My new lathe

[bojan]

Because my home-made lathe (http://www.iceinspace.com.au/forum/showpost.php?p=1406819&postcount=40) was not good enough for larger pieces and harder materials (plastic was OK, but that was about all), last Friday I decided to invest in "real" machine (https://www.ebay.com.au/itm/600W-Mini-Metal-Turning-Lathe-Woodworking-Tool-Bench-Top-DIY-Processing-Drilling/322701159722?ssPageName=STRK%3AMEBI DX%3AIT&_trksid=p2060353.m2749.l2649)...

It arrived today, and it is quite a piece of iron.. and actually bigger than I anticipated or hoped for.

Unfortunately, it worked (erratically) only couple of seconds.. and then the motor stopped.

I didn't smell any smoke, but it is likely the faulty control board.

Now I have to through the hassle of servicing or returning the whole thing... which is not easy considering 65kg of weight.
Anyway..

[miker]

Have you checked the safety cut outs on the chuck guard, gear end cover etc? Maybe a sticky or incorrectly adjusted one.

Michael.

[bojan]

Everything checked.. Head rotates freely by hand.

Motor was turning quite erratically for couple of seconds and that was it.
Pity, the machine seems to be mechanically quite good.

[Symon]

Nice little lathe, I have an old Hercus they were made in South Australia a lot of them are the lathes that were in high schools, my one was out of Blackwood high my 9 year old told me that's where the hip hop band the hill top hoods went to school lol, anyway my wife says the only thing I seem to make with my lathe is more things for my lathe:lol: hope you enjoy
Simon
Actually was just looking at the lathe you made its pretty neat just wondering how you managed to drill through what looks like vernier gauge with the heads cut off? that must have been tougher than a cat's head

[peter_4059]

At least you don't have the same model as me - mine's 280kg!

[JA]

Sorry to hear of your problem Bojan. Capacitors?

Best
JA

[bojan]

I had a look at the control board.. nothing obvious.

It seems the output driver of motor controller is single FET (or SCR), on heatsinks (next to coil must be Graetz rectifier). On the attached image, red and black are mains (left of the PCB), red, green, black bundle (next to the adjusting pots) is speed control pot on the front plate, DC motor wires are blue and white (in the middle).
The supplier wants me to provide demonstrating video of the problem, to come up with solution (which will be replacement of the controller board (my preference) or the replacement (more likely) of the whole machine)
I think I made mistake with supplier.. but no one else had that particular model, and form mechanical point of view this machine is exactly what I wanted.
If supplier becomes difficult or the whole situation gets too complicated, I may attempt to repair controller board myself, it seems it is very simple circuit, probably just PWM driver and SCR or FET output stage (motor is DC, no visible marks on it.. I will try to connect it to 12V just to see if is operational).

[bojan]

I connected 12V to motor - and it works (couple of rev/sec).
So- no doubts, it is faulty controller board.

[bojan]

One pair of calipers (now on standing drill cross vice) was really hardened and not easy to drill.
Another one (currently on that home-made lathe) was made of "white brass", so easy to work with. I would like to get some more, but when I ask vendors about material they swear their products are made of "best quality stainless steel"...

[Symon]

Thanks for the reply it actually made me smile hope you get sorted soon
Simon

[bojan]

I just couldn't resist...
Some pretty nasty hand-soldering here.. machine soldering is OK.
I found cold joint on "A-" motor faston tab, I re-soldered it but still no joy. It must be FET or it's associated driver or both.

[The_bluester]

Are they using solder to increase the current capacity of the tracks?

[bojan]

Yes... not very good idea, but it works.
Also, there is no proper ground for the rest of electronics, because they used Graetz rectifier directly on mains.. so the "GND" is actually live every 10ms.
Very unsafe practice, actually it is illegal (in Europe definitely, here in Oz most likely).
That also means it is very tricky to debug this circuit as is, to do that isolation transformer is needed (or test gear should float... ).
Or, separate external supply voltages for motor and for driver electronics must be applied.

[The_bluester]

I have been out of electronics work so long my brain is bubbling trying to work it out. Ignoring the volts drop of the diodes (Insignificant) is that essentially about a 240VDC circuit but essentially a balanced circuit with no ground reference as such, the neg rail around -120V with reference to "ground" (The MEN grounded neutral in this instance) and VCC of +120VDC?

