Level lathe set up

[Ex contributorParticipant @mgnbuk]

No need to be like that Nigel.

Apologies for any offence, Russell – my remark was intended to be light-hearted & I forgot to add the "smiley".

The object of "levelling" the lathe is to return it to the state that the manufacturer built the machine in when they documented that it met the accuracy standards. This will then allow the user to confirm (or otherwise) the manufacturer's documented readings. I do not believe that the "Method" (be it Rollie's dads or Rolly's dads) will do that. That does not make it an unreasonable method of checking headstock alignment if a suitable test bar in not available, but after the bed has been set "level" .How does this "Method" differentiate between a twisted bed and a mis-aligned headstock ? There is more to using a lathe than just turning parallel – will twisting the bed to get parallel turning when the headstock is mis-aligned cause problems with the tailstock alignment at different distances from the chuck that put drills off centre, for example ?

Is £40 unreasonable for a piece of precision test equipment ? Is it really a lot in terms of the cost of the lathe, the tooling require to be able to use the lathe & the other measuring equipment to check the output from the lathe ? As to how often it is used, that depends on how often you wish to use it & that will depend on how often the machine moves out of "level". One machine I worked on stated in the manual that the "levels" were to be checked monthly after installation until there was no change between two consecutive checks, then the checks were to be moved out to 6 monthly.

Nigel B

[mechman48Participant @mechman48]

…When I was briefly in the Merchant Navy the lathe in the engine room was rarely, if ever, level. Come to that you had to hang on to it just to keep yourself level sometimes!…

​snap!… same experience when I was in MN, Blue flue & ED's, the amount of flex the ships hull is built to take is surprising.

George.

[Ian Skeldon 2Participant @ianskeldon2]

Nick Hulme, David Standing is right, it was an unspotted typo on my part, sorry to have caused any confusion.

[duncan webster 1Participant @duncanwebster1]

Posted by Nigel B on 17/08/2017 10:42:17:

No need to be like that Nigel.

Apologies for any offence, Russell – my remark was intended to be light-hearted & I forgot to add the "smiley".

The object of "levelling" the lathe is to return it to the state that the manufacturer built the machine in when they documented that it met the accuracy standards. This will then allow the user to confirm (or otherwise) the manufacturer's documented readings. I do not believe that the "Method" (be it Rollie's dads or Rolly's dads) will do that. That does not make it an unreasonable method of checking headstock alignment if a suitable test bar in not available, but after the bed has been set "level" .How does this "Method" differentiate between a twisted bed and a mis-aligned headstock ? There is more to using a lathe than just turning parallel – will twisting the bed to get parallel turning when the headstock is mis-aligned cause problems with the tailstock alignment at different distances from the chuck that put drills off centre, for example ?

Is £40 unreasonable for a piece of precision test equipment ? Is it really a lot in terms of the cost of the lathe, the tooling require to be able to use the lathe & the other measuring equipment to check the output from the lathe ? As to how often it is used, that depends on how often you wish to use it & that will depend on how often the machine moves out of "level". One machine I worked on stated in the manual that the "levels" were to be checked monthly after installation until there was no change between two consecutive checks, then the checks were to be moved out to 6 monthly.

Nigel B

Surely the bed will be flat if no external load is applied? With small lathes it is easy to twist the bed when you bolt it down. What I did with my ML7 was to put a long bar in the chuck, put a DTI against it at the tailstock end, and make sure the DTI didn't move when I bolted it down. My stand was made from RHS so it was pretty stiff.

[HopperParticipant @hopper]

Posted by duncan webster on 18/08/2017 11:37:05:

…

Surely the bed will be flat if no external load is applied? With small lathes it is easy to twist the bed when you bolt it down. What I did with my ML7 was to put a long bar in the chuck, put a DTI against it at the tailstock end, and make sure the DTI didn't move when I bolted it down. My stand was made from RHS so it was pretty stiff.

This is something that seems to get overlooked in discussions of mounting lathes, or any other machinery. Standard procedure, assuming the machine is in good condition, straight, no external load etc, is to set it on the bench or floor where it is to sit and then check under all four feet with feeler gauges. Any gaps should be filled with shim before cinching down the mounting bolts. Otherwise, you risk puling your nice flat lathe bed out of shape to match the (relatively) unflat bench surface.

[Ex contributorParticipant @mgnbuk]

Surely the bed will be flat if no external load is applied?

