PDA

View Full Version : Lathe build: Part 1B, spindle housing



Evan
01-06-2009, 10:25 AM
I would have posted this on the first thread if it hadn't morphed into the Hydra. I cannot believe the number of views.

At any rate I have finished the housing for the spindle. To review some of the primary criteria for this lathe:

CHEAP
Big swing, 16" minimum
Make from materials on hand if at all possible without compromising performance
Use scrap and recycled materials as much as possible
Form follows function

Some secondary considerations:

Weight to be kept down so that it can be moved in sub assemblies by two people without special rigging
Explore the use of novel or unusual materials and geometries
Automation/electrification of some functions
Compatability with existing tooling and accessories
Discussion of headstock bearings is substantially complete

-----------------------------------------------------------

http://metalshopborealis.ca/pics5/lathespin1a.jpg

The housing is made from portions of two old automotive generators, one from my Land Rover since I replaced it with an alternator years ago. To insure alignment and strength the spindle nose bearing housing is threaded onto the smaller central housing. The visible threads may appear to be partial depth because they are filled with aluminum antiseize compound. The nose bearing is a light press fit into the housing. The brush ventilation opening will serve nicely to mount a hall encoder sensor to monitor the spindle rpm.

http://metalshopborealis.ca/pics5/lathespin1b.jpg

At the tail end is the drive pulley and a spindle extension. The spindle extension is drilled through with a half inch hole to permit the use of a tommy bar to assist in removing the threaded chucks.

http://metalshopborealis.ca/pics5/lathespin1c.jpg

I intend to buy an 8" 4 jaw chuck sometime in the future. It will become the more or less permanent chuck on this machine.

Here is something for some discussion: I am seriously thinking of using some composite construction the help keep the weight down. What I have in mind is to laminate jatoba wood with steel strap to make some structural members. Jatoba is one of the hardest of all woods and has excellent mechanical properties as well a very good damping properties. It is exceptionally dimensionally stable with a coefficient of expansion with temperature less than half that of steel. It takes threads well and is better worked with metal working tools than with wood working tools.

Has anyone tried something like this?

JCHannum
01-06-2009, 11:04 AM
I would suggest that something with more mass be employed in the headstock. The gen housing looks a tad lightweight for a headstock.

I don't know what you intend to use the jabota wood for, if for the bed, wood construction was employed early on on metal lathes and soon replaced with cast iron.

Lightweight subassemblies handleable by two men translates to something like 200# for me.

Evan
01-06-2009, 11:20 AM
Jim,

The spindle housing is to be mounted in a mounting frame for attachment to the headstock. It will be subtantially supported and restrained.

I was thinking of using the jatoba/steel laminate as cross bracing in the bed since the bed will be of built up construction. 200# is at the upper end of what I consider movable. I only have one way into my basement and that is a full length flight of rather steep stairs. There is very limited possibility for any sort of rigging at the top of the stairs because of the orientation of the entrances vs the stairs. It is a poor design.

I would like to thank you again for those threading dies you sent some time ago. They have been very useful for the threading portions of this project so far.

john hobdeclipe
01-06-2009, 11:54 AM
The problem with using any kind of wood is not thermal expansion but rather dimensional changes due to changes in moisture content. I haven't seen any numbers for jatoba...It's not in the charts in my 1961 Dry Kiln Operator's Manual.

But bear in mind that it won't take much of a change in moisture content of any wood to destroy the accuracy that you are building into this project.

That said, I'm following this project closely, as I'm thinking of building a spinning lathe, and have already acquired some rather massive laminated mahogany beams to use for the bed and possibly the headstock.

Evan
01-06-2009, 01:54 PM
Jatoba is very stable with changes in moisture. If well sealed it will be even more stable. More important is that the humidity here hardly varies and even then it only varies from about 20% RH to the highest that I have ever seen at 55%. This is a very dry climate, officially a semi-arid zone. Average precip per year is 15 inches. Most of that falls as snow when the humidity is bone dry.

noah katz
01-06-2009, 03:29 PM
Evan,

Jatoba sounds expensive, so I guess you have some on hand?

Even so, I'd think it more economical to ebay it and build the lathe structure out of steel for stiffness, and add some form of CSD (constrained layer damping), i.e., thin plywood adhered w/Liquid Nails for Subfloors (stays compliant).

Evan
01-06-2009, 03:47 PM
I have a nice plank of it that has had a good chance to dry to local conditions. I want to add some damping to the structure since it will be steel instead of cast iron. I can do plywood too. I have some nice baltic birch that will do for that. That does give me some ideas. I have about a half sheet of new 0.100" steel and about the same amount of 0.125 birch ply. A stack of maybe 5 layers of steel and 4 layers of ply would make a *VERY* strong and rigid structure and would weigh about 60% of what a solid steel plate would weigh.

topct
01-06-2009, 04:23 PM
Would you glue that stack together?