[bojan]

Not quite..

DC voltage between "+" and "-" lines is 340V (240*SQRT(2)).
If neutral is at GND potential (it never is), then "+ " oscillates between 0V and +340V, and "-" oscillates between "0" and -340V with frequency of 50Hz.

This could be deadly.

[miki63au]

Hello Bojan,


Test that kit with RCD!! I bought (for work) a SMD de-soldering kit. The heat-gun keep tripping the RCD till I reduced the heat on the 3rd unit below spec. The mica insulator with fiberglass reinforcement burned thru and keep tripping the RCD. These items from China should not pass testing, and sold without modifications for safety! Check yours before using it!





Mick

[bojan]

Well, this is end of my road. I applied the external power (12V) to controller board, in parallel to existing 240V->12V supply regulator.
LED marked with black arrow is supposed to be ON.

HT66F018 (A/D Flash MCU with EEPROM) is PWM generator, and it is dead, there is no output activity from it.
I need another controller board....

The good thing about all this is, there are DC motor controllers on the market (I can even make my own with Arduino), for couple of $$ and better/safer than original one.

[bojan]

I think this circuit (https://www.tehnomagazin.com/Motor/DC-permanent-Magnet-Motor-speed-Control-schematic.htm) may be good enough for the purpose..
It doesn't have overload protection (like chinese deign) but it is simple and low cost.

[Symon]

Great news :thumbsup:

[JohnF]

Not counting you repairing that Lathe, I used the Metal working Lathe at the Tweed Heads Men's shed -- am learning to use a lathe -- to make a couple of adaptors to use Weightlifting weights as counterweights on my Skywatcher EQ6 -- it came without Counterweights or a tripod, but cost only $200 with an Argo Navis, etc. plus I gave him a Marklin train set that I would have been willing to sell for just $200.

So what I am saying, is once the Corona Virus is over, your local men's shed may also have a Lathe. Only $50 a year and $5 a week for coffee and a raffle ticket at our one.

Bojan already has his Lathe, but others may be interested in this. Good luck in getting yours going Bojan. Sn some small ones that I would love, but no money.

[bojan]

Vendor promised the new controller in couple of days..
In the meantime, I tried the circuit discussed earlier (it is basically dimmer, 20Acapacity).. and it works OK, as expected. Probably not any worse than original controller (apart from overload protection and whatever else non-obvious features are implemented).
So - lathe is operational as of now :)

[Symon]

That's great news:D Seeing your lathe, has renewed my enthusiasm and I have been tinkering with mine making a hand mandrel for it, at the moment to help with threading.
Simon

[bojan]

:thumbsup:
My next project with lathe will be adding steppers... turning manual into CNC.. but not now :)

[multiweb]

Good to hear you're getting a replacement board. :thumbsup:

[bojan]

Now, this situation is becoming ridiculous...
The new controller arrived today, I installed it in the machine... and nothing, no "m" of motor moving... but this time indicator LED (at one of processor's output) was ON..
All connections seems to be OK....
I applied external 12V to board and there is PWM on FET's gate.. so this time I have board with faulty FET.
Plan now is to transfer (probably) good FET from old board to new one and see what happens...

[Symon]

Sorry to hear about your setback . Yes cnc would be nice
Simon

[Pepper]

Did you end up fixing it?
I’ve been looking at something similar but reviews are very mixed.

[bojan]

Yes..
The original problem was in faulty wire-wound potentiometer, the wire was broken on one place and when I was turning the knob the wire touched the metal enclosure of the potentiometer, so the processor input was put directly to 340V and consequently destroyed..

After replacing the pot with new one, the second board works as intended.

[Symon]

Great news look forward to seeing some of your projects
Simon

[bojan]

So am I :)

[bojan]

Saga continues...

Electronics is still not sorted out 100%, I discovered the minimum head speed is ~ 180rpm, too fast for threads..

I am sure they send me the wrong controller, it is probably suitable for US (110V AC).

I tried with variac, and yes, min speed is correct at 110VAC, but then the safety switch doesn't work.

