New lathe arrived today : The ongoing saga

[AjohnwParticipant @ajohnw51620]

I've mentioned slideway oil before Brian. It will help. I can't search ebay oz for some reason but there may be 1l available on there. 5l is available from several sources in oz. A bit too large a quantity for you. There maybe some one who supplies smaller amounts.

John

–

[Brian JohnParticipant @brianjohn93961]

Do you mean this ? Does it really make a difference ?

**LINK**

I have just had another closer look at plate 219. You can see the 2mm strip either side of the clamping plate which makes contact with the bottom of the ways (top and bottom of the photo). These areas should definitely be machined or filed as it is still very rough. It is not as obvious in the photos but quite clear in real life. I know Jason has said his plate is similar but my lathe is very poor quality so things like this could make a difference.

Edited By Brian John on 25/10/2015 13:07:32

Edited By Brian John on 25/10/2015 13:07:57

Edited By Brian John on 25/10/2015 13:18:58

[AjohnwParticipant @ajohnw51620]

Posted by JasonB on 25/10/2015 12:36:24:

John the grub screw tips go nowhere near the rails they just set the gap, if you look at the photo of the underside of the carrage marks the grub screws have left

I got curious and rubbed my finger where the similar sort of plates on my 280 bear against, also quite rough milled finish compared to the rest of the bed, now as I can get a good finish on my lathe and a freely moving carrage then is there a need for this surface to be smooth?

A shim is all well and good when packing between two flat surfaces, the 4 grub screws are likely to be a better option if the plate and carrage underside are not flat.

I missed the slight wear marks at the edges.

Gibs of any sort that use screw arrangements like that are always a pain to set but once set the cap screws should lock it all up so they should stay set.

Some people reckon that the underside gibs are added so that things can be milled on the lathe. They just prevent the saddle from lifting. I'm not so sure. A lot depends on how the cutting forces are transferred to the V and flat. There is often one on the back to stop that end from lifting but as you point out these generally don't need to be tight. The whole idea of a prismatic bed is that cutting forces push the saddle hard down on the V so it should remain in place. The fact that it's a V can also mean that there is a decent contact area to resist wear.

If the V in the saddle isn't straight/flat and has a slight bow it will rock. Past that problem pass but bed width to centre height of the lathe also has some bearing on it. I'm not convinced this is down to science more down what has been found to work. The lead screw might not be straight either which could also cause the saddle to rock in use..

Perhaps Brian's rock is side to side due to a warped saddle or front to back because when the V in the saddle is in firm contact the flat isn't.

Some one wouldn't see me trying to fix that with a scraper despite all of the hype, I'd use a file.

John

–

[frank brownParticipant @frankbrown22225]

The locking tab is the bit of the clamping plate that is meant to lock the saddle. The wear patterns simply show that the grub screws were not set properly.

While all the surfaces are rough, I would presume that the thickness of the bed shear is constant (roughly), the clamping plate is flat (roughly) and the machined surfaces of the saddle were done at the same time so are flat (roughly). The grubs crews cannot compensate for the shears being of unequal thickness, or the saddle bottom not being flat, just if the clamping plate is twisted. They seem to be fairly useless as they have cup points and no anti vibration locking devices on them. One easy job would be to rub the whole top surface of the clamping block on a large sheet of emery laying on a very flat surface (surface plate, sheet of glass), hoping to smooth it out a bit. That is why I thought shims would be the answer.

Frank

[VersabossParticipant @versaboss]

Just went back to #1 in this thread and saw it was on Aug. 25th. Now we have Oct. 25th, so two months of almost daily entertainment. Long may it continue – it is always the first one I have a look at when I click the site.

To add something (hopefully) positive to this story, I'm quite sure that the application of a good file in strategic points would help a lot. Scraping would be over the top, both for the lathe and for Brian

Frank, may be you should add that the locking tab is the corner which is separated from the rest by a saw cut, aand also that it is the smaller part…

Regards, HansR.

[Brian JohnParticipant @brianjohn93961]

FRANK :

1. I have rubbed brass work pieces flat on a sheet of glass I have for that purpose but would that work for steel and cast iron ? What grade of sandpaper would you recommend ?

2. I could install longer grub screws (say 10mm longer) so they project beyond the surface and put a locknut around each one. Is this what you mean by an antivibration device ?

Okay, now I understand what you meant by a locking tab. Yes, it is very ragged but it does the job.

 

Edited By Brian John on 25/10/2015 15:32:47

[AjohnwParticipant @ajohnw51620]

It may well be worth telling us which way the saddle rocks when the plate isn't attached Brian. Push down firmly over the V in the centre and see if it rocks in any direction.

As Jason pointed out and I expanded on a little there shouldn't be any need for the plate to be firmly up against the bed – other than if for some reason the basic design is screwed up.