With what?

Evan
01-06-2009, 05:36 PM
I will use glue, probably contact cement after painting all layers. Then I would rivet it with tinner's rivets. Maybe then cut some lightening holes with the mill. I am seriously considering doing this. The lathe will have plenty of mass in the areas where it needs it. In other areas it needs rigidity more than it needs mass. The two are not synonymous.

TheAndroid
01-06-2009, 07:00 PM
It is exceptionally dimensionally stable with a coefficient of expansion with temperature less than half that of steel. It takes threads well and is better worked with metal working tools than with wood working tools.

Evan,
Does this mean the strap will expand at twice the rate of the wood? If so, what adhesive would allow this shear without failure and still provide the strength you obviously need? I would think that at the very least, the pieces would not remain in the same planes.

Keep in mind that I live in Texas and we have WILD temp changes here.

Evan
01-06-2009, 10:47 PM
It won't depend on the adhesive. I will use fasteners. The rate of change with temperature of steel isn't that large. It's less than brass, bronze, copper, aluminum and most other common metals. Jatoba is close to nil. The main thing is to always make the lamination system balanced so there isn't any possibility of making a giant bimetal thermostat strip. Those considerations apply to all machine tools and are especially important when different metals are used together.

Wood/metal laminated structures are used sometimes in aircraft. That's where I first ran into them. The floor panels on the Swiss Pilatus Heliporter utility aircraft are very thin aluminum with a core of balsa. They weigh almost nothing but can stand just about anything up to and including full fuel drums. Composite structures have much better strength to weight ratios than conventional single material structural elements. This is especially true of trusses and beams. Laminated carbon fiber is unbelievably strong for its weight with a really unusual degree of stiffness.

It would be really cool to build an ultra rigid lathe using only ultra light composite materials so that something the size of a SB9 could be picked up with one hand. Even more interesting is to implement active vibration cancellation so that chatter doesn't have a way to start.

quasi
01-07-2009, 12:45 AM
I have always been under the impression that lots of mass in a machine tool was necessary for rigidity and vibration dampening. I don't understand how making a sandwich of wood and steel would achieve this.

aostling
01-07-2009, 12:52 AM
Jatoba is one of the hardest of all woods and has excellent mechanical properties as well a very good damping properties. It is exceptionally dimensionally stable with a coefficient of expansion with temperature less than half that of steel.

The C.T.E. for wood is anisotropic, if this matters. According to http://74.125.95.132/search?q=cache:xi1DaTfKASgJ:www.fpl.fs.fed.us/documnts/fplgtr/fplgtr113/ch03.pdf+c.t.e.+wood&hl=en&ct=clnk&cd=1&gl=us


The thermal expansion coefficient of
oven-dried wood parallel to the grain appears to be independent
of specific gravity and species. In tests of both hardwoods
and softwoods, the parallel-to-grain values have ranged from
about 0.000031 to 0.0000045 per K (0.0000017 to
0.0000025 per F).

The thermal expansion coefficients across the grain (radial
and tangential) are proportional to wood specific gravity.
These coefficients range from about 5 to more than 10 times
greater than the parallel-to-grain coefficients and are of more
practical interest.

I expect you know this, just mentioning it in case you did not.

MickeyD
01-07-2009, 12:54 AM
That is cool, a lathe spindle cartridge. Wooden lathes were fairly common here in the states prior to the civil war. White oak was often used for the bed and head stock, and lignum vitae was used for bearings (still used in some industrial bearing applications). Now, most of these lathes were used to build artillery pieces, and were not the most stable in the world. Have you thought about using a cast iron engine block as the headstock? Something like an old four cylinder Volvo block is very rigid but can still be carried by one person. I just got a 2kw mitsubishi servo in tonight that is either going to go in a home rolled 4th axis or a dedicated cnc lathe, so I won't be too far behind you in building something.

2ManyHobbies
01-07-2009, 02:10 AM
It would be really cool to build an ultra rigid lathe using only ultra light composite materials so that something the size of a SB9 could be picked up with one hand. Even more interesting is to implement active vibration cancellation so that chatter doesn't have a way to start.

I follow this idea and 6-8 others come to mind at the same time.

I figure of active dampening on the ways, you'd need something akin to hydraulic tensioning which could be done with less mass if you could space and brace cross members. A single (or group of) 1" diameter piston(s) with a 1/2-1" stroke could be used to alter any resonance though the only trick would be to do it without changing the positions of the headstock, tailstock, and tool when it happens.