So I asked for another one, it is on its way now...
In the meantime I had a look at mechanics.. and found one thrust bearing broken.. already replaced.
The cross slide lead screw doesn't have any bearings, 10mm shaft sits on edge of Ø9mm bore in aluminium bracket !!!!.
So I widened the bore (using step drill on the same lathe, mounted in tailstock) so I can insert the ball bearing (8x16x5mm, the only one that fits into available space). It is radial type, not ideal for the purpose but it should be way better than original "design". I will also insert the front bearing, but not right now, for this I need access to another lathe.
I can't say I am bored as SIS :lol:.

[Merlin66]

Bojan,
This was a brand new machine, yes???????

[multiweb]

Wow! a whole ball is missing?

[bojan]

Yes, brand new...

[bojan]

Well, not missing really... only crashed, I found half of it.

[bojan]

This is original situation with compound lead screw.. no bearings, by design.

[multiweb]

:lol: what was it made out of? Takes serious pressure to crash a bearing ball. Usually the track or cage are the first to go. You might want to go around the whole lathe and replace them with equivalent timken or other.

[bojan]

That's what I am doing...


I think this bearing was installed like this in first place... because the feed spindle was turning OK when I first inspect the machine. I realized it was faulty only when I went to investigate the reason why the nut self-unscrewed and fell on the floor.

[bojan]

New compound bearing plate (aluminium flat bar, 60x12mm)...
I will cut off the excess later (with circular saw).. the current size is larger to fit the part into 3-jaw head, properly centered for bearing seat.

The seats will wait for thrust bearings to arrive (in couple of days).

[alval]

[QUOTE=bojan;1474873]Saga continues...

Electronics is still not sorted out 100%, I discovered the minimum head speed is ~ 180rpm, too fast for threads..


It sounds like youre getting there, and by what you are doing will be way better than stock lathe. Imagine the person that cant problem solve and fix things would be getting sick of sending things back

My lathe has geared speeds and 160 is the slowest, no brake, so it just winds down when turned off. I thread away from the chuck. Same set up except tool is upside down and you run the lathe in reverse. Start from the run out for the thread. Make a carriage stop so you start in the same place each time. This way you can thread right up to big shoulders, or in bores you you can thread a blind hole up to the end. Much simpler and you dont have to worry about quick winding out or crashing the tool. (On bores thread off back face with an opposite hand bar and thread is always visable). Speed is then not an issue and most things cut better with a bit of speed.
Al

[bojan]

Yep, I was thinking about this method if they send the wrong controller again.
But, since the machine is new and sort of under warranty, I will insist on full functionality as advertised.. Anything else may be considered as improvement (the seller actually offered to pay for cost of repairs),

I sort of enjoy solving problems like those and at the end of the day I will know what I have.

[bojan]

Daniel Reardon's post (Astrophysicist with magnets in nose) gave me an idea - how to firmly attach dial bracket to lathe bed, for precision readout of tool position... and to be easily removable at the same time. I used Ø8mm Neodymium magnets (I found them in some old faulty toys, and until now I kept them on kitchen fridge door... I think this application is much more useful.
Picture is self-explanatory...

[alval]

Yeah neodymium magnets are great for quick attachment of all sorts. Might I suggest a bit of tape over the magnets as they attract lots of swarf and chips, when you need to pull off the tape and the magnets are clean, put on some fresh one for next time. They are a pain to pull chips off of but really handy.
Al

[bojan]

:thumbsup:

[bojan]

Today, after 1 w waiting, I received the parcel (from China via DHL).
But.. instead of new, correct motor control module or new processor (as I hoped we agreed after many messages), I found the potentiometer in the box :shrug: (not that I wasn't expected something un-expected after what was going on previously).

I will have to give negative feedback to this seller..

They really, really don't know what they are selling (and neither manufacturer know what they are doing.. or they think I am stupid. Which is probably true because I was too patient with them).


Anyway.. not a big deal really (machine works), but this is not the way how to treat customers.