It might be that the saddle will move across the bed when the side on the flat is pressed down. Have a fiddle about with that sort of thing and tell us what happens. Pressing down an various places and checking for any sort of rock/movement across the bed elsewhere. The V and the flat need to mate at the same time when it's on the bed so if the flat is too high, the V too wide etc it will move about in directions other than along the bed.

John

–

Edited By John W1 on 25/10/2015 15:35:41

Edited By John W1 on 25/10/2015 15:36:57

Edited By John W1 on 25/10/2015 15:47:22

[Brian JohnParticipant @brianjohn93961]

It's seems perfect : there is no rocking or movement at all. I had already tested that this afternoon and I double checked it now.

Maybe I should give the clamping plate a polish with 400, 800, 1500 then 2000 wet/dry sandpaper then put it back together. I doubt it would make things worse. I might also wait for that slide ways oil to arrive before reassembly as it could make a big difference.

Is this an unusual arrangement ? How do other lathes hold the carriage firmly in place ? I do know that my next lathe will have the carriage adjustment accessible from the top !

[AjohnwParticipant @ajohnw51620]

You've reported that you can rock it / move the saddle around in the past Brian so what has changed ?

John

–

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Brian John on 25/10/2015 17:31:03:

Is this an unusual arrangement ? How do other lathes hold the carriage firmly in place ?

.

Brian,

You might find this old book interesting

MichaelG.

[BazyleParticipant @bazyle]

Perhaps the clamping plate distorts the main saddle when fitted tightly. This would happen if one of the grub screws was set protruding too far.
The milling marks on the plate look to be partly worn along the 2mm of contact you mentioned initially even with your small amount of use. If you think about it that plate should not be in contact at all as it is underneath. It should be very very close, but not in actual contact so it should not wear.
You might try retracting the grub screws, fitting the plate firmly but not over tight with a piece of paper along each contact strip but not going as far as the grubs. Then tighten the grubs until they make contact with the saddle, just. Now tighten each grub by an equal amount in small steps until the paper just pulls out. You now have an even gap of a few thou.

Other people may think it should be in contact. Other lathes, well bigger ones, primarily rely on weight with a plate only at the back for when a dimwit takes too big a cut on a large diameter.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Bazyle on 25/10/2015 19:35:58:

If you think about it that plate should not be in contact at all as it is underneath. It should be very very close, but not in actual contact so it should not wear.

.

Thank you for the endorsement, Bazyle

That's the point I have been making for a while.

MichaelG.

[Brian JohnParticipant @brianjohn93961]

Okay, so the theory is that the plate should not be in contact…I would have to agree. That is all very well but it obviously IS in contact as can be seen from the signs of wear. So if the plate was uneven then that might cause unwanted movement in the carriage. I still think I should give it a clean up and polish before reassembly so that if it does make contact again then at least it will have a smooth surface to slide on.

I am puzzled by the fact that the carriage seems to fit on the rails so well without the clamping plate. It is very frustrating but I think I will have to have a go at reassembling everything now that I know how it SHOULD work.

Edited By Brian John on 26/10/2015 01:08:58

[Brian JohnParticipant @brianjohn93961]

Still not finished yet…a bit more to go. It may be pointless but I will do it anyway.

 

 

Edited By Brian John on 26/10/2015 06:56:06

[JasonBModerator @jasonb]

Posted by Brian John on 26/10/2015 01:06:24:

I am puzzled by the fact that the carriage seems to fit on the rails so well without the clamping plate.

Brian, are there any signs that the sides of the plate have rubbed against the bed, could be pushing the carrage out of line front/back or simply be the tightness you are feeling.

[Brian JohnParticipant @brianjohn93961]

No, there are no signs of wear on the sides.

[HopperParticipant @hopper]

Brian, can you get your digital caliper in through the bed to measure the distance between the top ways and the bottom way that plate runs on, at each end of the bed and several points between? You said it was tighter or looser at the end vs the normal working area. It would be real helpful to know just how much taper there is in the alignment between teh top and bottom ways.

Another way to measure it would be to set the clamping plate just touching the ways at the tightest spot, then slide the carriage along to the loosest spot and measure the gap between the plate and the ways with feeler gauges or bits of shim or paper etc of a known thickness as a makeshift feeler gauge.

It would pay to rub a fine file or your diamond stone etc along those rough bottom ways just a few times to knock down any burrs and ridges that are sticking out and could give a false reading on the above measurements, and cause your carriage to tighten up in use.

Edited By Hopper on 26/10/2015 10:34:49

[Brian JohnParticipant @brianjohn93961]

I am sorry but I cannot get any meaningful measurements. I do know it is difficult to get the carriage with clamping plate to move freely along the whole bed.

