Portable Production Collet Lathe Build Thread

[briuz]

Hello all,

I've posted a few pictures in the "What Did You Machine Today?" Thread thread (https://bbs.homeshopmachinist.net/fo...-machine-today) and I was asked to post a build thread. While this is rather belated as the lathe was started about 2 years ago today, here it goes.

The shop I have is very small and hard to reach as it's in my basement. Some years ago I was very, very busy making and selling a product and would have loved to have a production lathe that could be moved into my shop, make small parts, have a 5c collet, and didn't take up much space. At the time, the Tormach Slant-Bed lathe wasn't yet released. Even if it was, it would have been a little large for my needs. So, about 2 years ago, we started this lathe build with the goal of having the following features in as small a space as possible:

• Is small enough that it can be carried by two people.
• With the exception of a computer for CNC operations, there should be no other external components. The frame contains everything from the turret to the electronics.
• Has a 16 position turret for tooling.
• The spindle takes standard 5c collets with stock up to 1″ in diameter through the spindle.
• The spindle has an electrically operated collet closer that is controlled by the computer.
• Uses a common 115 VAC power cord.
• Has threading and rigid tapping.
• Uses a variable speed motor.

Originally, a small 2 cubic foot frame was built that was 2 feet by 1 foot by 1 foot with the idea to see if everything could be crammed into that space. Unfortunately, there are no pictures of that attempt. The electronics eventually ended up sticking out the left side as we ran out of space and the motor was too undersized to function. But, based upon what was learned, many parts were re-used to make version 2 of the machine. At that point, there are some pictures.

[briuz]

Below is an early picture of the second prototype. Here a 3/4 hp 3 phase motor is added on top to drive the spindle. The electronics are still in the original location of the first prototype and are rather miserable to reach. Visible is the second attempt at a collet closer which uses a ball bearings to lock the collet in the closed position. A stepper motor is used with a "ladder assembly" to lock and unlock the collet closer. The spindle bearing covers are 3d printed and use felt seals. The spindle encoder is driven by 3d printed gears which turned out to be noisy and had some backlash. Note that a 5c collet is shown on top of the machine for scale. The VFD turned out to be the wrong VFD for the job as it didn't allow for a braking resistor. So it had to be replaced. In this picture there aren't any handles as I figured we could just get some off-the-shelf handles and add them to the ends. (It turned out that off-the-shelf handles are mainly for drawers and are pretty expensive. It was cheaper to make integrated handles.) Version 1.0 of the turret is visible. Eventually, the turret had to be pulled and modified as it was not possible to locate the 3/8" tooling on center with that design.

[briuz]

Below is a close-up picture of the original belt drive and encoder. Partially visible is the Centroid Acorn that controls the machine. Also visible to the right is version 1.0 of the turret.

[briuz]

An early attempt at making a part with the machine. Picture is dated September 21, 2019. At this point, the spindle used felt seals and greased bearings. During development of prototype 3 the bearing lubrication was changed so that it now uses spindle oil.

[briuz]

As I stated earlier, running the encoder with the printed gears was noisy and had backlash. It seamed to work and the original plan was to eventually machine proper gears but it just didn't seem to be the best solution. So a new encoder mounting system was tried. This system worked much better, but took up a horrible amount of space and uses those expensive timing pulleys. Also, the pulleys needed to be balanced as they vibrated horribly otherwise. To balance them, I made up a U shaped partial "box" of sorts that had levelling feet. Then levelled it with my machinist's level on one of the milling machines. By doing so, I could put the pulley on a shaft and see which way it rotated to. Then, I drilled out some metal on the pulley and repeated the process until the pulley no longer would rotate to the heavy side. This worked. Also visible are those cheap chinese stepper motor drivers I started out with. It wasn't long until I had problems as they were skipping steps.

[briuz]

Three more pictures with a new front panel, some threads made on the machine, and where I started adding wire looms.

[briuz]

Here are some pictures from January of 2020. There is an overall view of the finished 2nd prototype, a pieces of stock in the lathe, a cut-off tool holder that I ended up machining, and one of the handles.

[briuz]

In March of 2020, it was realized the turret had problems as variations in insert tooling resulted in the tool height changing. Not being able to adjust the tool height was a major problem. So, arcs were machined into the existing turret and rockers were machined so as to be able to adjust the tool height. In the interest of not machining an entirely new turret, these rockers are on top do to space considerations.

[briuz]

At this point, Prototype 2 was at the end of it's life as it was clear that another machine was required. So, Prototype 3 was built. There are now integrated carry handles. Note that now "pads" are machined for the rails and there are extra threaded holes around the spindle so as to allow for accessories. Also note the extra hole in the lower part of the front panel for an accessory port.

[briuz]

A couple pictures showing how the spindle bearing races are aligned.

[briuz]

Here are four more pictures from May of 2020 showing a new front view of the machine, the new electronics layout with Gecko stepper motor drivers, the new spindle motor location, and the spindle nose. Everything is much cleaner and more professional looking.

[briuz]

Here is a picture from underneath the machine showing the new encoder. A magnetic ring with a read-head is now used. The timing pulleys and that big encoder assembly have been eliminated. Also, there is now a spot for the braking resistor as well as proper wire management. A custom machined spindle pulley is also visible as this pully not only is part of the belt drive but has the magnetic encoder ring integrated into it.

[briuz]

This is out of order as I just found these pictures. These are of prototype 2 where the stepper motor drivers were changed out from those awful chinese stepper motor drivers to the Gecko drivers.

[briuz]

Also, I decided a parts chute is required for a lathe of this type. It took several iterations until an acceptable parts chute design was created but here it is. The parts chute moves into place in about 5 seconds and is controlled by a TeensyLC microcontroller with a stepper motor driver and stepper motor. Great lengths have been taken to ensure that no air compressor is required to run this machine. The circuit board was prototyped with KiCad and milled on a small Sherline milling machine. A cover will be 3d printed for the electronics.

Near the top of the last picture there is a brass bar sticking out above the spindle. Setting up up to 16 tools in the lathe turret can be time consuming to say the least. So something had to be done. This bar holds a tool setter that also plugs into the accessory socket that the parts catcher is currently plugged into. Using Centroid's software, lathe tooling can now be setup rather quickly. Once the tool is installed and centered vertically, it takes only seconds to setup the tool.

[briuz]

The latest project has been to build another turret from raw stock. This is being done for several reasons. But one of the reasons is to see how long it takes to make a turret assembly. Also, the old turret is what I refer to as version 1.5 of the turret because it is the original turret and the rockers are upside down as it was modified. So, making a turret with the rockers positioned correctly seems to be a good idea. So, here is a picture of the latest turret.

A side project is to build a stand-alone lathe turret assemble that can be added to another lathe.