[bojan]

Since the issue with seller isn't going anywhere, I decided to modify existing controller by replacing original processor with AtTiny85.
The following "sketch" (arduinian for "source code") does the trick:

****
// HT66F016 application for lathe DC continuous variable speed motor controller
// to run on attiny85 as replacement

const byte pwmPin = 0; // PWM output pin defined
const byte LED = 1; // It is also used for onboard LED
const byte PWM_EN = 3; // PWN enable LOW, no activity from original processor
const byte analogInPin = A2; // input pin defined

// Original HT66F016 PWM frequency is 15kHz. Default ATtiny85 PWM frequency on P0 is 500Hz
// To change that:
// https://digistump.com/wiki/digispark/tricks
// MS_TIMER_TICK_EVERY_X_CYCLES in arduino-1.0x/hardware/digispark/cores/tiny/wiring.c is set to 8 (default is 64)
// resulting in PWM frequency on P0 =4kHz (If set to 1, freq would be 32kHz, possibly too high)
// FAVOR_PHASE_CORRECT_PWM to 0 in arduino-1.0x/hardware/digispark/cores/tiny/core_build_options.h file, it's possible to double the frequency on Pin1
// How to increase this frequency to ~15kHz?

void setup() {
//TCCR2B = TCCR2B & 0b11111000 | 0x01; // This doesn't work

pinMode(pwmPin, OUTPUT);
pinMode(PWM_EN, OUTPUT);
pinMode(LED, OUTPUT);
digitalWrite(LED, HIGH); // LED ON
digitalWrite(PWM_EN, LOW); // another wired-NAND for PWM, no activity observed form original /u processor, so here just in case..
}

void loop() {
int analogIn = analogRead(analogInPin);
analogIn >>= 2;
analogIn=analogIn/1.5;
// Alternative statement is: analogIn = map(analogIn, 0, 1023, 0, 255);
analogWrite(pwmPin, analogIn);
}

***


I will couple the PWM output to the rest of controller electronics with opto-isolator, which will give me the option of using GRBL for spindle speed control in the future.

EDIT: I added enable output (LOW), the same functionality was on Holtek HT66F018 processor (but no activity observed), probably to prevent activation of high voltage output during processor startup.

[bojan]

This is how $2 Digispark ATTINY85 board fits onto existing controller PCB without any modification. I may solder the chip directly on board later, but it's mounted perfectly OK as it is now.
I also reverse-engineered the controller circuit to some extent..
HT66F018 (faulty now) is full-blown 8-bit processor, but it's only function here is generation of PWM from pot position (exactly what Attiny85 with my program from previous post does).. and nothing else.

Overload protection is facilitated by separate Opamp and comparator IC's, signal from this circuit disables PWM externally in separate wired-OR circuit.. Maybe they used here processor from, say, vacuum cleaner power regulator to save the effort of designing more sophisticated firmware? This would explain my troubles with minimum head speed....

Anyway, it seems I will have exactly the same functionality after this modification is finished, except I will have the desired lower speed.
And another controller board as spare :-)

[bojan]

The motor controller works on bench, as expected :D
It is ready now for test in the machine.
If all goes well after tweaking speed, I will need to figure out how to get rid of 5 sec delay (used by bootloader) so the main code starts immediately after power-up.

[multiweb]

:eyepop: This is turning into an expensive DYI. For the money you've paid the seller should ship you a new one. I wouldn't modify it.

[bojan]

Well.. no....
Seller send me already the wrong items anyway, I think they are helpless here as well. They are just sellers..
The new machine may have numerous other problems, I found heaps of videos (https://www.youtube.com/watch?v=gzJnjm4Va9E&list=PLooB2EjBotjLpN5sL1KflRLwNUS7t SpBH) about them.. and I didn't want to pay 3x or 4x more for professional machine. If I found something second hand I would have gone for it, but, I was not that lucky.

Then...
First, what else to do in isolation?

Second, it is only my time that is really spent here, plus couple of bucks for additional electronics (which I already had in my drawers doing nothing), and a bit more for bearings ($5 each, some of them I also had in my stash).
And, this story may be useful for someone. Like, what to do in similar situation, or what not to do to get into situation :P.

[bojan]

Last update:
Lathe works now with repaired control module (original faulty processor replaced with Attiny85).
I did some more measurements on both modules...
Minimal pulse duration (determined by external buffer circuitry) is ~8us, and combined with ~15kHz (generated by Holtek processor) results in quite high duty cycle (~12%), and that means high minimum speed.
My program for Attiny85 generates ~4kHz for PWM and with ~8us PW duty cycle is ~4%, so the minimum speed is considerably lower, ~70rpm, acceptable for threading.

This is one very primitive controller.. obviously made with very low level of expertise in analog design.