Bearing material for the ways shouldn't be too difficult either, but what happens when something does impact the bed? Blue and stone? Relaminate? Epoxy fill/resand/refinish? Could the saddle be made longer and have bearing parts ride under or into the headstock and around the tailstock? H-shaped vs square or rectangle when viewed from above?

Of course, you don't have to sacrifice mass to be lightweight and portable. A composite structure could be carried into position then mass added from a garden hose. Pop the drain plug(s) out while you pack tooling or remove the headstock. Any structural members that get filled could have baffles included for extra strength and be designed to limit any harmonic sloshing. A bucket of water grabber would gel everything, though flushing might take more work. In that case, some warm saline and a long plastic stir stick in a drill might really speed up any flushing efforts.

I'm sure this isn't the last related idea I'll have in the next 48 hours, but if you are building the composites yourself, you could include strain sensors to let the computer modify spindle and feed speeds on the fly. On the opposite side, you could load up a knurler when you are done for the day and let it sing you to sleep.

Now I'm going to be having strange dreams about an engineer core backpacking in a full CNC machine shop to somewhere insanely remote. :D

Doc Nickel
01-07-2009, 02:32 AM
Bearing material for the ways shouldn't be too difficult either, but what happens when something does impact the bed? Blue and stone? Relaminate? Epoxy fill/resand/refinish?

-I assumed metal (iron, steel, etc.) ways on a (proposed) composite machine. Use the metal for the bearing surfaces, build the structure from composites.

Although, the Moglice and Turcite epoxies work well as bearing surfaces- though I'm not sure how well it would work epoxy-on-epoxy, and not the usual epoxy-on-metal.

The idea of making it carryable is an interesting one. Barring that, it'd seem the "epoxy granite" technique would be perfect to add mass and damping to what would have been an otherwise fairly lightweight framework.

How about removable mass? Make the "pockets" with tapered sides, cast the epoxy-granite in place, pop 'em loose after they've cured, and fab up some sort of clamping mechanism to hold 'em in place.

Doc.

DickDastardly40
01-07-2009, 04:03 AM
My experience of single point cutting male threads in Jatoba was that it splintered quite badly. I expect a lot depends on the piece of wood, how it was dryed, grain position and orientation.

You are obviously content with how it takes the threads you want, I am interested in how you will form them.

I didn't try cutting with a die as my application required a 24mm x 3mm thread for which i didn't have a die. In the end I cut it with a 6mm pitch to half depth so it looked like a 2 start thread with the second start not cut.

Al

Evan
01-07-2009, 04:32 AM
As for threads I was referring to drilled and tapped holes. When ever I have to make threads in wood like that I soak the wood with cyanoacrylate glue first and let it cure. It makes for much better and stronger threads.

I am not yet sure where I am going with the wood idea as making laminate stacks would be pretty time consuming. I have a tendency to get carried away with that sort of thing so I might just drop some or most of the idea.

Mass does not equate to rigidity, pretty much the opposite. What more mass does is lower the resonant frequency of a structure. As the quantity of mass is increased by increasing the size or density of the mass the volume of the mass grows by the cube of dimensions while the strength grows by the square. Carried to the really large scale this is what limits the size of mountains on Earth and other planets.

Right now I am testing possibilites for the way system. Candidates include flat top ways like the Atlas lathes and precision ground round rod ways that are fully supported and use a 3/4 round linear bearing, most likely of cast iron. Probable is to use both, a round rod way for the front and to avoid over constraint and to eliminate one alignment axis use a flat way for the rear.

Bearing material for the flat way will probably be steel way with PTFE slider.

.RC.
01-07-2009, 04:58 AM
Evan have you turned all the parts on your lathe or are they cylindrically ground???

Evan
01-07-2009, 05:10 AM
All turned on my South Bend.

noah katz
01-07-2009, 02:54 PM
"I have always been under the impression that lots of mass in a machine tool was necessary for rigidity and vibration dampening."

Mass, stiffness (though rigidity is commonly used interchangeably w/stiffness, in engineering parlance rigid = infinitely stiff), and damping are three different parameters, and the ones required to describe the dynamic behavior of a mechanical system.

"As the quantity of mass is increased by increasing the size or density of the mass the volume of the mass grows by the cube of dimensions while the strength grows by the square."

This is true for tensile/compressive loading; for bending, stiffness increases with the cube of the height of a beam, and the 4th power of the diameter of a round bar.

So doubling the diameter of a solid round bar increases its mass by 8X, bending stiffness by 16X, and resonant freq by 1.4X.

