CNC Lathe Build

UPDATE: I'm revving up with some winter time energy to get this lathe completed. I had set myself a 6 month timeline and I'm now 6 months over that target. Such is life. In updating the project thread today, I thought that the thread was disjointed and hard to follow. Our software allows for formatting of threads into an article format. I've done that here so that the thread can be read uninterrupted without comment. Any replies and dialogue follows the thread. Also, I've added some navigation so that readers can get around the thread a little easier.


Table of Contents
Design
Beam Structure
Alignment Washers
Welding the Bed & Beam Structure
Welding the Frame
XZ stage Motor Mounts
Walk Like an Egyptian
Tool Inception
Goto End

The gear hobber project is on hold for awhile. I've decided, for various reasons, that I need (no, no, no - not want, but need) a CNC lathe. I will spare you the long ranging logic behind my reasons for building a CNC lathe from scratch, but suffice to say it boils down to money and what I think I need.

Basic Requirements, the guys and conceptual design idea:
  • Minimum 1HP spindle
  • Potential for C-Axis control
  • AC Servo on XZ axes with beef to drive them
  • Software that has a good online community and staff support
  • NOT a small desktop CNC driven by small stepper motors
That's basically it. The design I've settled on relied on re-using some existing hardware. I had to search for what I thought would be adequate. The three main components for re-use I acquired are:
  • Hardinge HC headstock spindle
  • 7" dovetail bed to mount the headstock
  • Omniturn XZ stage
The headstock and bed were removed from an HC that was slated for the scrap yard. Those were free and only cost me elbow grease and lots of back pain.

View media item 97999
View media item 98000
View media item 98001
If you know anything about OmniTurns, you know they are solid, slant bed, gang tool CNC lathes and from what I'm told - workhorses. Omni-Turn sells retrofit kits for Hardinge HLV, HC, etc. The nice thing about this mix of components is that the Omniturn stage is designed to clamp to a Hardinge dovetail bed. The convenience of not having to worry about headstock and stage alignment makes the design concept work in my mind. As I think you will see through this build most of my inspiration for this build comes from what I have researched on OmniTurns, Hardinge CNC conversions and through Centroid, which is the CNC control board and software I chose to go forward with.

Goto Top
Here is the conceptual design:

View media item 98002
Headstock, bed and stage will be bolted to a welded "beam" structure made up of tubing and channel. That beam will then be filled with concrete for the purpose of vibration dampening. Then, the whole beam will be welded to the supporting frame structure below. Missing in the photo are leveling feet, panels, wiring, guarding, etc. - because I haven't designed all of that yet.

The bed is 43" long x 7" wide and ~3/4" thick. The frame is approx. 43" wide, 24" deep and approx. 36" high. I need to be able to fit this in my small shop. The footprint will consume a little over 2% of my shop floor - but it will feel like 25% of my shop if I get carried away with the design. Space is at a premium. So I may tighten up these dimensions while trying to increase mass and maintain stability.


Goto Top
Beam Structure Build

Materials used, 3" channel, 2" sq. x 3/16" tubing and a leftover 2"x5" steel tube from some long forgotten pack-ratty hoarding grab 20 years ago. I think it came from a semi-conductor machine I purchase before it was scrapped, I parted out a lot of the linear motion components that were mounted to this beam which was machine flat.

The 7" dovetail bed will be bolted through the channel webs with 5/16"-24 SHCS bolts. Lot's of fabrication work to do here and need to make sure things are nice and square, clamped flat.

Process Steps:
  • Cut materials
  • Tack weld channels to main beam
  • Layout dovetail hole pattern in channels
  • Drill 28, 21/64" clearance holes for bolts
  • Tack Weld up the assembly
  • Clean-up
Let's get started:

View media item 98003
View media item 98004
View media item 98005
View media item 98006
View media item 98007

Goto Top
Alignment Washers

The bed weldment is tacked up. The next task is to clamp the bed up to the weldment and then finish the welding. The problem with clamping now is consistent clamping pressure cannot be achieved with misalignment of the screw head on a 10 degree bearing surface. This next section is the tedious work of creating alignment washers that will allow me to clamp up to surface angled at 10 degrees. I hope you find this next section useful.

