My future home-made lathe...

[bojan]

I simply couldn't resist the challenge when I saw similar modifications on google..
On the image, the guiding rods, tool holder and lead screw are missing (I am still thinking how to put all that together, so they are not mounted yet)..
It occurred to me that I could use a small worm gear for coarse movement, similar to this solution:
http://www.traderscity.com/board/products-1/offers-to-sell-and-export-1/hand-wheel-screw-jack-handle-worm-gear-jack-screws-174144/

However, it is not easy to find the suitable gear, so I decided to try and cut it myself, using the tool I made for threads for my 3D printer z-axis.
The tool was made from left over piece of lead screw (10x2mm), but since the nuts were made of delrin (acetal) there was no need for hardening the tool.

Now I want to harden it properly, to use it on brass and aluminium.
Does anybody know of the service (preferably in Melbourne) that could do it?
Or, in case I decide to do it myself, what would be the best way to heat up the tool?

Thank in advance!

[mswhin63]

Damn I sold my old Ryobi just the same. Maybe it could have served me well.

[bojan]

I didn't bother selling it when I acquired a bigger drill...

This is what I need:
http://www.cnczone.com/forums/linear-and-rotary-motion/110965-acme-screw-r-amp-p.html

And this is how some folk solved the hobbing problem:
http://pounceatron.dreamhosters.com/shop/hob/index.html

I will also try this (http://www.ebay.com.au/itm/2Mod-20-21-22-23-24T-Spur-Gears-45-Steel-Gears-Motor-Gears-Thickness-20MM/161893577154?ssPageName=STRK%3AMEBI DX%3AIT&var=460858865634&_trksid=p2060353.m2749.l2649)
If the gear mates smooth enough with lead screw (the shafts will have to be at angle, 90+6°), then I won't bother with hobbing (for now).

And of course, I will also need split nut, to be able to use lead screw for fine movement of the tool holder.

[ZeroID]

I considered doing something similar a few years back but got offered a good deal a while ago from a friend in the business.
Oil bath hardening if required. Take to near red heat and dump into a bucket of oil. An ordinary domestic fire can get to that point.

[Baza]

How are you controlling the float that exists in the drill head?

thanks
Barry

[sheeny]

I don't know if this will help or not, but have you seen this:
https://www.amazon.com/Metal-Lathe-Build-Working-Scrap/dp/1878087010/ref=pd_lpo_sbs_14_t_1/144-3963615-8242432?_encoding=UTF8&psc=1&refRID=WGWVP8SVBPFX8K4N3MB2

I thought there might be some tips in there you could use.

I got the books to build the foundry and I reckon they'll be a good project(s) for retirement.

Al.

[Boozlefoot]

Hi Bojan,
I once converted a massive old Russian bench drill into a mill by changing the bronze bushes in the head to needle roller bearings, and then using a compound vice for my moving table. I think it would be possible to use the compound vice and convert the head section of it to become your tool holder maybe? The size you're chasing is about $78-$80. This would give you two axis movement. (I'm currently converting my cheapie hydraulic log splitter to a workshop electric press..........necessity, the mother of invention!)

Cheers, Rod

[skysurfer]

Is it mild steel ?
In that case you can also carburize it, that means wrapping the workpiece in powdered charcoal in a crucible or a steel container and heating up the whole stuff to 900ºC for a few hours. After that, quench the cherry red hot object in oil or water as soon as possible.
Then reheat it to about 200ºC and allow it to cool.
There will be a hardened surface with a higher carbon content on the object.
I made a hammer that way.

[bojan]

First turned piece - pencil :P

[bojan]

Hi Rod,
Thank you for the tip - I know about this thing, actually I have one and using it with larger press drill.
I am still thinking out the details... how to and what exactly to do...

[bojan]

Not sure what kind of steel it is..
the rod came from China, I bought it for my 3D printer.
I have one more piece left, so I will experiment a bit before I go for the tool..

[Baza]

If that is the linear bearing material, the 10mm that I have appears to be case hardened which is softer steed in the centre.
There used to be a case hardening powder which was sprinkled on the red hot steel then quenched at an appropriate temp. the temp was established by colour. It requires polishing the red hot steel so the cooling colour change is apparent. It is no longer cheap https://www.lprtoolmakers.com.au/case-hardening-powder-200g/

This link may be useful http://www.model-engineer.co.uk/forums/postings.asp?th=64390

[bojan]

Hi Barry,
Thank you for the links.

