PDA

View Full Version : Why are lathe ways shaped the way they are?



SmoggyTurnip
01-24-2009, 10:05 AM
This is the shape of my lathe ways:

http://i513.photobucket.com/albums/t337/SmoggyTurnip/Ways.jpg

Seems to be a pretty common shape.
Seems pretty complicated.
Why?
Why so many surfaces? Why is one side higher than the other? Why is one side larger than the other. Why is there 2 seperate sides at all, Why not just have one flat top and 2 perpendicular sides?

tony ennis
01-24-2009, 10:08 AM
Wow, what a can of worms you've opened.





*makes some popcorn and gets comfortable*

A.K. Boomer
01-24-2009, 10:14 AM
Its like bring up comparisons in boat hulls:p

loose nut
01-24-2009, 10:31 AM
There are many way configurations. Flat, inverted vee or a combination of both and don't forget the round bed ways or the bar beds like the Barker, Unimat and the triangular bed lathes.

People have been arguing the merits of bed configuration for at least 150 years, each type has its advantages and disadvantages. Inverted vee type tolerate wear better, holding there accuracy but some people think flat beds are more sturdy and less prone to having the saddles lift under heavy loading. Bar beds like the Drummond Round bed were used to make lathes cheaper so amateurs like us could afford a lathe but they had trade offs in use.

It's apples and oranges.

Check out books like "Lathe Design and Construction and Operation", Lindsay Press sells it but you mite be able to download it off of the Internet archive or google books for free. It's old but still relevant on this subject.

John Stevenson
01-24-2009, 11:01 AM
A lot of it has to do with hereditary design, the country of origin and the price of kippers.....


.

SmoggyTurnip
01-24-2009, 11:06 AM
On the shape that I have shown, how many of the surfaces are actually in contact with the carrage? They can't all be... can they?

aboard_epsilon
01-24-2009, 11:13 AM
One inverted v is for the carriage the other is for the tailstock.

They are great until they wear ...then you have to be a master craftsman to get them right again ..

The all flat ways Machines are much simpler and could be put right with the minimum of specialised tooling when they wear out.

all the best.markj

moldmonkey
01-24-2009, 11:13 AM
Usually on the style in your picture, the carriage uses one vee and one flat and the tailstock uses the others.

brian Rupnow
01-24-2009, 11:28 AM
And the machined underside towards the center is where the carriage lock rides against.

lazlo
01-24-2009, 11:32 AM
The all flat ways Machines are much simpler and could be put right with the minimum of specialised tooling when they wear out.

Oh, I thought it was because British engineers hadn't discovered prismatic ways yet? :D

Peter N
01-24-2009, 11:44 AM
They're lips to stop your files, screwdrivers, and hammers falling off the edge, which happens on flat slideways.

Peter

Spin Doctor
01-24-2009, 11:49 AM
One option I really like in the later LeBlonds was the replacable front way. Either the whole way could be replaced or it could be taken off and reground. Of course either way you are looking at scaping the saddle . One thing to consider is that for the most part the tailstock ways will wear fairly evenly as when not in use the TS tends to get pushed to the far end of the ways while the wear on the saddle ways tends to be concentrated towards the headstock

dp
01-24-2009, 12:01 PM
You left out the bottom ways for the carriage. Your diagram shows distinct ways for the carriage and another for the tail stock. The shape solves the requirements for good four point surface contact, good surface area without being excessively wide (pyramid shape), side force resistance, and are self-aligning for the carriage. The structure between the rails provides rigidity for torsional forces. It is also easily reproducible by casting, and is maintainable through resurfacing.

It's but one of many designs, of course, but they all have similar objectives.

Teenage_Machinist
01-24-2009, 01:05 PM
To me it seems like an angle way and a flat way would be good, not a vee way and a flat way.

spope14
01-24-2009, 01:07 PM
As mentioned, the front V is for the carriage and the flat on the opposite side also guides the carriage. The flat way on the front guides the tailstock and the v on the back side also guides the tailstock. The "v"'s keep things aligned and straight, the flats allow for glide and pressure. The carriage glides on its own set of V and flat and the tailstock the same to minimize the wear that would occur if they rode on the same way V's and flats.

Just what I learned......

John Stevenson
01-24-2009, 01:37 PM
Oh, I thought it was because British engineers hadn't discovered prismatic ways yet? :D

Read Whitworths history of the British machine tool trade with reference to Mausdley

The US trade at this time was still using wood and buffalo hide [ although not much changes :D ]

.