Of course greater freq gains can be made w/hollow sections.

Evan
01-07-2009, 05:05 PM
I suspect you meant to say "decreases resonant frequency by 1.414...x" :)

Spin Doctor
01-07-2009, 05:11 PM
Here is something for some discussion: I am seriously thinking of using some composite construction the help keep the weight down. What I have in mind is to laminate jatoba wood with steel strap to make some structural members. Jatoba is one of the hardest of all woods and has excellent mechanical properties as well a very good damping properties. It is exceptionally dimensionally stable with a coefficient of expansion with temperature less than half that of steel. It takes threads well and is better worked with metal working tools than with wood working tools.

Has anyone tried something like this?

If you want to build up a solid wood housing using Jatoba look for deal on "Brazilian Cherry" hardwood flooring. Same stuff.

PS It looks realy great as flooring. Froms the specs I have seen it is just about 3 times as hard as Red Oak. Also would Baltic Birch plywood work also?

Evan
01-07-2009, 05:17 PM
I just did some testing to see how well it takes standard cap screw fasteners.

Amazing. I treated it just like aluminum, drilling and tapping using the maximum thread percentage tap drill and tapping with a spiral point tap. I tried a 1/4-20 and a 5/16-18. Across the grain it is about as strong as aluminum. In order to strip the threads with around 3/4 to 1 inch of thread engaged I had to use a cheater bar on the allen wrench and I really had to reef on it on the 5/16 screw. Also, with just a standard washer under the head it doesn't sink into the wood at all. It's a lot more like metal than wood.

noah katz
01-07-2009, 05:26 PM
"I suspect you meant to say "decreases resonant frequency by 1.414...x""

No; (bending) stiffness increases more than mass, so freq goes up.

John Stevenson
01-07-2009, 06:09 PM
It's a lot more like metal than wood.

Can you weld it :D

.

Evan
01-07-2009, 06:59 PM
I suspect you meant to say "decreases resonant frequency by 1.414...x""

No; (bending) stiffness increases more than mass, so freq goes up.


I was assuming an increase in all dimensions including length. In that case the resonant frequency goes down as mass goes up.


Lookie what I found in my junk pile. I totally forgot about this until I stumbled on it today. :D

http://metalshopborealis.ca/pics5/leadscrew.jpg

John Stevenson
01-07-2009, 07:21 PM
Lookie what I found in my junk pile. I totally forgot about this until I stumbled on it today. :D

http://metalshopborealis.ca/pics5/leadscrew.jpg

Hospital bed ?

At one of the shows a couple of years ago this guy had a big pile of ball screws, I mean BIG, 200 plus or so.
He was selling these for 25 each and doing well. I had a word with him and had a ring round then did a deal with him.

I bought 30 for 12 each, I had 5, 3 for an X1 conversion that never came off as we did the X3 which led to the KX1 and KX3 and 2 for the cross slides on two lathes, again not done as i bought some Denfords from a college.

The rest went to others I had contacted and they agreed to buy them.

These came off hospital beds and were brand new. Policy said 4 years and change them but these came out of some wards that were not even open !
Later beds had acme screws on to reduce cost.

So now the question is? where are those 5 ball screws ?

lazlo
01-07-2009, 07:59 PM
Hospital bed ?

That looks like the carriage drive for a 20 year-old printer.

Evan
01-07-2009, 08:18 PM
It's from some sort of precision linear actuator. The motor is rated at 36 volts dc which is an odd value but at 12 volts it still has a lot of push and is a perfect speed for long lathe cuts. I haven't tried it at 36 volts but I bet it will be perfect for traversing. I expect it will be ideal as an electric leadscrew for my lathe project.

wierdscience
01-07-2009, 08:41 PM
I have one identical,mine is a dish drive off an old 16' wire mesh dish.

Those will lift quite a load,something like #1200.

Threading wood,old trick to add strength to the threads is to dope the tap hole with super glue,then tap.The glue soaks in and binds the fibers across the open pores of the wood.

I tried it on some white oak once and a 1/4-20 gr5 bolt could be wrung off 60% of the time.

Peter.
01-07-2009, 08:52 PM
Old trick for mounting engine mounts on the firewall of model aircraft that. Drill the hole & soak in cyno.

Evan
01-07-2009, 09:00 PM
Yep, I mentioned that in post 18. :)

tony ennis
01-07-2009, 09:12 PM
using Jatoba .... "Brazilian Cherry" .... Same stuff.

Froms the specs I have seen it is just about 3 times as hard as Red Oak.