The lead screw is one of (or similar to) those:
http://www.ebay.com.au/itm/TRPZ10x2-400mm-TRAPEZOIDAL-SCREW-ACME-LEAD-CNC-ROUTER-MOTION-3D-PRINTER-MACHINE-/112147375510?hash=item1a1c812596:g: f50AAMXQVgBRGidh

During my first year at uni, we were required to spend couple of days in workshop (I worked on lathe and was very good at that)... I remember they used the "bath" of molten (red hot) salts mix for hardening the lower carbon-content steel, not sure what was used...
I also made my own hammer then (500g), that I used later for many years.

[Baza]

My best guess is case hardened, if at all given the price point. Precision ground items may be of better parent material but more expensive.

[xelasnave]

I expect you already know this.

To harden heat to red hot and drop it in water. You then need to temper the item.

To do this.....

Polish the item back to metal.

Heat slowly and look for the colour appear. When you reach the appropriate colour drop the item in oil.

Straw to brown is probably where you need to be but remember as it goes blue you are making something as brittle as a file... I have a colour chart some place and if I can find it I will post it but there must be something on the net...

But you may only have case hardened steel in which case hardening and tempering is not going to work.

alex

[bojan]

Alex,
Thank you for the tips...
Yes, long time ago I was doing this, however this ACME tapping tool, made for me by a friend, is sort of unique - so I didn't want to ruin it as it is perfectly good for the purpose it was made (Delrin nuts for lead screws, for my 3D printer).
Funny thing, recently I bought on ebay another two lead screws (for lathe, from another supplier as I couldn't find the previous one any more) and those new screws are of slightly larger diameter (older ones were 9.8mm, those are exactly 10mm in diameter) so they do not fit into already made nuts... it seems I will have to bother my friend again...

[bojan]

Moving forward...
still missing split nut and gears for fine lead screw rotation.

This (https://www.thingiverse.com/thing:270309) or similar is what I have in mind.... of course, not made of plastic.

And dead/live centre on the other side.

As for parallelism between chuck and sliding rods, I will try to use laser collimator. I need to achieve better than 0.1mm, 30cm away from chuck

[bojan]

I was thinking...

In order to do threads on that machine, I would need to add encoder to the main shaft and stepper on the lead screw.
A processor would then move the tool in sync with chuck rotation along the axis of the thread.
The radial movement can be done manually....
No need for gears... :thumbsup:

[miker]

Have you read this thread on making leadscrew nuts from Acetel/Delrin using the leadscrew as the form?

http://www.denfordata.com/bb/viewtopic.php?f=50&t=3727

Apparently it has been quite successful.

Michael

[rally]

Bojan,

There are many different hardening processes, the process depends entirely on the type of hardening you need and the type of material that you are hardening.

Are you needing to through harden or case harden ?

If you have a high carbon steel or a tool steel or silver steel then heating it to even dull red and quenching it vertically may work and that will give you a through hardening
Most Tool steels are typically vaccum hardened and gas quenched.

If you have a low carbon steel then you need to get carbon into the steel and that takes some sort of active process. Or maybe you need to get Nitrogen into it eg Nitriding
These processes work with very low carbon steels and will case harden.

. . . This all depends entirely on the steel alloy you are trying to harden

If the shaft is 4140 then you can Nitride (specialist process - cant do it at home) and that will give you a thin hard outer case.
The same material will through harden if just heated and quenched

If its some of the carbon mild free machining steels such as 1020, 1025, 1045 then that process may through harden (if quenched properly)

If its an old axle shaft or similar - EN25, EN39, EN36, 8620 etc then you would need to carburise and case harden.

Carburising can be done in a carbon rich gas environment, or done in a salt pot using molten cyanide salts - the two most popular commercial processes.
But there are others.
Sealed quench carburising or the old fashioned way to seal the part in a small container wrapped in carbon/charcoal - both processes but can take many hours - 10 to 13 hours is typical as the carbon literally has to permeate into the steel down to a mm or two. (Big gears are even longer)

The common thing with almost all hardening is there is a quenching process.
But again there are many different methods - usually depending on the material and the process.
The rapid cooling literally locks the crystalline structure of the material as it was at high temperature, do it too slowly and you might just temper the material back to normal !
Do it wrong and you will distort your shaft.

Quench from too high a temperature or with Oil/Water too low in temperature and you can distort or crack the steel.

Some processes and materials should be tempered after hardening (too brittle, or core refining), others not.

Without knowing the type of steel alloy you have - its either a case of lucky dip or potentially damaging the steel you have by trial and error.
Although most will temper back to normal.

A heat treater will probably be able to identify the steel and use a process to give you a good result and that is my recommendation.