Evan
01-24-2009, 01:58 PM
American Pacemaker has replaceable hardened ways as well. I like the round rod ways as they are trivial to replace. I can replace a pair of ways on my mill in less than 30 minutes and don't even need to realign anything.

Evan
01-24-2009, 02:26 PM
The US trade at this time was still using wood and buffalo hide [ although not much changes

Wood is superior to steel and iron in several areas.

Plate buckling resistance, column buckling resistance and composite structure stiffness as well as strength to weight ratio. In the first three categories it may surpass or equal steel for the same volume, not just the same weight. That's why it was used in aircraft. The main reason it isn't still used is because of weathering in outdoor applications.

Teenage_Machinist
01-24-2009, 02:29 PM
Ideal would be a flat way and a halfround inserted way. This is less over constrained than a vee and a flat. Also the halfround way could be in a seat.


I sometimes wonder why nobody has tried Tin coating ways, or ceramic inserted ways on scraped or ground seats. It works for calipers.

I guess there are a lot of anti friction coatings.

spope14
01-24-2009, 02:47 PM
Bridgeport chrome plated their for a while - I have two Bridgies with this. TIN may not have the "slip" factor needed in ways. Calipers may not need the slip that a way set needs, and i wonder about the overall wear resistance of Tin? Sounds like a promising idea though!!!!!.

I also have two Mitutoyo sets of tin coat calipers - VERY NICE!!!!!!

SDL
01-24-2009, 02:53 PM
A lot of it has to do with hereditary design, the country of origin and the price of kippers.....


.

Wrong forum for Kippers:D

Steve Larner

Teenage_Machinist
01-24-2009, 02:55 PM
One might imagine layers. If tin could be put on easily:

Scraped ways.
Tin Coat- Easy wear quantification. If the tin is gone, its worn.
And then a super slippery coat.

There are ceramic coated linear shafting.


I like kippers!

John Stevenson
01-24-2009, 03:05 PM
Already done, do a search on Turcite

.

Teenage_Machinist
01-24-2009, 03:09 PM
So it's like a bronzoteflon.
I heard of turcite and seen it advertized but did not know that it had bronze. That presumably helps the wear but seems a bit abrasive.

S_J_H
01-24-2009, 03:29 PM
Check out books like "Lathe Design and Construction and Operation", Lindsay Press sells it but you mite be able to download it off of the Internet archive or google books for free. It's old but still relevant on this subject.

Good advice. There are several very good old books dedicated to lathe design and you can find most of them for download free.
I have read several of them and for me at least, very interesting reading!
In the hey day , there was a lot of competition and many thought they had the best designs. As the years went by some were proven better than others from a wear and manufacturing standpoint but most all of them work.

Steve

John Stevenson
01-24-2009, 03:29 PM
That presumably helps the wear but seems a bit abrasive.

So are some posters :D :rolleyes:

Norman Atkinson
01-24-2009, 03:33 PM
You rang me Lord?

lazlo
01-24-2009, 03:33 PM
So it's like a bronzoteflon.

That's a great expression Ian! :)

Turcite, Rulon 142, et al are bronze wear strips laminated with Teflon. Bronze has about 1/3 the coefficient of friction on cast iron, and Bronze laminated with Teflon is much lower still. Turcite is normally used on the female side of the way, and because the Teflon is compressible, you can actually run the saddle with nearly zero clearance.

Many (most?) modern machine tools (like the Hardinge HLV-H) that still have dovetail ways use Turcite, because it dramatically reduces drag.

Paul Alciatore
01-24-2009, 03:40 PM
Perhaps a few of the "whys" you wanted.

In engineering, form often follows function. In a lathe or other metal cutting machine, it almost always does.

The first consideration is that the three primary elements of a lathe (headstock, tailstock, and carriage) must be mounted on and two of them must travel along straight, PARALLEL paths. And this must be maintained while considerable and DIFFERENT forces are acting on each of these.

The headstock is the least difficult as it is generally mounted once and never disturbed. It only needs to be aligned initially and unless there is some breakage or unusual accident, it stay put.

The tailstock must be movable but is locked in position for any particular operation. The forces on it are primarily along the axis and upward. The upward force comes from the cutting force acting on the work piece: it pushes down on the tool and carriage and up on the work. The axial force can come from tightening the center into the work or from a drill mounted there being forced into the work. A very important consideration for the tailstock is that it be held down. This is accomplished with a clamp between and under the front and rear ways. This clamp explains the open area between and under the ways.

The carriage must move back and forth along the ways while cutting. It must not change position in the other directions and should not rotate in any manner. All this while variable cutting forces are imposed on it.

