Actually Building the Machine!

[Nena]

Even that i have done a lot of programming for the algorithm it will still be a pain in the *** to controll the machine. But there are several way's to go to Rome Mayby ill try it with Arduino, sounds silly but Arduino is a good stable C platform witch is easy to program. The thing is i dont need to use all axis at once, there threatet all separately at first stage. When i want to go for other things like rounded facets i should use 2 axis at once but this schould also be no problem becourse you work with triangles. Even more fun is to use the sines/cosines then you can calculate both at once. When you use encoders like what i want to do is that you let the machine run on the encoders. (Only possible if you run not tooooo fast).

But... I use 2 sensors directly under the bearings for testing sound if the stone toutches the lap, i can make a measurement if i'm actualy cutting or not. I hope this works by the way lol. Not tested yet. Thus there are more aspects to look at.

Buyed today some more linear rails today, could buy a lot from somebody i know, hopefully there is some usefull guides for my stone cutter. Also i need to build that one myself. My dad buyed one for 1200 euro's a long time ago and he uses the machine now for everything you shouldnt do on one lol. I can build one under that price. Second hand is a problem here in the Netherlands becourse there are so little people that uses one here... and a new one costs a lot of $$$.

Greets Christian

BTW i added a aluminium extrusion to it but i think i need to use a another linear rail for the mounding, i can better use slot 10 for the mounting and use 4 guide cars to hold the weight. Why one guide car? Becourse these cars may have 2000kg of force on it. So one car is inaf but if 2 needed ill add two The 4 guide cars are for under the aluminium profile for the X axis. The extrusion is 90x90mm.

Also still have litle problems to add my ballscrew mayby i need to search for another one.

 

[vonG]

Funny you should mention Arduino, was locking at using that myself. But then I thought it couldnt do what I had planned. Can it only drive one axis at a time or did you mean that it doesnt have to?

I have some ideas for the depth of cut that I can share with you if you like, doesnt look like Im going to need it myself anyway. I think it would be something that you could use in your machine.

By the way, what do you think the pros and cons for the vertical vs horizontal lap is? Why use vertical?

Did your dad by a faceting machine or what did you mean? A new UltraTec cost almost 7000 euros if you include the shipping cost when sent to Sweden, and that is just for the machine. :|

Same aluminium extrusion I was looking at for my small machine. My machine is only 900 x900 (mm)

 

[Nena]

I have a simple calculation that i found out by mistake It uses the mid point from the stone and the angle and with the cosines from those two i can exactly plan my depth of cutting the stone.

Machines are expensive here in the Netherlands as well at your place. My dad has build his own for faceting but he have buyed a saw new. And the saw was 1200 euro's that time. (2500 Gilders/Floryn). I also need to build one. I can get his faceting machine but i wont, too mutch error in it lol hehehe but he did make good stones on that machine!

About vertically cutting,

For automatic faceting it is easyer to cut vertically becourse you dont have to control the Z axis ;- ) And you have more stiffness to the machine!

My machine wil be around 1000x600x600mm And ill use most of it like 300x600x600 Dont have to move the head everytime 990mm to the left and right! LOL

 

[vonG]

But then you dont know when the rough stone first hits the lap, just an estimation. When preformed the stone first it will off course be easier.

My way you will know exactly when it hits the lap and how much it has taken off the stone after grinding. And how much force it has used to do that. That information can be stored in a databank and used the next time cutting the same mineral.

Aha, so it was a saw. There are cheaper saws than 1200, all depends on what you need. But brand saws are expensive.

How will you manage the water and oxides or other lubricants on the laps when its vertical? Will the water not spin of the lap easier when vertical?
How do you clean the used laps if you are going to have it auto changing the laps?

 

[vonG]

Oh, so your machine isnt going to be that big. I thought it would be gigantic, specially when you have that head that weighs so much.

 

[Mr Magoo]

Hi Nena.

Watching with interest from afar. I assume you will be using differential encoders, yes?

I've just finished a project to measure the angle. The encoder I used was 4000 PPR (16,000 CPR) quadrature optical encoder giving an accuracy of 0.0225*.

I used a program called 'Flowcode' https://www.matrixtsl.com/flowcode/ . A very simple to use, fast program used by amateurs and professionals alike. Can be used with ARM, AVR, PIC and PIC16. Capable of CAN, I2C, SPI, RS232, LIN etc, etc. I am no programmer. A spell of ZX81 basic would have me stumped.

Here's a screen grab of a routine that 'get's' the encoder count, converts it to an angle that can be displayed to two (my choice) decimal place and send it to this screen http://www.4dsystems.com.au/product/uLCD_32PTU/ by RS232 (shown in the open window) then reset a flag.
Ironically it takes more effort/characters to write this explanation than to do it.

 

[Nena]

For laping i use diamond laps also for polishing 200.000 gritt is i thougths the highest so i wont use oxides. First on cheap laps then i want to go for sinterd laps. The two sensors in the axis tels me when the stone toutches the lap. So there is no guess in this part the pressure will be a problem becourse i can push te stone with a force of 58Nm to the plate lol an momentairly over the 150Nm it means that i can press the head trough de lap lol. I hope that wont happen

Im all ears what you have in mind!

The distance from head to lap needs to measured each time when i change the lap and cleaning should be done by hand or by over flooding lol

Preasure measuring is a problem becourse the stiffness of the machine!

About the arduino is called i2c master Goes to the pc and handles the slaves... Works perfect and easy so you can get to the 200khz for spinning your motors

 

[Nena]

Hi...

I dont use encoders on ppr the encoders i use are absolute encoders with 1.200.000 positions per rotation. But i have looked for some biss encoders as wel. But i have problems to read out the signal. The one i use is almost 20bits and gives already problems with direct usb.