[bojan]

I haven't seen this article, but I have delrin nuts made for my 3D printer ans CNC mechanism (project in progress), and they are OK.
(see here (http://www.iceinspace.com.au/forum/attachment_browse.php?a=217626).. the black thing on the table)

Recently Chinese suppliers are selling lead screws + brass nuts on ebay (http://www.ebay.com.au/itm/10mm-T10x2-Lead-Screw-Rod-Trapezoidal-ACME-Brass-Nut-Kits-200-400mm-3D-Printer/272769445457?ssPageName=STRK%3AMEBI DX%3AIT&var=571906150430&_trksid=p2060353.m1438.l2649) for ~$13... I am using them for this project, they seem to be quite good.

[bojan]

Hi Rally,
I have no idea what kind of steel the rod is made of.
I guess I will have to take a risk and try.

[bojan]

Someone "stole" my idea (https://www.youtube.com/watch?v=BaBK9teKUaA)... :D:thumbsup:
Doesn't' matter, It will be easier now for me to actually finish this project

[dimithri86]

Looks good. I was thinking of doing similar, but I found a metal lathe in the council pickup (can you believe that?)

[bojan]

Yes, I can... but it never happened to me :sadeyes:

[bojan]

Firmware can be downloaded from here (http://www.chipmaker.ru/files/go/3382360807aefbbffbdc8bed6f902582/)..
(looks like registration is required for downloading more that 3 files... so I registered)
It seems there are "lite" and commercial versions (? my Russian is a bit rusty..)..

[dimithri86]

Have you read the books by David J. Gingery?

In particular this book
https://www.amazon.com/Metal-Lathe-Build-Working-Scrap/dp/1878087010

[bojan]

Too late for that now, the concept of my machine is already determined by what I have done so far...

[bojan]

Axial motor mounted.
Unfortunately, the lead screw is double start (ebay seller's mistake)... so I am not sure if there will be enough torque, but is should be OK while I am waiting for the single start one.
I will probably finish design of radial slide compound by then as well.

Now it is time to start thinking about electronics....
I am playing with idea to use encoder instead of knobs....

[MattT]

:eyepop: the Cryobi drill press is only just recognisable....

A question....what did you do with the red light centreing thingy and switch?

I have an idea to use them as illuminators for setting circles on my old mount.

I have one from my drill press....care to donate the other one, for a small consideration of course :question:

Matt

[bojan]

Matt,
the "thingy" is actually ø6mm laser with diffraction "cross generator", focused at ~20cm from lens.
Unfortunately, I already used it for my 3D printer alignment.

However, those things are easy to get on ebay ... This one (https://www.ebay.com.au/itm/650nm-5MW-Red-Laser-Line-Module-Focus-Adjustable-Laser-Head-new-Laser-Module/142169224600?hash=item2119f23598:g: 4-QAAOSwGIRXa6vw) is line only (possibly better for your application), but if you want cross, look here (https://www.ebay.com.au/itm/Red-Cross-Line-Laser-Focusable-5mW-650nm-Module-Focus-Adjustable-laser-Head/322720338712?hash=item4b23a19318:g: v1IAAOSwrblZplS0).
They are both powered by 2xAAA, 3V

[bojan]

Today I tried to compile the arduino sketch (By Oleg) (http://www.chipmaker.ru/files/file/9787/)
(the link may not work... so I translated some comments and attached the archive).

After some initial troubles with errors ("not defined in current scope" etc.) fixed (probably) by adding couple of libraries, the code compiles OK.

This firmware version (7a_lite) supports handle encoder (one knob for both axis, switchable by selector switch), limit switches, adjustable feed rate (pot), thread cutting (this works with chuck encoder)... almost everything I may ever needed (or so I think at the moment).

[bojan]

While waiting for keypad/LCD board to arrive, I was thinking to temporary use Digispark board as quadrature decoder (conversion of encoder output to STEP/DIR).
While I know that this is not really needed (because encoder could be connected directly to STEP/DIR inputs of the Pololu driver (of course, the back/forward transitions of STEP could be interpreted as continuous movement in one direction, but it is not the real issue here), my reason for using this $2 board is the 5V regulator, needed for optical encoder power supply (Oak-Grisby (https://www.ebay.com/itm/2-X-Oak-Grigsby-Rotary-Encoder-128-Pulses-Rev-TTL-Compatible-Gray-Code-/281491428561)).

I found an interesting code (https://thewanderingengineer.com/2014/08/11/rotary-encoder-on-the-attiny85-part-2/) for controlling the intensity of LED (tried it, works on Digispark after changing the port for LED). The idea is to replace the "analogWrite" function with STEP/DIR pulse outputs (1us duration)...

If I try hard enough, I will probably do it myself but... it may be easier if someone could point me to suitable piece of code? Thanks in advance!