View media item 98008
View media item 99009
View media item 98010
View media item 98011
View media item 98012
Cutting, drilling, deburring and test fitting one piece is one thing, making 28 of the same piece is another.

The shape of the web makes them difficult to quickly reproduce 28 of these buggers. I made a quick and dirty fixture using my drill press vise and a cutoff from an old broom stick handle. Precise? No. Precise enough? Yes.

I thought this little picture summary might inspire those hobby machinists out there in that you can make fairly accurate parts with a simple operation. In this case, I used my angle grinder, a vise, file and drill press. Not glamorous, but represents what simple jobs can be done quickly for your shop needs.

View media item 98013
You can see the hold down clamp as a stop on the left of the vise. The angle is a front stop for the vise. And my quick and dirty broom stick fixture is a bottom stop for the clamp that I need to drill a hole through. The kant twists are just holding the drill press vise up against the stops and down to the table.

View media item 98014
View media item 98015
View media item 98016
View media item 98017
View media item 98018

Goto Top
Welding the Bed

Now that I can clamp the heavy, thick dovetail to the bed, next is to finish up the welding on this structural component. First is the bed pieces that I laid out previously. Two channel pieces, a 2x5 rectangular tube and two 2x2 square tubes make up this weldment.

View media item 98019
View media item 98020
View media item 98021

Goto Top
Welding the Frame

Main support frame is next. And then the bed will be welded to the top of that. 2"x2"x1/4" wall tubing was used.

View media item 98022
View media item 98023
View media item 98024
View media item 98025
@JimDawson - you are right! It rings like a bell! I'm going to fill the bed with concrete to dampen the frame.


Goto Top
Omniturn Motor Mounts (XZ stage)
The existing motor mounts are too small for the DMM 750W AC servo motors. I need to make up my own. 1/2" aluminum plate was used. I used the drawing specs from the DMM motor drawing and closely measured the holes on the Omniturn castings. Then, using those dimensions, sketched up some prints and used those to make up the motor mounts.

First mount will allow for some movement to adjust for belt tension. The other mount will be stationary allowing for no belt tension. I will have to add some sort of tension idlers at a later date.

View media item 98026
Once the centers were punched, I set out to remove the bulk of the material by drilling a circle of holes out of the middle of the part.

View media item 98027
Continued to layout hole positions around the circle.

View media item 98028View media item 98029
Once the bulk of the material had been removed, I was able to setup on the Atlas MFC and bore to size.

View media item 98030
After the Z axis mount was finished, I set out to repeat the process, albeit more complicated, with the X axis mount.

View media item 98031
The process for this was to spot and drill all of the centers, removed the circle of waste in the large bore, mount the blank to the stage and spot the bearing hole (second largest hole on the part) and then finish the pilot for the motor. I left out removing excess stock to bring the blank to shape because it deserves its own special image:

View media item 98032
View media item 98033
Goto Top
Walk Like an Egyptian (or moving very heavy things in simple ways)

The Hardinge chucker headstock, by my estimate comes in at a minimum of 150 lbs. It is VERY heavy and while I have moved it, I don't want to hurt my back (again). The easiest way to move the headstock onto the dovetail bed would be to use a hoist of some sort. I don't have that capability in my small shop. So, how do you install a 150lb. headstock onto a 7" dovetail about 36" off the ground...easily?

The following photos illustrate how I improvised some lifting, leveraging, pivoting, sliding and pulling movements into a scheme that would make a pharaoh proud.

View media item 98034
View media item 98035
View media item 98036
View media item 98037
Goto Top
Tool Inception - Episode 1

When you are eyeballs deep in a project, sometimes you need a tool you don't have. Or you have a tool and it doesn't quite do the job. Either way, you find yourself in a position of needing a tool, to make a tool, so that tool can make a tool. Eventually, you will get to the part you want to make, which is what this lathe is eventually going to do. My wife calls this, Tool Inception, aptly named after the movie about a dream within a dream - which is where my wife thinks my head is with these projects most of the time.

