CNC CONVERSION - 6" ATLAS LATHE

[RICH]

The following  posts in this thread will show how I am converting my 6” Atlas lathe for CNC use. I rebuilt this lathe a number of years ago and it is mint condition. One of the conversion criteria was to ability to return the lathe back to manual. I have a Sherline CNC lathe that is quite nice for making small parts, but, the need for something more robust and  yet small was desired.

There are a few things one needs to consider before doing a conversion. What kind of screws will be used and how will they be mounted.  Motor selection and associated mounting. 
I will making a lot of posts as i go along with the conversion. Hopefully there will be a little something for everyone.

RICH

[RICH]

The following are just some preliminary considerations for the conversion.

MOTOR
Currently a ½ HP / 1725 motor is mounted for use. An option would be to use an existing VFD with a 1 HP DC motor. I will first try the AC motor since backgears can be easily engaged for threading and heavier work. It’s very desirable to have a constant spindle speed for threading. This was one of my dislikes with the Sherline lathe. Regular turning on the Sherline was not a problem, but the lack of HP at the spindle during threading left something to be desired. I will probably be making comparisons to the Sherline from time to time. You can calculate the HP requirements but the calc’s can be  somewhat subjective, and if the motor is marginal, you really don’t know how it will behave under load and the resulting small rpm changes. Here’s a look at some of the Atlas lathe speeds.

ATLAS LATHE SPEEDS
M pulley  S pulley  SPINDLE    BACK GEAR  APPROX
                                RPM            RPM                  RATIO
1                 4           410                55                      7.45 
2                 3           870                120                    7.25
3                 2           1435              195                    7.35
4                 1           2350              320                    7.35

When the backgears are in use there is a considerable increase on torque delivered to the spindle.
½ HP x 7.35 = 3.7 HP ( approx ) which is a significant HP increase ( > 12X ) as compared to the Sherline.
RICH

[RICH]

BALL SCREWS

I already had ball screws. Both the ball screws are easily adjustable for backlash. The intent is to minimize the backlash ( you never have zero but you can try ). You can do all kinds of calculations for the ball screws, but practically speaking, experience on sizing and mounting  from past experience is all that is needed for this conversion.

The Z axis will use a 16mm diameter x 5MM ball screw. The X axis will use a 12mm diameter x 4mmm screw. You are somewhat limited on a conversion as to what will fit and yet the basic mechanical integrity for supporting them is very important.

One important criteria will be resolution ( just how small of an axis move can be made).
Steppers will be used since the existing controller is for steppers, is portable, and is used for other CNC machines. Here are a steps per unit calcs and a look at resolution:

Z Axis
5mm x 0.0394 mm/inch = .197” / revolution   ( 1/0.197 = 5.0761 PITCH )  along with a 2:1 pulley reduction
STEPPER   x  MICRO STEPS x SCREW PITCH x PULLEY REDUCTION = STEPS / UNIT
200 Steps / rev  X 10  X  5.0761 = 20304 STEPS / UNIT

RESOLUTION:  1/ 20304 = .00005”

Note that the mounting plate will be such that the pulley ratio can be changed as I may want
to increase the Z axis velocity later after testing.

X axis
4mm x 0.0394 mm/inch = .1576” / revolution   ( 1/0.1576 = 6.3452 PITCH )
STEPPER   x  MICRO STEPS x SCREW PITCH x PULLEY REDUCTION = STEPS / UNIT
200 Steps / rev  X 10  X  6.3452 = 12690 STEPS / UNIT

RESOLUTION:  1/ 12690 = .00007”

There are tradeoffs to made. I choose to do a 2:1 pulley reduction on the Z because I wanted more torque delivered while cutting threads. The X resolution is fine but doing a reduction makes the mounting somewhat more complex. Another consideration will be how much holding torque will be available. The Sherline never skipped when using 300 oz-in steppers and the steppers were direct mounted. I'll need to take a look at the speeds I will be running when doing heavier work along with threading considerations. 
RICH

[RICH]

AXIS VELOCITY

 A preliminary desirable velocity for rapids is 60 IPM. The Sherline normally ran at 40 . I won’t know just what kind of rapid movement I will get until the conversion is done since a lot of things come into play such as the stiffness of the axis system,etc. Calculations can be done to see the affects of velocity and axis loading as they relate to stepper motor sizing. The following data will be used later on when I look at stepper motor information.