For example, they are using diode in series with transistor base in buffer circuit (no idea why.. maybe to have better noise immunity.. but then they have too slow transistor switching speed? And they have no control over duty cycle at low speeds.. except via PWM frequency).
Bypass cap (1nF) across this diode reduces duty cycle down to 1.5%... Or, using series resistor instead of diode) could also have the same effect.

They should have used dedicated Gate Driver IC for FET imo.

No wonder people are buying better electronics (which is not cheap, ~US200-300).
Anyway.. Another project is finished.

[bojan]

One more update on electronics and no more...
After simulating circuit with LTSpiceXVII, conclusion is they screw up the design.
Diodes are not supposed to be in series with bases of the transistors, they should be connected between base and collector, similar to Schottky-TTL logic family output circuit.
This way transistors do not go into saturation, so their switching-OFF speed increases up to ten-fold.
For actual mod I simply replaced diodes with short jumper wire, and soldered BAT54 diodes on top of existing SS8050 transistors.
The actual minimal pulse width is now ~1/us, 1/8 compared to original.

[iborg]

Hi Bojun


Sounds like you know a lot about electronics. Damn good job on improving the lathe circuitry.


Philip

[bojan]

Thanks :-)
Analogue electronics design is my profession...

[iborg]

I suspect just a tad more advanced than my TAFE "Advanced" Diploma in Electrics Engineering.

[bojan]

Well.. that plus additional 50 years of working experience in the field...
I am much slower problem-solver these days compared with what was 10 years ago...

[bojan]

How to modify cross slide parts of the lathe on the same lathe ;)

Well, it is possible... with help of compound slide rotated by 90°.
The backlash is almost non-existent at cross slide now after insertion of small ball bearings, and, more importantly, there is no friction of aluminium to aluminium between dial and rod barrel (machined as shown on pictures).

[bojan]

Finished....

[multiweb]

Bravo! :clap:

[bojan]

Thanks :-)
Now I can start business ;)

[DavidU]

Great project Bojan.

[JohnF]

My son was a Welder/Boilermaker until some one talking on there phone rammed his bosses work van and broke his back -- at last he is still on his feet. They got in some best quality steel girders from China. At one paint you could see the engine number of a Car engine block that was made into that girder. Parts of the steel were very hard, other parts were soft.

So wondering if your "Bst quality Stainless steel" has the same problems.

[bojan]

My calipers were made of "white brass" (copper alloy with zinc added).
When I was cutting them, the material behaved exactly like brass (it felt like slippery) - only it was silver-white colour.
Casting signatures are easily visible on attached images...
Those calipers were cheap ($10) but very easy to modify...

[bojan]

Yesterday I discovered a potentially dangerous bug in original motor controller firmware..
Something went wrong while I was cutting a metal rod and my instinctive reaction was to quickly turn the potentiometer knob to the left to stop the spindle rotating (instead of hitting the emergency button..)... but there was no expected effect - the spindle continued rotating.

Later I experimented a bit and yes - only relative slow(er) movement of pot controls the spindle, but quick change doesn't.

This is another reason to replace original processor with AtTiny..

[bojan]

While searching for ideas for DRO (Digital Read Out) for lathe (because calipers are not long enough), I remembered I saw something about PCB strips for them...

Some more search and I found this:
https://www.ebay.com.au/itm/Replacement-Sensor-Strip-For-Wixey-WR-700-Saw-Fence-Readout/254616464933?ssPageName=STRK%3AMEBI DX%3AIT&_trksid=p2060353.m2749.l2649
The pattern pitch is the same as used with cheap Chinese calipers... Perfect solution for my case.
I plan to stick this strip to 16x3mm aluminium flat bar (https://www.aluminiumtc.com.au/store/product/view/16-x-3-303/), available at Aluminium Trading Centre, and use it with one of standard caliper reading heads.
Perhaps connect it to Arduino LCD display which could then be conveniently placed somewhere on the machine...


Something like this:
https://www.thingiverse.com/thing:1731982

[bojan]

DRO prototype works...
For now only one sensor, and only one readout (displayed as x and y).


I noticed when caliper is in imperial mode, DRO displayed value doesn't match reading on caliper LCD.. so, this f/w version works only in mm mode.

[AndyG]

This is absolute artwork that you have done on this machine Bojan.