I have attached the clamp to the carriage (no lead screw) and made an attempt to adjust it so that it just slides easily along the whole length of the bed but it is too loose at the tailstock end and gets tighter towards the headstock. This is the opposite of what I previously thought it was but I may have been confused. 

When it is too tight (ie. barely move it by hand) there is no wobble in the carriage but as soon as it starts to slide freely then the carriage wobbles. Please do not ask me why…I have no idea ! It does not wobble without the clamping plate attached so it should not wobble when the clamping plate is attached. It makes no sense. At least we now know there is nothing wrong with the lead screw so pulling it all apart has not been a wasted exercise.

It is interesting that the clamping plate is not wider. There is plenty of room to make it wider (about 3mm on each side) but they have not done that. There is just that narrow 2mm strip down each side of the clamping plate which makes contact with the bottom of the ways as indicated by the wear marks in the original photo. The fact that they have not made it wider reinforces the theory that it is NOT supposed to make contact.

I need to have a think before I do anything else.

 

 

Edited By Brian John on 26/10/2015 12:12:55

[Martin KyteParticipant @martinkyte99762]

Assuming nothing is rubbing that shouln't . . .

Just a thought. When you do your adjustments say at the tailstock end and then slide the carriage towards the headstock and it gets tight, is this repeatable? By this I mean when you slide back to the tailstock end again and then move towards the headstock does it get tight at the same place. Do this a few times to prove it to yourself. If nothing is moving (gib strip, adjustment screws etc) and it sticks in the same place each time it has got to be the bedways which are out. The other possibility is that the gib is moving and wedging the carriage which I would have thought more likely.

regards Martin

[Neil WyattModerator @neilwyatt]

> It does not wobble without the clamping plate attached so it should not wobble when the clamping plate is attached. It makes no sense. At least we now know there is nothing wrong with the lead screw so pulling it all apart has not been a wasted exercise.

This suggests that the problem is with the adjustment of the clamping plate. As the carriage of an inverted-V lathe sits on to of the bed and should engages nicely with the vee it should be wobble free and wholly free to move without the plate fitted, as it is. Any wobble must therefore come from it being forced out of clean engagement with the vee.

Lets have a close look at how it all fits together:

Two possible sites of problems are at 'y' – is the plate too wide and rubbing here, this would push the carriage out of alignment.

The plate is adjusted by working the big screw, Z2 against the grub screws Z1. Why opposed screws? because you aren't simply clamping against eh bed, but aiming for a small but discernable working clearance. Normal cutting forces will push the carriage down onto the V-way, however, it should be a small clearance to prevent any tendency of the saddle to rise up on the vee under twisting forces.

So, in normal circumstances adjustment of the front edge of the plate (which helps keep the carriage on the vee) is much more important than the back (which does very little most of the time, unless the cross-slide is wound right out.).

One possible source of problems is the looking screw. It works by bending that tap on the plate – a crude arrangement. If over tightened with the saddle plate loose, that tab could take on a permanent set or bend, this would mean that it contacts the ways before the rest of the plate – and what will that do? Make it very difficult to get the plate well adjusted under the v-way.

I would check the plate by bluing it (a magic marker will do) and rubbing it along the bed. This will reveal any high spots 9the marker won't rub off) especially if that tab has been bent up. Ideally you want the plate to have good contact along both edges and no wobble.

Personally I would discard the locking screw (when you eventually need a carriage lock for something, you will be experience enough to improvise a better solution) and turn the plate through 180-degrees so the longer edge runs beneath the v-way.

Another thought – when adjusting the six screws Z1 and Z2, start loose and gradually nip up the Z2 screws until wobble or resistance is felt. The gently tighten the Z1 screws, this should marginally loosen the plate to a good working clearance.

Don't forget to check for rubbing at the points marked Y in my diagram.

Neil

[Neil WyattModerator @neilwyatt]

Just another thought, the pre-polishing picture of the plate shows extreme wear on one corner of the tab – this suggests that it has been set to tight – bear in mind that it's screw should be left loose for normal use.

Also the polishing may have made the contact area worse – the plate looks rounded off now. I strongly advise doing the marker test, then rubbing down any high points, until good contact is achieved along both sides.

Neil

[Brian JohnParticipant @brianjohn93961]

Okay, that's it…I am handing it over to Hopper to have a look at.. I feel that if I do anything more then I will only stuff it up further.

Today I removed the metal screen which blocks removal of the cross slide. The metal screen will not be replaced. The Perspex screen just slips between the four steel L brackets and is easily removed ; it also allows more light to come in. I will need some sort of work light soon. It is surprising how dark things get when you want to look closely at something !

I am also fiddling about with some aluminium angle you can see in the photos. I want to build two 80mm lead screw covers to go on either side of the carriage.