EDIT:

This code is tested and works OK, output is DIR and PULSE at each interrupt (5 usec duration), depending on encoder movement direction.


******************************
#include "avr/interrupt.h";

volatile int lastEncoded = 0;

void setup()
{
pinMode(0, OUTPUT); // DIR
pinMode(1, OUTPUT); // STEP, on-board LED

// set pins 3 and 4 to input
pinMode(3, INPUT);
pinMode(4, INPUT);
digitalWrite(3, HIGH); // and enable pullup resisters
digitalWrite(4, HIGH);

digitalWrite(0, LOW);
digitalWrite(1, LOW);


GIMSK = 0b00100000; // Enable pin change interrupts
PCMSK = 0b00011000; // Enable pin change interrupt for PB3 and PB4
sei(); // Turn on interrupts
}

void loop()
{
}


// This is the ISR that is called on each interrupt
// Taken from http://bildr.org/2012/08/rotary-encoder-arduino/


ISR(PCINT0_vect)
{
int MSB = digitalRead(3); //MSB = most significant bit
int LSB = digitalRead(4); //LSB = least significant bit

int encoded = (MSB << 1) |LSB; //converting the 2 pin value to single number
int sum = (lastEncoded << 2) | encoded; //adding it to the previous encoded value

if(sum == 0b1101 || sum == 0b0100 || sum == 0b0010 || sum == 0b1011)

digitalWrite(0, HIGH); // Direction forward

if(sum == 0b1110 || sum == 0b0111 || sum == 0b0001 || sum == 0b1000)

digitalWrite(0, LOW); // Direction backwards

// Pulse after delay
delayMicroseconds(5);
digitalWrite(1, HIGH);
delayMicroseconds(5);
digitalWrite(1, LOW);

lastEncoded = encoded; //store this value for next time
}

EDIT 2:

I stumbled on un-expected hardware problem with Digispark Attiny board - because of protection zener diodes on USB connector (marked as 'W4', supposedly 3.6V), input voltage levels from encoder were around 2.5V only - so the whole thing worked very erratically at 5V due to the noise from Pololu drivers. After I removed those diodes ( I think they were not really needed, they were placed "just in case" and they were obviously not 3.6V as marked in some schematics found on web), the circuit worked as expected.

[bojan]

Work on my lathe is progressing, now I have both axis properly assempled and equipped with stepper motors.


The web search for firmware and PC application resulted in this project (https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&uact=8&ved=0ahUKEwj-hLOV-q7bAhVCFZQKHdxtCBcQFghRMAY&url=http%3A%2F%2Ftlc.iiitdm.ac.in%2 Fux2%2Fwp-content%2Fuploads%2F2017%2F08%2FCNC-Lathe-Manual.pdf&usg=AOvVaw3ioLtrcz8Sw8OKIpmLrUXr), lathe is controlled by GrblGru.


Is there anything else/better available? GbrlGru simulation is behaving strangely when I imported very simple LibreCAD drawing (dxf), the 3D model looked OK on the screen but the tool movement would damage my machine in reality.

[bojan]

Tried today om my workbench..


All works OK.
All I have to do now is to put back motors on lathe...


And perhaps to build joystick & controller, to have a lathe independent of PC or laptop.

[JA]

I love the industrial chunk and heft of your modification - Well done Bojan :thumbsup:

Best
JA

[bojan]

Thank you :-)
I tried to make the design no-nonsense & simple, using available standard materials (aluminium extrusions, standard bolts), parts from my drawers and minimal modifications to achieve required functionality. The whole thing works, that's important.

[bojan]

Re joystick...

I found this webpage (https://www.brainy-bits.com/arduino-joystick-tutorial/)...
The code works as it should, now I have to modify it according to this page (https://github.com/gnea/grbl/wiki/Grbl-v1.1-Jogging) and/or to this page (http://cncsimulator.info/blog/?p=393)to generate jog commands for grbl controller, for lathe movement to be independent of laptop.
Also need to mount verniers.

[bojan]

It's working....


The calipers are $10 each from ebay... nor stainless neither hardened (well, it is some hard aluminium alloy or "white brass" (it behaves like brass when filing, and it was cast).. so the printing on them is not totally false).


Cross slide one is positioned a bit too high (so it may be in a way of tool in some cases), and of course it indicates radius, not diameter.

[bojan]

Finally, it is complete (almost).
Controlled with GRBL (Arduino-Uno) + another Arduino Nano which reads hand encoder and switches on front panel and sedns commands to GRBL via serial interface.
This way, I have full manual control of the machine, but if I need to execute program, this is possible by connecting computer/laptop via USB.