In a better world, you would be able to sell your design back to the manufacturer. Bugs like described in post #64 (http://www.iceinspace.com.au/forum/showpost.php?p=1477767&postcount=64) are unforgivable on their account.

In practise, they'll keep selling these units as is, to everyone ongoing. Only you have a good one now :)

[bojan]

Yes, unfortunately this is true..
However, there is a lot more to do, I've only started..

For example, I am still not sure about both slides precision.. most likely I will have to ground them.. There are lots of videos on youtube discussing those issues.
I did offer the processor solution to them, but they ignored it ;)


Attached below is where to mount the scales....

[dikman]

I just found this thread, very interesting bojan. There are two versions of what is commonly referred to as the Chinese mini-lathe. Yours is the slightly bigger, and supposedly better, version. A chap on another forum I'm on bought the same one you did and had no problems with it. I bought the smaller, more common version. My controller lasted two weeks! The seller couldn't supply a replacement board, which I asked for.:rolleyes: I won't go into the details but I fought long and hard with them (and ebay and Paypal) and got nowhere! In the end I bought a stand-alone controller off ebay ($30) and fitted it to the lathe. It's been fine since then.



I spent time on mine taking out high spots on the ways, smoothing and adjusting the slides, modified the compound slide so I don't have to remove it to change the angle and fitted a tacho. There's lots of info on the 'net about these lathes and the controllers. The best option for these lathes is to replace the motor and controller with a 3-phase motor and VFD as this is their weak point.


I was extremely annoyed with the dealer I bought this off so bought another one!:lol: My intention was to replace the controller and then send the faulty lathe back for a refund. Long story short after three months fighting with ebay and paypal and after involving the Australian finance authorities Paypal grudgingly gave me a refund (and I kept the lathe).


A bit of advice to anyone buying off ebay, if you have problems only use ebay to fight any claims, paypal policies are NOT consumer friendly and once you seek help from paypal ebay will no longer help you.


My little lathe is very useful for working with small items, much better than trying to work with small stuff on the bigger lathe.

[bojan]

Yes, it is very useful.
The last hurdle will be obtaining the new cross lead screw, as the original is eccentric...

[dikman]

It's a real lottery when you buy one of these as to how it's been built. There are several factories that build these things, supposedly to the same specs. The second one I bought, from the same seller, looked the same physically but was noticeably different in quality. It came from a different manufacturer, the ways were smoother, with no binding, the controller board was built differently but was a straight swap and the motor was only 400w. This was the point of contention as it was advertised as 550w (the seller told me it didn't matter what was marked on the motor - 400w - it was really 550w!!!!!). I have realised that most of these Chinese sellers on ebay are, in general, crooks.


I was going to sell the second one but instead gave it to my son-in-law.

[bojan]

Well, that depends on definition of "crook".
I think they are just sellers who want to make a buck, without knowing what they are selling and doing really.. at least in my case.
The outcome is the same...

But.. at the end, the machine will be worth $900 I paid for (I think..)

[dikman]

In some cases you are right, but it's when you have problems and contact the seller that you discover what they're really like. You haven't done too badly because at least they tried to help.
In my case the lathe was made in China but stored and shipped to me from a warehouse in Sydney. One would assume that a return would be to the shipping address it came from (which, I believe, is what the ACCC says). In order for Paypal to give me a refund they require it to be returned first and the return address I was given was in England??!! Shonky? Just a little bit.:lol:


The other chap I mentioned that bought one like yours is very happy with it and found it to be quite accurate, given that it's not a high precision build. It's certainly a better design than mine, I looked into them but it was going to cost double what my little one did and seeing that I already have a larger lathe I didn't want to spend that much. As you've no doubt found there is a lot of info on youtube and the 'net about using these little lathes, plus various mods to do to them. I bought a quick release toolpost, which I see yours has, and a set of replaceable tip carbide tools, both of which make life much easier.


Once you get it sorted I'm sure you'll feel it was worth it. If you like making things then a lathe is almost a necessity, in my opinion.:thumbsup:

[bojan]

Quite so.

The quick release toolpost I had from my DIY lathe (http://www.iceinspace.com.au/forum/showthread.php?t=161739&highlight=lathe). As well as some tools I accumulated when working with it...
Well, this one is certainly better than DIY :P

[dikman]

I just read your thread, that is an impressive bit of fettling you did!:thumbsup:
Now you start all over again modding this one.:D

[bojan]

So it seems.. :)

[bojan]

While waiting for new cross lead screw (original one is eccentric.. obviously done on couple of different not so good machines, seller finally promised a new one), I am working on steady rest.... at the same time trying to improve my currently not-so-good welding skills..