On the style of ways you show, the headstock and tailstock are usually mounted on the inner ways. These will have the least wear so the alignment of these two elements will be maintained better than that of the carriage.

The outer ways are usually reserved for the carriage. As stated by others above, it will produce more wear and more irregular wear.

As for the shapes used, they are well suited to the lathe's needs. Many machines, like mills, use dovetail ways. Dovetail ways are good at resisting forces in all directions: up, down, fore, aft, and rotation. However, dovetail ways can not be adjusted for equal tightness at all places along their length after unequal wear has occurred. They will be tighter at the ends and looser in the middle.

In a lathe, we do not need equal amounts of resistance in all directions. The cutting tools are mounted above the front way and the cutting force is directed almost straight down toward that front way. Fore/aft forces are a lot smaller and there is no upward force. The cutting force tends to rotate the carriage about the lathe's main axis and a tool mounted to one side could produce rotational forces around afore/aft axis. But there is little force to rotate about the vertical axis.

The inverted Vee way will take care of all of these forces. It has an additional advantage in that the downward cutting force forces that presses against the front Vee holds it in intimate contact at ALL positions along it's travel. This Vee will then lock the carriage position in two degrees of motion and all three rotational axies. Square ways would not exhibit these self correcting characteristics and would show differing amounts of play at different positions on worn ways.

The rear way is a plain flat. This is sufficient as it has a lot less force on it and is used mainly to prevent rotation about the left/right axis. If a vee or other shape was used here, it would be harder to make and more expensive. Aligning two Vees would be much more difficult. So one Vee per set is sufficient.

The flat below and in the rear is used to hold the back of the carriage down against rotational forces and the plate that rides on it should be adjusted for best operation.

Having two sets of ways serves another purpose (I spoke about the wear thing above). The headstock and tailstock are locked down so the narrower, central ways can serve their needs well. But the carriage must move and still resist all the forces on it. The wider that it's support points are spread out, the more stable it will be (remember the ads for the "Wide Track Pontiac"). I don't know about your carriage, but the one on my SB sits on the outer ways and has extensions to the left and right. Both of these give it a larger footpring and more stability. This larger footprint on the ways will also help average out any local errors in the ways better so any unwanted change in position or angle will be lessened.

Other forms of way, like the round rods of my Unimat, have even fewer of the desirable characteristics for a lathe.

Another thing you can observe is that the Vee/Flat way design is not very good for other kinds of work, like milling. Milling can produce forces that act in many different directions so the lack of upward restraint of the Vee way is a definite disadvantage. This is entirely born out in my efforts to mill on my SB lathe which has much the same shape as you have drawn. Although it is possible to mill in the lathe, it is a lot more difficult than on a mill and a lot slower. Cuts must be shallow and feeds slow. Also every effort must be made to direct the cutting force downward as a upward force on the carriage will lift it and cause a lot of clatter.

All in all, form does follow function.

lazlo
01-24-2009, 03:48 PM
Excellent post Paul!

S_J_H
01-24-2009, 04:41 PM
Some pics, we all love pics right?
My SB9A which has triple inverted V ways and one flat. The front way is worn yet it still cuts amazingly straight. Having 2 inverted V's For the carriage meant more work. I'm not sure why South Bend went to the trouble. http://i109.photobucket.com/albums/n48/S_J_H/misc/lathebed004.jpg

My Artisan 11x24. This machine is just unusual all over. The ways are dovetailed on the outside and flat on the tops and insides and all surfaces are used.
The carriage has a tapered gib for the flat inside front way and a another gib for the far end of the saddle in a dovetailed manner.
It uses both dovetails, top flats and the flat side of the inner way. Very rigid setup, would have no problem doing milling with this machine and since this machine has not much wear it's still very accurate.
The tail stock is a very accurate sliding fit between the ways and uses the outer dovetail and inner side to locate it.
http://i109.photobucket.com/albums/n48/S_J_H/misc/lathebed001.jpg
http://i109.photobucket.com/albums/n48/S_J_H/misc/lathebed002.jpg
http://i109.photobucket.com/albums/n48/S_J_H/misc/lathebed003.jpg


I'm very temped to put a VFD and modern belting system on this old lathe and put her back to work.
Steve

S_J_H
01-24-2009, 04:55 PM
one more pic of the Artisan tailstock to better see how it was kept aligned, inside flat and angled outer dovetail with an adjustable gib strip-
http://i109.photobucket.com/albums/n48/S_J_H/misc/tailstock003.jpg