Maybe, maybe not. That could depend, for example, on whether the bed casting was properly stress relieved at the point of manufacture – it may have been naturally straight at the time it was built & gradually move over time as the stresses worked themselves out over time. In that case, it would need "un-twisting" when you installed it later.

I used to work for a company that rebuilt & updated CNC machine tools & cannot recall any vertical borer, large lathe or milling machine installation that would sit naturally "true" – all required bolting to the floor (or other foundation, depending on the sub-soil & floor concrete depth) & carefull tweaking of the levelling bolts to achieve the required "levels". In "our" factory, which was purpose built, the floor was 600mm thick concrete & the machines being worked upon were bolted down with chemically anchored high tensile studding.

It may well be that a small machine, like a mini-lathe, would sit naturally "true" if the casting was properly stress relieved before maching, but I suspect that anything bigger would be prone to movement – castings are a lot more flexible in practice than you might think.

Boxford used to stress relieve the South Bend style beds naturally over many months. They sat stacked on pallets in the yard & were hosed down daily if it wasn't raining. After 3 months or so, the castings were shot blasted with steel slag, "topped & tailed" on a planer & put outside for another 3 months. Then another shotblast, filled & painted before final machining. The later geared head beds came in supposedly stress relieved from the foundry – they were found to "move" during heat treatment (induction hardening of the slideways) & after a lot of trials & jig was used to bend the bed hollow before it was hardened. After hardening, the tension was released, then re-applied at a lower torque setting before grinding – the finish ground bed then came off straight.

On the cabinet mounted lathes, the beds were mounted onto a bed of Isopon polyester paste on the base fabrication & set true with a level. The excess paste that exuded as the bed was pulled down was trimmed off before it hardened.

Nigel B

[David Thompson 1Participant @davidthompson1]

Thank you to everyone who responded.

Duncan Webster's contribution to the debate is thoroughly sensible, Nigel B's information on Boxford production is fascinating (you should turn this knowledge into an article for publication Nigel), I will certainly take Peter G.Shaw's advice and consult Tubal Cain's book, and I will consider Hopper's idea for encouraging coolant towards the drain. However, I remain unconvinced about the value of precision levelling and tend towards the Rick Kirkland view.

In some occupations and workplaces "rituals" develop. Things are done because they always have been and when examined are seen to have no value. Surely engineers are too hardheaded to allow these rituals to occur. I can see that it is sensible to have a lathe level, as you would a workbench, but I can't see why it should be necessary to use sophisticated instruments for this purpose, a builders level will do.

Incidentally, it would be interesting to hear of any "rituals" that readers have observed in engineering.

Thanks, David Thompson.

[ChrisBParticipant @chrisb35596]

Been following with interest as very soon I will need to set up and level a lathe and mill on benches I've fabricated. Was not planning anything fancy, my thought was the lathe being a casting should be more rigid than a frame, the top being marine ply should conform to the lathe as well, so I was not bothered by warping issues. As for leveling in terms of the lathe relative to the ground I was going to use a common spirit level and adjust via adjustable feet on the bench. I believe this should do for a small hobby lathe such as a WM280. If it were a bigger lathe on its own stand such as a Colchester Master and the likes it would be another story…

[Niels AbildgaardParticipant @nielsabildgaard33719]

Posted by ChrisB on 19/08/2017 17:15:37:

my thought was the lathe being a casting should be more rigid than a frame, the top being marine ply should conform to the lathe as well, so I was not bothered by warping issues.

As for leveling in terms of the lathe relative to the ground I was going to use a common spirit level and adjust via adjustable feet on the bench.

I believe this should do for a small hobby lathe such as a WM280.

 

My expirience is that ligth lathes (including WM280s) shall not be left free.

 

They shall be bolted directly on top of a very heavy concrete or rock (I use granite) block.

Using a test bar or Rollie method it will be a matter of minutes to put shims under one of the four feets and enjoy cylindricity in all eternity..

The added mass and damping enables lathe to make more pleasant sounds.

 

http://imgur.com/UL2l90I

Edited By Niels Abildgaard on 19/08/2017 19:58:49

[ChrisBParticipant @chrisb35596]

Posted by Niels Abildgaard on 19/08/2017 19:41:07:

My expirience is that ligth lathes (including WM280s) shall not be left free.

 

They shall be bolted directly on top of a very heavy concrete or rock (I use granite) block.