[dikman]

Welding tube like that with an arc welder requires lots of practice and a delicate touch - I am also very good at blowing holes in metal!!!! I eventually bought a MIG (much easier to weld thin stuff) and then went to shielding gas. Not a cheap option but worth it, particularly for doing small jobs.

[bojan]

The lead screw arrived today, I installed it and since it is concentric, I was able to reduce the backlash down to 0.01 mm, with knob still moving smoothly.
Now the machine is ready for hyper-tuning :-)


EDIT: backlash is not 0.01, it is ~0.04 mm, this is probably the limit (screw/nut play)

[dikman]

0.01 mm! You've got to be happy with that.:)

[bojan]

Oops, typo.. 0.04mm, that's what it seems... when turning the knob I can feel when lead screw engages inside the nut.

[dikman]

That's still pretty good, given that wherever you've got mechanical couplings it will be almost impossible to completely eliminate backlash/play.

[bojan]

Yes, I am happy with this, especially when comparing with original situation, when there was a hole in aluminium and steel shaft turning inside..

Actually, cross slide is sort of better (almost no backlash but tighter), this is due to the way nut is attached to the slide (with screw), so when tightened, nut is also rotated a bit (as allowed by screw.. which is then hold tight),

[bojan]

Caliper ready for installation...
Now I need to drill/tap couple of M3 threads on the cross slide to fix it in place.

[bojan]

Done...
Also, yesterday I received measuring strip (https://www.ebay.com.au/itm/Replacement-Sensor-Strip-For-Wixey-WR-700-Saw-Fence-Readout/254616464933?ssPageName=STRK%3AMEBI DX%3AIT&_trksid=p2060353.m2749.l2649) (it is 600mm long PCB with capacitive encoder pattern), it works with Chinese caliper reading head.
I plan to mount it under the apron for X-axis.

[JA]

Is there no end to you creative drive ? Rhetorical

Best
JA

[bojan]

I am just bored :-)

[bojan]

After having both calipers installed some time ago (pic 1), I decided to go one step further: more sophisticated DRO with some math (radius-diameter conversion + some memory functions and calibration of tool position).
The idea was to use Arduino UNO for reading both calipers and display on LCD with buttons for commands.

For sketch (Arduinian for "f/w") I found couple of examples on web, but I had some initial problems with pin terminology (Arduinians are strange folks with some strange ideas of how to name things). After some fiddling with the code I have basic functionality to use with my lathe.
For level converter between caliper 1.5V logic and Arduino 5V logic I used cheap converters from ebay (https://www.ebay.com.au/itm/1-10X-2-channel-Bi-Directional-Logic-Level-Shifter-Converter-3-3V-5V-For-Arduino/274093085149?hash=item3fd13891dd:g: POEAAOSw~bldylaK) (LV needed to be dropped from 3.3V as suppied from UNO by means of series diode to achieve better output voltage span).

Pics 2, 3 are showing working circuit (here one caliper is used for both inputs so the displayed values are the same).

Next step is to mount the processor and LCD with buttons in the metal box and wire the whole thing to the lathe.

[bojan]

Caliper mounting details.
Both of them are as far away from tool as possible, but still easily visible by lathe operator.

[bojan]

Command box with buttons..

[bojan]

Installed...

[JA]

Very neatly done :thumbsup:

Best
JA

[bojan]

Thank you, yes it turned out nice...
For caliper connectors I decided to go with this design (https://www.thingiverse.com/thing:3141366) (from Thingiverse).

With first prototype I just soldered wires directly on caliper PCB, but this could damage the board... so perhaps connector is better idea (and hopefully reliable enough).
And, if anyone is interested, code (still under development) is attached as well.

[bojan]

Finished. Tray for swarf and 300mm DRO strip in place, all functional.