.RC.
01-24-2009, 05:30 PM
Some pics

First lathes with 1 V and 3 flat ways

http://img.photobucket.com/albums/v606/OzRinger/a0ed_3.jpg

Note, Dean, Smith and Grace went to using this configuration for their modern manual lathes like here

http://img.photobucket.com/albums/v606/OzRinger/DSCF0556.jpg

I won't bother with a picture of the most common type of two v's

Next three V ways like this Reed and Prentice lathe

http://img.photobucket.com/albums/v606/OzRinger/reedprentice2.jpg

Four V ways like this Pacemaker

http://img.photobucket.com/albums/v606/OzRinger/16x102american.jpg

.RC.
01-24-2009, 05:30 PM
Dovetail ways like this Hardinge

http://img.photobucket.com/albums/v606/OzRinger/Hardinge1982-4.jpg

Round bar ways like this Unimat

http://img.photobucket.com/albums/v606/OzRinger/um3.jpg

Drummond lathe ways http://img.photobucket.com/albums/v606/OzRinger/img8.jpg

Flat ways http://img.photobucket.com/albums/v606/OzRinger/myfordtk.jpg

loose nut
01-24-2009, 06:43 PM
Seems like this thread should be renamed "AnyWAY you like it".

As long as the machine gets the job done it really doesn't matter what type of ways they have.

Evan
01-24-2009, 09:39 PM
Hey, that's my Unimat.

wierdscience
01-24-2009, 09:42 PM
Then there are Graziano lathes,which are in a class by themselves-

http://www.lathes.co.uk/graziano/index.html

These lathes had several unique features,first was the double height ways,one lower heavy set that the carrage rode on and one upper lighter set the tailstock rested on.

The second was the gap bed that allowed the machines to have a gap bed,without losing rigidity.This was achieved by stopping the upper set of ways short of the chuck by one carrage width.

The third feature was that the lower ways were covered and out of the way of swarf chips and dirt.

http://img.photobucket.com/albums/0903/wierdscience/img1.gif

Spin Doctor
01-24-2009, 10:59 PM
Well Hardinge on the old TLs http://www.lathes.co.uk/hardinge/page8.html that preceded the HLV used a set of dovetail ways for the saddle and another set of ways that matched the old spilt bed for their turret and plain turning lathes. Then there are the Wilson Slant Beds http://www.lathes.co.uk/willson/index.html as on the UK lathe site. Leblond with their replaceble front way had two flat ways for the saddle but the front one was rotated maybe 30D on its long axis. Unfortunately the tapped holes where not on the centerline and perpindicular to the mounting surface. Then you could of just swapped them end for end.

miker
01-25-2009, 02:17 AM
The real reason for the design shown by the Original Poster, is that after watching the English, Americans, Europeans etc. srew it up, the Chinese made a Quality design that we are all fortunate to be able to purchase today.

Now no more telling Porkies!!!

Rgds

SVS
01-25-2009, 06:08 AM
A friend of mine has a near look-a-like to that Graziano. We've questioned why Graziano gave up so much potential swing with the raised tailstock ways, but I now see that his way covers are missing.

I like the idea, but not the execution-Takes a potential 24" swing down to 18" or 20".

.RC.
01-25-2009, 06:22 AM
Hey, that's my Unimat.

I stole that pic fair and square....In fact I have had it on the HDD for years.....

Evan
01-25-2009, 07:24 AM
No problem. I wish the bench still looked like that though. :rolleyes:

winchman
01-25-2009, 07:46 AM
"They are great until they wear "

They work pretty well after they wear, too. Because of the geometry, it takes a lot of wear on the ways to make a significant difference in the accuracy of the lathe.

Roger

Teenage_Machinist
01-25-2009, 01:33 PM
That is WAAAYYYY out htere ! :D

That would be nice though to cover the cairrage ways:eek:

wierdscience
01-25-2009, 02:34 PM
A friend of mine has a near look-a-like to that Graziano. We've questioned why Graziano gave up so much potential swing with the raised tailstock ways, but I now see that his way covers are missing.

I like the idea, but not the execution-Takes a potential 24" swing down to 18" or 20".

Well,yes,but remeber lathes are limited to weight between centers as well.If you chucked something 24x80" it had better be aluminum:eek:

quasi
01-25-2009, 06:46 PM
Rivett 10-20 and 10-30's have removable ways made of tool steel. I love Monarchs but those American Pacemakers make me" Happy in my pants."

ckelloug
01-26-2009, 12:50 AM
Teenage machinist made a good but overlooked point which is borne out by MIT Professor Alexander Slocum's book Precision Machine Design. When designing a machine tool, Slocum teaches us that you want to use kinematic design wherever possible.