Edited By Brian John on 27/10/2015 10:34:29

[julian atkinsParticipant @julianatkins58923]

this is all very interesting. i look forward Hopper's report.

a thought occured to me that the 'V' on the front of the bed may not be parallel to the flat sliding surface on the rear of the bed. this would cause a 'wobble' one end and tightness the other.

cheers,

julian

[HopperParticipant @hopper]

Well, the fun has begun. We got Brian's little lathe up on the bench this afternoon/evening. Overall it does not look too disastrously bad but needs a bit of "fine tuning", as far as can be done in the home workshop without specialist scraped templates etc.

Here is what a preliminary inspection/play around found:

The carriage is indeed a loose fit at the tailstock end and gets tight when pushing it by hand to the headstock end. The carriage indeed has quite a bit of what Brian has described as "wobble" when at the loose end of the bed. The wobble is such that you can rotate the carriage back and forth when viewed from above by about .004".

So it seems the  lower bed ways are not parallel to the top ways, being further apart ("thicker"  )  at the headstock end. Simple measurement  with a mic revealed a .oo4" (0.10mm) variation on the rear ways and about half that much on the front ways.

A second test using a small DTI gauge mounted to the carriage with the pointer running on the under way confirmed these measurements.

 

This runout is over a 12" (300mm) distance along the bed. To give an idea how far out this is, for a full-sized toolroom lathe, the standard traditional spec for this lower carriage gib way is a maximum of .0007" over a 48" length. IE, pretty much a zero-zero reading over a 12" distance.

The machining quality on the lower ways on Brian's lathe is OK for the purpose. It looks to have been end-milled and is reasonably smooth to the touch, not ground or scraped at all. As this is a non-bearing surface under most circumstances super-fine finish is not required. (Although on my old Drummond it is hand-scraped dead smooth. But many other lathes use a milled surface here.

So it's looking like one of the first steps will be to file/scrape a few though off those lower ways to smooth out carriage movement all along the bed.

BUT, further points of interest popped up when we removed the lower gib plate (called the clamping plate in the manual and sometimes called the lift plate on English lathes.) Without it the carriage sits much more firmly on the major inverted V way and the rotational slop there (we could fit a .004" feeler in the gap on the V way) is gone.

So it seems the set-up with the gib plate being a piece of cast iron almost as thick as the carriage, the four jacking grub screws on the outer edges of the plate exert an upward pressure on the little carriage while the clamping cap screws exert a downward pressure on the carriage and the force is sufficient to distort the carriage.

So final set-up is going to include remachining the gib plate truly flat (it has slight drop-offs on the critical edges) and scraping flat. And then either rounding the ends of those grub screws and setting it all up very carefully with miinimal pressure – which is leaving it open to slacken off under working conditions — or replace the jacking screws with somelarge squares of flat shim to sit the gib plate where we want it and so the force of the cap screws then is spread evenly across the carriage to minimise distortion.

 On the good news side, the carriage sliding ways, both the inverted V front way and the flat rear way, are not too bad. They are roughishly machined, end mill for the flatway and maybe a shaper or planer for the V, but a bit of bearing blue on the bed ways showed reasonably good contact areas for unscraped ways on the carriage.

The upper bed ways are pretty nicely surface ground so it will not be too hard to scrape the carriage ways to match . I laid the blue on pretty thick for initial sightings and to show in the photos but still not too bad.

One end of the flat way is a bit light on in the centre area, but has some reading all along the edges, so I don't think the light spot is due to the rear way being not parallel to the front way. Checked the bed for level with my old Rabone precision level and it seems OK, but that was using the flat on the top of the inverted V as a reference point. I would rather check it to the cheeks of the V later in the week by making a little jig that rides two cylinders on the cheeks and a third on the flat way and then borrow my mate's fancy new digital super precision level to sit on that.. My money is on it confirming the old Rabone but it pays to be sure that top flat is ground parallel to the V cheeks. You never know with unknown machinists.

Will also check the carriage's  V way for parallel to the flat way on the surface table with a round ground precision bar (aka endmill shank) laid in the V groove and a DTI ..

Tthe cross slide ways seem OK. Again, one set (on the slide) appear to be nicely ground and the others (on the carriage) have been milled but not too roughly and then the nominal scraper marks thrown in their general direction as seen in Brian's earlier pics. So again, the milled ways can be scraped to match the nicely ground ones. With one set of  good ways we can fix the mating set. If both were bad, there is not much you can do without making special scraped straight edges, V templates and dovetail templates — more work than the lathe is worth for a one-off job!

So overall, nothing catastrophic but a series of small errors/faults/poor finishing adding up to a problem.

Will keep you posted as we progress.

[HopperParticipant @hopper]

Well I dont know why my last post went all skinny columns and stuff. Maybe a mod can fix it?

[Author]

Posts