Now, I have to start using it :-)

[alval]

It looks good and functional but too clean 😊, unfortunately once you start using it you’ll find other things to modify 🙄. Things will annoy you, like swarf over the back so you make a larger guard, having to move to get to tools so make an accessible holder for quick change tool holders, another for centres. The list goes on but its all good. At least now you can make things, in between you can tweek things. Mines been going for twenty years, thinking of trying to make it variable speed now, so the tinkering goes on and on and ........
From the other stuff I’ve seen you put on here, you’ve got it covered.
Al

[bojan]

I cleaned it for presentation :-)
Yes, tweaking/adjustment things is always necessary before start of making anything..
BTW, I updated the code a bit, however there are still couple of details I would like to fix before I will be satisfied - one of them is the consequence of keypad multi-functionality.. I have to implement adequate de-bouncing to this essentially analogue keypad.

[JimsShed]

I like the way you used a diameter symbol instead of a "Y" on the Y axis. Unambiguous. No second guessing when you're busy.

[alval]

Dont know what language you are using there, Im sure the debouncing Im thinking of has nothing to do with the one youre talking about 😂.

This is a picture of the archaic (literally coming off the arc) motor control system on my lathe, it only allows for forward and reverse, with a timer when switched off so I cant immediately reverse. 40-50 years old. Hence why Im looking to convert to variable speed, if this dies not sure Id find the parts to repair.

With your variable speed do you find you still have the torque on low speeds? Ive heard varying accounts on this, using slower speeds on larger diameters tending to bog down. Its not an issue on the geared system.
Al

[bojan]

Yes, this is pretty old.. no electronics whatsoever.
The term "debouncing" is used to describe the behavior of the push button switch.. because electronics is much faster than closing contact of the mechanical switch, it detects every "bounce" and interpret them as multiple pressing of the switch.. which leads to unwanted behavior of the system.
In my case, "Up" and "Down" buttons have two functions, depending on previously pressed or not "select" button. So, when I press "select" then say, down, it executes first command (setting ref for Y axis) and immediately after that it executes diameter ø or Y. So I have to slow down the buttons somehow so only first command is executed. Not a big deal but it could be annoying.



This lathe has brushless permanent magnet motor, so it is supposed to have speed proportional only to applied voltage. But in practice, torque is significantly lower at very low speeds because of coil resistance (I can stop the chuck by hand) but I think it is not a big problem (unless perhaps cutting the thread on big diameter).
The better solution here would have been a speed controller with encoder and feedback (interestingly, this lathe has encoder, but it is used only for speed display). Maybe this will be one of my future tweaks :-)

[alval]

Might have to have a rethink, cant have it stall when turning 6 inch cast iron flywheels for my model engines. Low speed high torque is required. I suppose if it aint broke I should keep using it, been working fine for me for twenty years. Just annoying i have 8 speeds, 4 high 4 low. The two used most are high low and low high (as luck would have it) so I have to pull the belt off and change a gear quite a bit. Its worked til now, and Ive got it down pat like a speed change. Just wishful thinking I suppose and getting lazy.

[bojan]

:thumbsup:
However, if in the future you stumble on suitable motor and associated electronics, you could go for it.

The controller board I have is very basic and low cost solution, without speed feedback (only overload protection is implemented.. which doesn't activate at low speeds anyway)

[bojan]

Couple of bugs in arduino "sketch" fixed in this version...

[JA]

Hi Bojan,

I too am looking at a mini lathe possibly the same model as your lathe as it has a good collection of features/size, not least the ability to cut 0.75mm pitch threads for astro / camera adaptors so when I saw your comment wanted to ask: Were you using the 50-1250 RPM or 100-2500 RPM range when you noticed the low torque at low speed?

The reason I ask is that I was considering exactly the scenario you mentioned: cutting some larger threads. I suppose one could be gentle with the depth of cut and use brass or Aluminium rather than steel to try and accommodate any issues. I'd be interested in your thoughts on the issue as you've observed it directly and stopped it with your own hands, how hard I don't know.

Best
JA

[bojan]

I was using 50-1250 range.. where torque is higher due to higher mechanical reduction. It is pretty easy to stop it by hand...

The issue is in control board, the motor speed is determined by voltage (generated by PWM from processor) and there is no feedback.
You are welcome to have a look yourself - PM me so we can arrange something.
Cheers,

Bojan

[JA]

Will do, thanks Bojan.:thumbsup:

Best
JA

[Constant]

No fan of the expression "bucket list" but.... I'd love a metal lathe. Power to your arm, well done!