60 IPM= 1 INCH / SEC
1 IN/ SEC x  20304  STEPS / SEC = 20304 STEPS / SEC = PULSES PER SEC ( PPS )
IPM    PPS
60       20304
30       10152
20        6768
15        5076
10        3384
5          1692
2.75        930

Figure #1  is a comparison of spindle speeds to axis velocity and motor PPS when doing a ¼ - 20 - UNC thread. Note that the motor will need to turn twice as fast as the ball screw ( 2:1 pulley reduction ) so the PPS is double of what was given in the other table.
RPM     IPM     PPS
600       30         20304
400       20         13536
300       15         10152
200       10          6768
100         5          3384
55        2.75        1860
Attached is a graph of  the above. Note that as the RPM goes up the axis feed rate increases along with
 the required acceleration during threading.
RICH

[RICH]

Figure #2 shows the decreasing stepper torque as speed / PPS is increased. Added to the motor curve is the RPM and IPM when threading for a ¼ - 20 thread. The graph is for the Z axis. Won’t know how much torque will actually be required until after installing and adjustment of the ballscrews are done. At least it’s obvious that with the back gears engaged, there will be enough motor torque to do the threading. Note that in using the SmoothStepper , velocity increases of 2X and greater were achieved on my other machines so speed is  will be increased relative to the torque. Also not that the motor curve is for a lower voltage than what will actually be supplied to the motor. Even though you have a motor curve, if the curve is not truly representative of your actual drive, current and voltage and wiring configuration, you can only use it as a guide for  anticipated performance. Knowledge from past experience really helps!

RICH

[RICH]

 Just thought it would be helpful to provide some of the considerations from a technical point of   view relative to the conversion. 

Why steppers instead of servos? Well it was a rather easy decision as the controller I built has drives for steppers and is used for other machines. Here is a picture of the movable cart I use.

RICH

[RICH]

Z AXIS BALL SCREW INSTALLATION

The first thing was to see if the ball screw would fit. I had a saddle from an old lathe that can be modified and thus no need for changes to the original. All the bosses were removed and some rough outside machining was done just to mount the nut and square things up. The carriage will have a housing on the front of it with two small hand wheels. The hand wheels will turn encoders for the Z and X axis. Things looked very close but looks like it will fit.  The screw was held in place parallel to the bed and Z axis and temporarily attached to the carriage. One hole was drill for the ball screw nut and alignment checked, and then the other hole was drilled.  In the figure #4 the ball was just rough aligned by using parallels at two ends to check the mounting holes. A little clearance is provided in the holes for later refinement when testing.
RICH

[RICH]

I'll continue to post more as time goes on. Feel free to ask questions or comment.
RICH

[RICH]

The attached pictures show the trust bering assembly being mounted.
Rich

[RICH]

THRUST BEARING
Two back to back bearings are used top take axial thrust. The bearings sit inside the housing. The end of the screw will be tapped for a bolt. The bolt when tightened will push against the retainer and thus provide for a preload of the bearing thus eliminating any movement of the ball screw along the Z. An additional bearing will be placed at the tail end. No machining of this end of the screw is required. It was a real bitch to machine the screws for my mill as the hardening went way down below the bottom of the screw lands. No fun, and heating / annealing the screw was not a choice and grinding it wasn’t an option. Figure #11 shows the mounted bearing assembly.
Note that the retainer will have a number of set screws which will tighten against balls and not directly against the screw. You can get away with this for smaller loadings.
RICH