Kinematic design, as best I can summarize it without a mini-thesis here, is a type of design in which the reaction forces on the structure to loads can be predicted using simple physics rather than requiring a deep understanding of material properties and deflection computations.

A lathe with a large diameter round bar with a keyway and key would be a primarily kinematic design as the bar provides resistance in the y and z directions leaving the x axis free to slide while the keyway provides resistance to torque but has minimal effect on the constraint of the y and z directions.

Most designs are a compromise between truly kinematic designs and the opposite, overconstrained designs. In an overconstrained design, many machine elements attempt to constrain the motion in a given direction. This results in a competition between the parts to determine which one deflects the most.

A V and a flat way design on a lathe is a semi-kinematic design. The V resists y motion and torque about the y and z axes while the flat resists torque about x without forcing the carriage to deform if there is a discrepancy between the straigtness of the V way and the straightness of the flat way. In short, the V way alone determines the direction that the x axis will move in while the flat merely keeps the carriage from falling off and resists cutting forces that apply a torque about the x axis.

Various designs have been tried and are preferred by different manufacturers and almost all designs are a compromise between truly kinematic design and other requirements.

Evan
01-26-2009, 01:08 AM
I read his lectures and used a lot of his ideas in the design of my milling machine. For instance, if the design has the potential to be over constrained as my dual round rod ways do then it is a very good idea to make the carriage bearings at the corners of a golden rectangle or greater. This avoids cocking and jamming during translation. On the other hand I am making my lathe just adequately constrained with the rear way being just a flat bar with a underhanging gib to prevent lifting. All axial and radial constraint is provided by the fully supported front round rod way which is essentially the same as a vee way with an underside gib. The round rod in this case is used because it is a lot easier to obtain the required precision without resorting to time consuming and difficult to implement alternatives that require scraping or grinding.

lazlo
01-26-2009, 11:24 AM
When designing a machine tool, Slocum teaches us that you want to use kinematic design wherever possible.

Most designs are a compromise between truly kinematic designs and the opposite, overconstrained designs. In an overconstrained design, many machine elements attempt to constrain the motion in a given direction. This results in a competition between the parts to determine which one deflects the most.

There's a whole chapter in Moore's "Foundations of Mechanical Accuracy" about kinematic constraints of machine ways, and the various design trade-offs. Moore points out that one prismatic way and one flat way is ideal, because it's not overconstrained, but that the problem with that design is that the carriage will rock when a non-symmetric thrust load (the leadscrew) is applied.

That's why Moore went with the overconstrained double V-ways: because a Jig Boore needs absolute positional accuracy, any radial thrust loads are auto-centering.
The downside is that, because the system is overconstrained, Moore had to use an extraordinary amount of hand-fitting to get the two V-ways aligned perfectly.

The same effect applies in a much bigger way to a lathe, where the leadscrew is outboard the nearest way. So an extreme amount of non-symmetric thrust load is being applied to the carriage: the leadscrew is pulling on the carriage at the halfnuts below- and forward the saddle. An elegant solution is to place the leadscrew in the center of the bed, like some of the Schaublins do, but that makes the overall design of the machine much more complex:

http://www.lathes.co.uk/schaublin/img1.gif

Teenage_Machinist
01-26-2009, 11:59 AM
I assume dovetails are somewhat overconstrained?

ckelloug
01-26-2009, 12:26 PM
Dovetails like on the Hardinge are overconstrained. Overconstrained designs can work very well but they require a more careful analysis and or testing to ensure that they are doing the right thing.

tony ennis
01-26-2009, 12:41 PM
kinematic design

Never hard of this before. You could get lost in that.

Evan
01-26-2009, 05:53 PM
Anything that moves has six possible motions. Three translations and three rolls. A lathe or mill carriage needs only one degree of freedom, the other 5 must be constrained. A true kinematic design constrains each of those 5 degrees of freedom only once. The problems arise when a particular degree of freedom is constrained by more than one mechanism.

For example, my milling machine has two round rod ways and the carriages are constrained from rotation and translation in undesired directions twice or even four times because there are four linear bearings spaced apart on two ways. It is a highly overconstrained design. This is not inherently bad but it requires close attention to the error motions that will always occur in any design. In the case of parallel bedways the main consideration is parallelism of the ways. The deviation from parallel must be smaller than the clearances allowed for the linear bearings. If it is not the system will bind. Assuring very accurate parallelism can be very expensive. In the case of my mill it was very easy. The end plates that support the ways were line bored together at once in one setup. Absolutely equal spacing at each end was assured and so was parallelism.