In this episode of Tool Inception, I needed to remove the flat belt pulley that is on the Hardinge HC spindle. The gear puller I have doesn't fit in between the the flange of the pulley and the opening of the casting. Time to make a tool to make a tool!

View media item 98039
View media item 98038
View media item 98040
I think I bought this gear puller either at Harbor Freight or an auto parts store. Hopefully the modifications I made here make sense. If anyone on Hobby Machinist ever needs to remove a drive pulley from a Hardinge chucker spindle, let me know - I'll mail the gear puller to you.


Goto Top
 
Last edited:
Wow!
Hey are you concerned that the weldment might not be perfectly flat and then might pull the base out of geometry? Do yo have a plan to get this flat?
Robert
 
Wow!
Hey are you concerned that the weldment might not be perfectly flat and then might pull the base out of geometry? Do yo have a plan to get this flat?
Robert
I'm not terribly concerned. The weldment will be clamped and welded as flat as possible. Any misalignment in bend (and I don't expect much if any) I plan to shim out. Once that is done any other alignment issues with the bed will be twist. The really nice advantage of using the clamp mounting style with the headstock and stage is that twist about the spindle axis doesn't really matter because the stage doesn't move once it is clamped onto the bed.
 
Cool. I didn't consider shimming the base. That could work. I am not sure I follow what you are saying about twist. Twist in the bed will move the headstock one way and the tooling the other way and result in cutting a taper? I suppose you could try to level out any twist depending on how it is mounted. I guess for short pieces this would not be an issue anyway. Will this machine have a tailstock also?
I am excited to see this project!
Robert
 
Cool. I didn't consider shimming the base. That could work. I am not sure I follow what you are saying about twist. Twist in the bed will move the headstock one way and the tooling the other way and result in cutting a taper? I suppose you could try to level out any twist depending on how it is mounted. I guess for short pieces this would not be an issue anyway. Will this machine have a tailstock also?
I am excited to see this project!
Robert
There are six possible degrees of freedom for misalignment. Looking at the headstock spindle from the tailstock: Left-to-Right bending, Up-to-Down and Axial Twist. In the case of the first two, I don't expect any Left-To-Right bending and I should be able to shim out any Up-Down Bending.

In the case of any axial twisting of the bed: even if the headstock rotated some amount about the axis of the bed, the stage doesn't move along the bed axis. It remains in one place since it is clamped to the bed, so any changes in any detectable twist in the bed are not translated to the axis of the stage. Hope this makes sense.

With all of that said, I don't expect there to be any real detectable alignment issues. Maybe that is my naivete speaking. We'll see! I'm wrong a lot!
 
Obviously you have thought about this. I think you will be able to correct for twist pretty easily even if you get enough to interfere with your desired tolerance.
Now you have me cruising eBay for lathe beds!
Robert
 
Curious to why you didn't use the original lathe base?

Hi Eddyde, the main reason is that I don’t have a good mode of transportation for hauling equipment, let alone loading and unloading. It certainly would have been easier.

There was also the other “stuff” that comes along with the rest of the machine. I don’t really have the time nor do I want to prioritize my time to part out turrets, handles, speed control gear, etc., etc. The other problem with taking the whole machine is that the 5hp 3ph motor was bad. I also did not want to get stuck with that. Finally, I just don’t have room for all of it in the shop.

At the end of the day, came down to where I wanted to spend my time.


Sent from my iPhone using Tapatalk
 
Ah, that makes sense. Reason I asked, I have a CNC lathe build project on the "if I live long enough" list. I was thinking of the same approach, using a Hardinge as a donor lathe.
Looks great so far!
 
Ah, that makes sense. Reason I asked, I have a CNC lathe build project on the "if I live long enough" list. I was thinking of the same approach, using a Hardinge as a donor lathe.
Looks great so far!

If you have the space, it makes sense to use the donor lathe.


Sent from my iPhone using Tapatalk
 
Back
Top