I used some Brazilian cherry (which is totally unrelated to what we in the US call cherry) for trim on a chessboard recently. It was easy enough to saw, sand, and hand-plane. I couldn't tell it from any other hardwood though I was using small pieces.

wierdscience
01-07-2009, 10:12 PM
Yep, I mentioned that in post 18. :)

I have an excuse,I've been welding aluminum since 8:00 this morning:D

darryl
01-08-2009, 04:42 AM
Ah, a Von Weisse actuator. Those were always my favorite. Dual ball bearing motor, smooth and quiet operation, fairly efficient. Dual needle roller thrust bearings- but some odd thread, like 6 tpi. I have two of those motor/gearboxes adapted to my mill for the x and y, and I use about 15 volts to feed them. They run just fine on 60 volts, but as you might imagine they turn a little quicker. At 15 volts there's enough torque to drive the axis against any cutter load I've been comfortable using. My TP grinder motor is made from two of those, with one can but both sets of magnets and most of both armature laminations. Was a fair amount of work to rebuild, wind, epoxy and balance the armature, but it sure is nice.

A couple thoughts on the wood- were it me I'd be laminating the pieces using coffee table epoxy. Two names come to mind- Envirotex, and Nu-Lustre. I haven't seen the former for awhile, but Nu-Lustre 55 is available locally at not too bad a price. It's tough, not brittle, not prone to chipping, and bonds well if it's applied within about 1/2 hr after mixing. My own feeling is that contact cement isn't the long-term adhesive to use.

Some years ago I built a micro drill press with an x-y table for pc board drilling. It was formed from mdf, then all corners rounded before laminating over the form with fiberglass and either epoxy or polyester resin, don't recall which. It is quite rigid and has good damping characteristics. I wouldn't hesitate to build that way again, and the ultimate strength depends largely on how many layers of glass is laid up on the form. Of course mdf isn't the ultimate choice for moisture resistance, etc, but Jatoba sounds like the ideal wood to build and laminate with. You're building with materials at hand, so maybe this avenue is out, but Jatoba and fiberglass might work optimally together- leaving steel out of the major structure for the most part. Threaded steel inserts could be bonded into it for fastening points where major parts assemble.

I could go on and on- I haven't followed the epoxy granite thing lately, but I did like that. The idea of using a granite surface plate as a base to build on also seemed good to me- epoxy the lathe bed parts directly onto it-

Evan
01-08-2009, 05:45 AM
I have decided not to use laminated construction at this point. I am going to use the Jatoba as main beams in the front and rear way structure but there will be plenty of steel in there too. After seeing how well the jatoba will hold fasteners I have no doubt it will work well and will provide an isolation function to the various structural steel portions. The way system will be made so that the front way determines alignment with the centerline and the rear way merely supports the rear of the carriage on a flat way. This avoids any issues with small changes in front/rear spacing or alignment. The way system is capable of actually standing alone without the jatoba beams.

When I design something like this I wander around my various storage areas and collect up everything I can find that looks like it might be useful. I then start mocking up configurations without actually cutting parts to size. I examine each configuration for advantages and disadvantages. I also depend heavily on a factor that I can't really explain. It has to look right. As I rearrange the parts the arrangements tend to converge on something that contains all the required functional elements AND for some reason is usually asthetically pleasing. I firmly believe that a good design will automatically look good too.

davidh
01-08-2009, 06:29 AM
evan, when do you sleep ?

Evan
01-08-2009, 07:55 AM
I don't sleep, much. I get at least 3 hours per day. I have a sleep disorder.

S_J_H
01-08-2009, 09:53 AM
When I design something like this I wander around my various storage areas and collect up everything I can find that looks like it might be useful. I then start mocking up configurations without actually cutting parts to size. I examine each configuration for advantages and disadvantages. I also depend heavily on a factor that I can't really explain. It has to look right. As I rearrange the parts the arrangements tend to converge on something that contains all the required functional elements AND for some reason is usually asthetically pleasing. I firmly believe that a good design will automatically look good too.

You have been at this hobby much longer than I, but that's pretty much how I went about building my little cnc lathe and how I am building my HBM/Lathe machine right now.:)

That's all a part of the "fun" for me with this hobby.
Steve

Evan
01-08-2009, 10:07 AM
Since I am doing this as a "public" build I think I'll take a few pictures of where I am going with the bed today. The bed is the next item to complete and will be the most work.

noah katz
01-08-2009, 09:42 PM
"I was assuming an increase in all dimensions including length. In that case the resonant frequency goes down as mass goes up."

Yes, in that case it does.

"Lookie what I found in my junk pile. I totally forgot about this until I stumbled on it today."

Wish I could, but for some reason on my work computer blocks some images.