Nah, should have mentioned, yes of course I'm bolting it, through the plywood and frame. Probably having plywood sandwiched between the lathe bed and the frame will act as a damper and absorb some level of vibration…just a guess! No granite slab for me, true the WM280 is not enormous, but a slab the thickness shown in that photo fit for my lathe would weigh alot!

Edited By ChrisB on 19/08/2017 21:38:55

[martin perman 1Participant @martinperman1]

Just to add my two penneth, until I retired recently I spent my working life either Installing, servicing, repairing or general maintenance of machine tools from workshop lathes, multi spindle lathes, vertical lathes,CNC machining centres, cylindrical grinders, centerless grinders and very large industrial washing machines and they all had one thing in common and that was that when they were being built and or installed you always had to set the base to a level position and this became your datum from which all else was set to. I was sent to manufacturers for training and followed the build process from the casting shop through to the final build and test, any base castings, bedways etc all spent months being stress relieved outside. In the 80's one manufacture of centerless grinders went away from using cast and fabricated bases and made theirs out of concrete with steel mounting plates set at various positions to which all of the other components were fixed.

When the concrete had completely set the bases were leveled and the steel pads were machined flat so that they were all at the same relative height to each other then the machines were built.

Once the machines were at the customer and in position the first job was to level the machine using a machine level or collimator depending on how long the machine was, once this was achieved production tests could begin to check for parallelism, taper, run out etc

My last job was the installation of large washing machines, tunnel washers, these all had to pass in kit form through standard doors and whilst installing these had to be leveled otherwise it was very difficult to bolt then together.

Martin P

[Joseph Noci 1Participant @josephnoci1]

Seems this subject does the rounds again and again…

Level is a complete misnomer. You could level the bed with the most accurate level possible, but this gives no hint as to headstock or tailstock alignment, etc. I realise the subject here is 'leveling' the lathe, but that means and achieves nothing.

What should obviously be important is if the lathe machines straight and parallel. Merely leveling the bed is no guarantee of this at all. There are various tried and tested way of machining a bar, held in chuck and another between centers, that provide all the info and measurements to achieve the best accuracy within the limits of that machine ( wear, original machining tolerances, etc). The aim is obviously to ensure that a line through the centre of the spindle passes through the centerline of the tailstock, and is parallel to the lathe bed, vertically and horizontally. This line does not need to be 'level'…

If the bed is 'naturally' twisted, then obviously counter-twist is required, as mentioned by others, shims under the 'feet' or whatever. Bolting a straight bed down onto a twisted surface that does not give obviously negates the objective..Bolting a straight bed onto a straight cabinet and then bolting the cabinet onto a twisted surface, likewise..

Much depends on the initial treatment of the castings – stress relieving, and then how the bed was machined.

I have an Emcomat 14D lathe, which is a toolroom quality lathe, easily equal to the top-line Hardinge of old. This comes fitted to a steel cabinet made from 2mm plate – a vertical box section under the headstock sectiom, and a similar box section under the Tailstock area. Joining the two is a horizontal box section, welded to each. This assembly – lathe and cabinet, sits on the floor on three height adjustable 80mm diameter steel pads ( 12mm thick) with a 4mm rubber pad under. Two pads on the left, one forward, one aft of the headstock cabinet, and one pad under the right section of the Tailstock side cabinet, in-line with the centerline of the lathe bed. The instructions from EMCO for 'leveling' are – adjust leveling pads to ensure that coolant comfortably exits the drainage hole…

The 14D on its cabinet weighs 650kg.

I have verified the Factory accuracy QA check sheets supplied with the delivery, and they are impressive..

I subsequently fitted my EMCO V10P lathe with similar mounts..

No bolting down to Planet Earth, no granite slabs, no high-tensile bolts, etc….

Just get the cabinet firm so it does not wobble or move about in use, ensure the lathe is firmly fitted on the cabinet, and then do the Bar Machining tests, and fit shims or use adjusting screws ( Note – I did not call these 'leveling screws'!) till the machining tests delivers the best accuracy you can get out of the lathe. If you mount the lathe on wood, unless the wooden base is a good Oak cord in size, the wood will probably follow the bed's twist, so not really useful…

Joe

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Joseph Noci 1 on 19/08/2017 22:26:55:

Seems this subject does the rounds again and again…

Level is a complete misnomer.

.

