Ball screw and electronic lathe conversion

[tonyfoale]

There are many ELS (Electronic Lead Screw) and Lathe CNC projects on the net. This post shows my take on it. It is a lathe conversion which I guess is between ELS and CNC. I just call it an electronic lathe.

Here is a list of the lathe's automatic functions in addition to retaining full manual control, albeit fly-by-wire;

Multipass turning, boring and facing in either direction.
Multipass RH or LH threading, external/internal, in either direction and taper threading.
Tapers, big to small and small to big
Curve machining, concave or convex.

I have just made a video to introduce a series of videos describing the whole build. You can find it here:

Enjoy the video and please like and share. Do not forget to subscribe to my channel and click the bell shaped icon to ensure that you get notified when the other videos in the series get published. Which will be located in this playlist;

https://www.youtube.com/playlist?lis...phh8KleeqPiqpz

These will show both the mechanical and electronic hardware construction and modification, as well as the control programming.

There is a bit more about this project in a recent post about the optimised threading that I programmed into the control Screwing without screwing up.

I did not start out to build this but the project just grew. Some time back I converted the cross slide ACME screw to ball screw. That is described in this previous HMT post
Lathe cross slide conversion to a ball screw

Click images for full size.

I was so pleased with the smoothness and precision that I decided to replace the main Z axis leadscrew also. The ball screws do not have half nuts so it meant that it would have to be motorised and that meant that I would need motor drivers and some form of control. I thought that if I was doing that to the Z axis I might as well do it to the X axis also.



Then I would have all the hardware needed for a CNC conversion except for a CNC controller which are available quite cheaply. I did NOT want a full CNC. I wanted to retain full manual control but with some common lathe tasks automated. The manual control is fly-by-wire controlled by MPGs (Manual Pulse Generators).

An Arduino was my first thought for being the controller brains but the base models could not produce a high enough pulse rate for the rapid movements or the coordinated movements of turning tapers or ball ends. I chose to use an ESP32 micro, which is like an Arduino on steroids. The ESP32 is circled below and the finger shows the small size of an amazing amount of computing power.



For entering the control demands I use a small and unbelievably cheap display, "LED&KEY" with buttons as shown here.
I was planning on using a touch screen display before I saw these. The buttons are very close together and could easily result in two being pressed together. To avoid that I made different length buttons shown in the second photo below.





Clough42 has a long series of videos about his ELS build and it was the first of these which alerted me to the "LED&KEY" devices. Thanks James, you unwittingly helped out. Here is his channel https://www.youtube.com/c/Clough42

[Toolmaker51]

Ohhh I like this.
Sure, I use lathes, ran many versions over the years. Many seem incomplete/ inadequate, or features mismatched to the footprint. Unsure why it's irritating to move a project between them to compensate that lack; we do it for other machines without thought. To me, billing as the 'universal machine tool', means virtually all inclusive. The only real limit should be swing, center distance, and beef to support full size material at correct RPM. Serious spindle bore? Just a fuzzy dream.

While not seeing my primary lathe [16 x 54] a candidate for tonyfoales control, only because the mechanical controls still impress hell out of me. But it's size and related tooling not ideal for certain work.
Judging machine tools, an exercise appraising 'bang for the buck' features, coupled with tooling, versus cash, a situation we all relate too. Been hunting the right 'toolroom lathe' a long time, so many haven't a decent taper attachment, a favored spindle, speed range, metric threading; yaddy-ya.

Long story, short...
This project opens a host of new candidates, including a spiffy P&W that never had a leadscrew. One puzzle of the missing piece syndrome.

[tonyfoale]

The only real limit should be swing, center distance, and beef to support full size material at correct RPM. Serious spindle bore? Just a fuzzy dream.

...... so many haven't a decent taper attachment, a favored spindle, speed range, metric threading; yaddy-ya. ......

Could not agree more with your limits. I have an MT5 taper in the spindle so the bore is large for the size of the machine.
Theading with the conversion is so easy. It doesn't care whether it is metric, imperial or other strange units. it is just a question of specifying the pitch. So in metric if I want a M1.5 thread I just enter 1.5, if I want (doubt that I ever will) a 16 tpi then that is a pitch of 1.59 and so I enter 1.59.
In fact in terms of data entry there is no difference between setting turning feed and thread pitch. When threading the only difference is that the Z axis is advanced on each pass so that it follows the thread angle. 60 deg needs an advance of 0.577 times the depth of cut for that pass. When doing multiple passes the depth of cut for each pass is automatically set such that each pass removes the same volume of material.

[DIYer]

Thanks tonyfoale! We've added your Electronic Lathe Controller Conversion to our Lathes category,
as well as to your builder page: tonyfoale's Homemade Tools. Your receipt:

Electronic Lathe Controller Conversion by tonyfoale

tags: lathe, electronics, controller

[Toolmaker51]

.....snipped.....So in metric if I want a M1.5 thread I just enter 1.5, if I want (doubt that I ever will) a 16 tpi then that is a pitch of 1.59 and so I enter 1.59.

I'm enthused with threading prospects just like that, figuratively speaking. Repair work can entail need of reproducing, more often only approximation of proprietary diameter and pitch combinations.
Now for CNC, not critical; but no way will I tie up funds and floorspace for those eventualities. Especially, when those cases demand few actual parts. Other features are handled as always, form tools, radiusing attachments, referencing stops and so on.

[olderdan]

Tony, this is just brilliant and typically neatly done. Retaining the manual function is a real bonus, you have found the best of both worlds. Your video tutorials are among my favourites, straight to the point and thoroughly explained.

[tonyfoale]

Tony, this is just brilliant and typically neatly done. Retaining the manual function is a real bonus, you have found the best of both worlds. Your video tutorials are among my favourites, straight to the point and thoroughly explained.

Alan,
Glad you like it.
I am still getting used to the new way of working with it. I am also updating the software with extra features as they come to mind. For example yesterday I needed to do some knurling and realised that I could that with less strain on the lathe by screw cutting with an appropriate pitch from right to left and then, instead of retracting the tool and moving back to the start in rapid mode, using the same pitch and DOC to cut from left to right.
That only took a minute to add.

[tonyfoale]

I'm enthused with threading prospects just like that, figuratively speaking. Repair work can entail need of reproducing, more often only approximation of proprietary diameter and pitch combinations.

You might like a feature that I have built into the software. For multi-pass threading if you use the same DOC (depth of cut) for each pass then you will remove more material on the final cut than on the first one. This is obviously not optimum because it means that the initial cuts are not very productive if you set the DOCs to suit the final heavier cut. I have programmed in an optimum DOC such that each pass removes the same amount of material. So the first DOC is greater than the final one but both remove the same.

This only applies to normal triangular profile threads, rectangular shaped threads need the same DOC for each pass. In addition the tool is advanced along the Z axis for each pass to force the cutting to the leading edge of the tool, just as you would do manually with an angled compound.

Here is the derivation for the algorithm that I included in the software, it assumes that the thread depth is small compared to the workpiece diameter.

[albertq]

Your electronic lathe is brilliant. It looks so simple which is a sign of good design.

[tonyfoale]

Your electronic lathe is brilliant. It looks so simple which is a sign of good design.

Many thanks Albert, coming from you that is praise indeed.
I like simple and always strive for it.