted over the years. This would be very difficult to fix, although a combination of local heating and application of controlled force might help..

Preload should be pretty minimal for a lathe.

Neil

Light preload for a lathe, always light preload, and adequately lubricated

Afterthought:

The lathe may not actually be designed to have its bearings replaced, it may have a built in finite lifespan

Edited By Ady1 on 16/05/2018 18:33:30

I have NOW used a test bar and it is pointed nicely at a centre held in the tailstock ... the error only shows up when cutting ...hmm ... hmm ... that might point the finger at the roller bearings...

But have you checked that the bearings are correctly adjusted before condeming them ?

Mount a chuck & chuck up a length of stout bar. Set a dial gauge from the bed to indicate at the top of the chuck backplate (12 o'clock). Grasp the bar & lift it away from the bed firmly. If there is lift in the spindle, it will show on the dial gauge. Adjust the bearings until you get no lift, add a small amount of preload & try a turning and boring test again - do both to see if the taper problem is apparent inside & out. If you cannot adjust to a "no lift" situation, then consider replacing the bearings.

When rebuilding CNC lathes in my last employment (much larger machines than a Proxxon !), we used to "bump" front bearings off the spindle. The spindle was stood upright on its rear end on a piece of wood (railway sleeper), then a stout bar was bolted across the chuck mounting face. Two of us lifted the spindle a few inches off the sleeper (the top of the spindle was about shoulder height with knees slightly bent from the floor/ sleeper) then dropped it vertically. With repeated "bumping", the front race walked off it's seat with no damage caused. The spindle also had a safety sling loosely mounted from the O/H crane so it couldn't fall over accidently if we got "out of sync" with the lift & drop sequence. The replacement bearing was warmed with an induction bearing heater & then it just dropped on.

Joe's experiences re: loose bearings appear to be different to mine - spindle bearing problems pretty much always manifested themselves as surface finish defects for me, not alignment issues. 0.002" lift on a newly rebuilt machine for an oil industry customer was enough to fail acceptance due to chatter while screwcutting - adjusting the bearings so there was no lift (which required the use of a crane on a bar bolted to the spinlde nose - couldn't tighten the adjusting nuts sufficiently by hand !) completely cured the problem. These were "NNU" type expanding roller bearings , though, not taper rollers - but a loose spindle is still a loose spindle regardless of the bearing arrangement ! At the employer before that, heavy lathe spindles were preloaded by running temperature - if the headstock didn't get to 40 C after a predetermined period at full speed, they were too loose, over 50C they were too tight. These were taper rollers, a "fixed" pair at the front end & a "floating" pair at the rear - as the hollow spindle passed 16 5/8" though, I would guess the bearings were around 24" bore.

Nigel B

.

From p20 of the book:

Main spindle

The 2 taper roller bearings of the spindle are maintenance-free for at least 6000 hours at minimum speed and 1800 hours at maximum speed. If slight play is evident after this period, the bearings can be adjusted by a specialist.

MichaelG.

That's the best way to do it. Practical in these days of cheap IR thermometers, although you can use your hand...

Neil

Wow. That's massive. That's 36 thou per foot of taper! Did it just suddenly start doing this, or has it come on gradually over the years?

Which direction is the taper? Larger at the tailstock end? Or smaller?

With that much error, it seems like more than just a tad of bearing adjustment or bed levelling needed. Could it be the aluminium headstock distorted over the years? Or the bed itself?

If so, you might have to resort to setting the top slide to turn parallel and use that for critical jobs.

Edited By Hopper on 16/05/2018 23:39:58

Leave the tailstock out of the equation for right now, it means nothing until after the headstock is aligned to the bed.

If you put the test bar in the spindle, and check it with a dial indicator while rotating the spindle to ensure there is no runout, you should then mount a dial indicator to the carriage with the plunger on the test bar. Run the carriage back and forth along the bed. The dial indicator will then give you a reading on the relative alignment of the spindle to the bed ways. Do this in the horizontal and vertical planes. It should be within a thou if all is well.

--If slight play is evident after this period, the bearings can be adjusted by a specialist.--

Who is?

The 2 taper roller bearings of the spindle are maintenance-free for at least 6000 hours at minimum speed and 1800 hours at maximum speed.

This is actually pretty poor

Pultra stuff ran 12 or 24 /7 for decades from the 1940s

It was basically all burned out by the 1960s

Edited By Ady1 on 17/05/2018 00:37:23

.

Ady,

Please don't shoot the messenger ... I was simply sharing what I had found in the book; not expressing an opinion.

I have no idea who Proxxon would suggest as 'a specialist' but at least it gives us some comfort to see that the bearings are considered adjustable.

I'm surprised that the exploded view does not seem to detail any spacer between the bearings; so perhaps only those 'specialists' are privy to the design details of the headstock arrangement.

MichaelG.

.

[my Pultra's bearings are fine, thank you ... but the PD 400 is surely intended for a very different market]