I was resisting the temptation to join this thread, Joe, but I feel obliged to defend the word

In my understanding, the problem lies simply with 'use of English'

We are checking to confirm that various reference planes are 'level' to one-another.

[alternatively; adjusting to make them so]

The process often makes use of a spirit level, or equivalent; which is where the unfortunate and inappropriate association with "horizontal" has crept in.

MichaelG.

[Niels AbildgaardParticipant @nielsabildgaard33719]

Posted by Joseph Noci 1 on 19/08/2017 22:26:55:

Seems this subject does the rounds again and again…

Level is a complete misnomer. You could level the bed with the most accurate level possible, but this gives no hint as to headstock or tailstock alignment, etc.

 

The process can be called lathe axis centralizing and the confusion disappears?

It is maybe of interest to se how much better my scheme of Box on Rock is than Box on tinplate or plywood.

Is there a soft mounted Boxford owner that is ready to do a little unpaid experimentation for the benefit of new lathe centralizers?

I have some M16 12.9 socket screws and that is about as nasty a turning steel as possible.

One is turned to a specified removal rate in England and one in Denmark and let us compare Youtubers.

Tool has to be the same and I have a couple of swiss CCMT 009 T304.

Something like :

https://www.youtube.com/watch?v=Squ7h5XtFj8

 

Edited By Niels Abildgaard on 20/08/2017 09:22:13

[duncan webster 1Participant @duncanwebster1]

Niels's use of a large lump of granite is sound, if perhaps overkill. I've read that some large machine tools have the bed castings/fabrications filled with concrete to increase the mass and stiffness, but mainly I think give some damping. Rather than a lump of rock I'd be inclined to weld some feet to a bit of tube (square for preference), fill it with concrete and bolt the lathe to that. If you can get the feet skimmed flat(ish) so much the better. Bolting your ML7 with its fairly flexible bed to a wooden bench which will warp and twist with the weather is not a good start

​

[Nigel RiceParticipant @nigelrice73086]

I have enjoyed following this thread with the differing elements concerned. I trained at the Humber Motor Co., Coventry during the early 50's and was involved with the installation of a new machine shop for the Hillman Minx OHV engine. The largest machine was a horizontal broaching machine for machining 5 faces simultaneously on the cast iron engine block. This weighed in at about 200 tons, stood upon 3' of concrete, if I remember aright. Machine tools were "leveled" using steel wedges and a precision level. I do not recall any being bolted down. The in house foundry produced the cast iron castings and these were left to "weather" for 8 to 12 months. I subsequent machine shops the only item I ever bolted down were bar feeds for capstan lathes.

With a retirement workshop and Myford ML7 as lathe, I have mounted this on two pillars of concrete blocks. In a previous workshop I cast a concrete base but had problems with the shuttering as I had not calculated the not inconsiderable weight involved! I "leveled" with a builders level but used a laser mounted at the two extremes of the bed to remove twist. This worked well, with the opposite workshop wall being some 4 metres distance and the laser "spots" initially having about 75mm vertical separation. With the use of shims and holding down bolts this was reduced to zero. It may not be "level", but I believe it is aligned! No vibrations are detected in this set up.

[SteveIParticipant @stevei]

Joe,

Your Encomat example is indeed comparable to the manual Hardinge lathes as they too are mounted to the cabinet on 3 points; 2 at the head stock end and 1 at the tail stock end. In this instance the whole subject becomes simplified as a 3 point approach is optimal and optimal results are achievable with such a design on a very rigid short bed tool room lathe. The HLV-H's are up around 1000Kg including the steel fabricated cabinet, a different beast when you consider it is basically a 10" (swing) x 20" lathe.

The issue at hand is how to level a lathe that has been built less rigid and which has not been designed with 3 points of contact.

On a side note Last year I scraped in some straightedges. One had been cast over 35 years ago and had been planed and then left in a barn. Others were cast in 2015. Then heat treated with a 24 hour soak starting at a very high temp (can't recall what) and held for 12 hours and then slowly lowered over the last 12 hours. The 35 year old continued to move about as I scraped it. Whereas the heat treated seemed to be stable. Time will tell of course but it does make me wonder what is happening to the castings out in the rain? I am confident only in that they rust and any hard skin gets broken up as a result but can anyone confirm that iron actually de-stresses out in the rain?

Steve

[Niels AbildgaardParticipant @nielsabildgaard33719]

Posted by SteveI on 20/08/2017 18:05:04:

Joe,

Your Encomat example is indeed comparable to the manual Hardinge lathes as they too are mounted to the cabinet on 3 points; 2 at the head stock end and 1 at the tail stock end. In this instance the whole subject becomes simplified as a 3 point approach is optimal and optimal results are achievable with such a design on a very rigid short bed tool room lathe. The HLV-H's are up around 1000Kg including the steel fabricated cabinet, a different beast when you consider it is basically a 10" (swing) x 20" lathe.

The issue at hand is how to level a lathe that has been built less rigid and which has not been designed with 3 points of contact.

On a side note Last year I scraped in some straightedges. One had been cast over 35 years ago and had been planed and then left in a barn. Others were cast in 2015. Then heat treated with a 24 hour soak starting at a very high temp (can't recall what) and held for 12 hours and then slowly lowered over the last 12 hours. The 35 year old continued to move about as I scraped it. Whereas the heat treated seemed to be stable. Time will tell of course but it does make me wonder what is happening to the castings out in the rain? I am confident only in that they rust and any hard skin gets broken up as a result but can anyone confirm that iron actually de-stresses out in the rain?

Steve

The beauty of my granite- flimsy cast iron lathe combination is that granite is in command of geometry and has been seasoned quite some years in changing temperature.

The moment my Boxford lathe bed left the grinding machine it was NOT straigth ,but it does not matter.

It has six machined flats on underside so three shimms will make it very straigth turning when keept by granite.

This is a smarter and cheaper way of making lathes than the way Schaublins and Hardies were made.

With laser aligning we can see micron deviations from ideal form.

[SteveIParticipant @stevei]

Niels,

Agreed. I like your approach. Accurate yet simple. I've found keeping the ambient temperature stable also helps. Not inexpensive though. How did you secure the bed to the granite? Epoxied inserts?

Steve

[Niels AbildgaardParticipant @nielsabildgaard33719]

Posted by SteveI on 21/08/2017 19:18:37

Niels,

Agreed. I like your approach. Accurate yet simple. I've found keeping the ambient temperature stable also helps. Not inexpensive though. How did you secure the bed to the granite? Epoxied inserts?

Steve

Bed and headstock are bolted with 3 pieces M12 allthreads through granite.

**LINK**

[SteveIParticipant @stevei]

Niels,

Interesting. Some of the schaublin small mills and lathes are on seemingly enormous cast iron cabinets for the size of the machine. As we know that cast iron is a "dead" material and dampens vibrations it certainly makes a difference. How does the granite behave?

Thanks,

Steve

[Niels AbildgaardParticipant @nielsabildgaard33719]

Granite varies much,but it has never been tried for church bells due to high internal damping.

The best material will probably be Dolerit or Diabas as used for real expensive monuments like the US Ivo Jima made from swedish Diabas.

My next lathe will get a bedmate of this beatifull ,black swedish stone if I can find some not to expensive.

The modulus of elasticity is higher(more rigidity) than any other easily obtainable rock.

[Michael GilliganParticipant @michaelgilligan61133]

In the early 1980s, Cranfield deveoped Granitan [effectively a reconstituted granite], and there are plenty of documented examples of its use … here is just one: **LINK**

Definitely worth a browse, if you're interested in the concept.

MichaelG.

.

P.S. … that hyperlink is rather long, but for the benefit of those who don't trust 'undisclosed' links:

https://books.google.co.uk/books?id=2rXtCAAAQBAJ&pg=PA85&lpg=PA85&dq=cranfield+granitan&source=bl&ots=h99I33spsR&sig=AM9hOQl_HUYdrebqTduSiChu0QA&hl=en&sa=X&ved=0ahUKEwiH0oPRrOrVAhVMaVAKHfbGCWIQ6AEILjAF#v=onepage&q=cranfield%20granitan&f=false

Edited By Michael Gilligan on 22/08/2017 08:56:51

[Douglas JohnstonParticipant @douglasjohnston98463]

I have my Myford Speed 10 lathe mounted on a thick concrete slab. The concrete was cast in its final place on top of a thick bench and the whole set up is very rigid. Granite is probably better than concrete but would be more difficult to handle and transport to the workshop.

Doug

[KWILParticipant @kwil]

For one of my lathes I built a wooden bench, but below the top surface there is "buried" a 3 foot long piece of 10" x 3" steel channel, this carries the lathe with steel bosses through the wooden top and provides extra mass to stabilise the lathe.

[Author]

Posts