 

[vonG]

Maybe I should look more into Arduino then, didnt think it could work with the machine. Thanks.

There are ways to measure the force against the lap, just have to get the right pressure by trail and error first. Measuring the impedance force of the motor on the x-axis will give that. Maybe that doesnt work with your encoders and steppers though. Works with permanent magnet dc motors.

The size of the rough stone and the location on the dop can be measured up as a 3d model using a simple cnc tool height probe, probably easy to make that program with your programming skills too. You measure several spots on the rough stone as you would with a CMM machine, but with the probe firm on the table instead.

This way you can move the cad file over the 3d of the rough stone and not lose to much of the stone when cutting it too.
Just as in a diamond bruting machines.

When doing the precut you go out and measure the stone again after each facet and then you will know how much you have taken off on that facet. Compare it with the original rough stones measurements.

The distance to the lap can be measured too with inductive distance sensor probes put on the side of the faceting head.

 

[Nena]

Hi Magoo,

I use absolute magnetic encoders for positioning (extra) i already have encoders on my motors but for exact positioning i use those ring encoders. They are very high resolution. They are expensive but i can set every thing as absolute, even when i turn off the power and back on then the sensor still knows its position. If you use ppr encoders you need to home everytime your position and thats also a leak in the system. And its not graphical inaf. The ring has 1,048,576 positions. An accuracy of 0.1* but i have a accurate reposition whats more important for me. If you need to home your system everytime it also gives you a loss in accuracy. So every encoder/system has problems. What i want is that i can reposition the complete system exact even when i turn off the machine. I always know where i have set my encoders on.

 

[Nena]

Hello Thomas,

I want to do it way more simpler lol. Check the sound if its cutting or not. I dont use cad or any 3d files, my software is generating the stone like you do on a lapidary. The endresuld is the cutted stone. I can change the size and the refraction index in a second. Also drawing a new stone is a metter of a few minutes. Or i download a gemcut file and convert it. The easyer the better! It can handle verry complex cuts im unlimited with the total facets or size or... I dont have to spend each time a half an hour to change a cad file And thanks to those encoders i dont have to keep everything in the memory. I do use a MySQL database for cutting and setups/photo's/materials/errors/etc.

Greetings,

Christian

 

[vonG]

You should go with whatever works for you, and the setup you have is pretty advanced in my mind.

How do you determine where to start cutting on the stone or evaluate how much you will lose of the rough?

By using a 3D image of the rough stone you will not lose that much of the stone when cutting it, will even get the exact position on the dop too.
My plan was to cut for value and not as a souvenir sitting in a cabinet somewhere.

I think your machine is going to work very differently from all the other machines on the market, and that is probably a good thing.
The encoders memory is a really good feature, less risk of damaging a stone if you get power outage and get back to where you were. If that should happen your vertical lap has a very big advantage too.

 

[Nena]

I dont know how the others do all the electronics. But im still searching for a good solution for the distance measurement to the lap. The coolest way is to use a distance laser for measuring. But havent found a nice sensor to work with with measuring in microns.

The encoder does have a memory but even when the power is turned off and i move its position by force and add the power back on, it has a new position and the sensor knows its new position. Realy its a great product Look at this movie:

[video=480,360]https://www.youtube.com/watch?v=P4hPnFMDB3Q[/video]

I use 2 of them for the rotate axis.

 

[vonG]

Haha, that explains the price tag on it.

The one I have seen use an inductive distance sensor, it will give an accuracy of 1-5 microns depending on the sensor used and the sensing distance. But the lap itself will not be that plane anyway.

One of the problems with these sensors is the material it can measure on though, have seen some complain about it on a different forum

 

[Nena]

I think i can do it better, using a laser sensor mouted at the back of the head and move the head till its breaking the laser beam and read out the x axis sensor. Then i should have the exact position of the lap. There are again more ways to Rome lol

 

[vonG]

Even if you would get a better measuring the lap will still not be completely plane. Measuring from the inside out on the lap will give different values, taking an average is not good but maybe use the highest to be sure not to cut to much on the stone.

All these sensor readings, laser or inductive will go to hell if they would get lose from the head and give wrong readings anyway.
Having a fail safe in max/min distance will probably fix that. The same goes for when cutting of the stone.

 

[Nena]

The best solution will then be to measure all laps on a ceramic flat table and insert those in a db hehehe, the lap isnt flat if its new and cant be flat. A scratch is already a problem. So i think i need to be a bit serieus in what i am doing. Thats why if i cut a stone under the 0.01 accuracy i would be verry glad

Things i like to do with automatic faceting: Fantasy Cutting! (also normal cutting) for learning. Professional cutting and large cutting. It would be verry cool if i can cut a large quartz from a cm or 10 with more then 1000 facets, or doing crazy, funky facets. It would be cool to make lenses as facets... So i dont want to be limited by cutting standard diamond shaped forms. I think that is also the reason i want this, not being limited by a standard faceting method.

If i can get the 4 stars for a certificate i would be verry happy! But I Luv Facetted Stones! Making from your hobby your job!

 

[vonG]

I like your dreams about further progress of the machine. Concave faceting shouldn't be that hard to do, needs more axes though.
Have you seen any stones with that many facets? The light reflection doesn't work that great on them in my opinion, to many facets.

Crazy facets is with the odd 77-index gear. That is why it didn't work in your calculations earlier, it's for odd calculated designs. There isn't to many of them though.

A clear quartz with a "normal" facet in the 400-600 ct range is worth a bit, but the rough will cost too. As long as there is a market for it it's worth cutting, don't want a shelf warmer.

I would go for lab grown rough in odd colors or the colors that's in demand, clear stones with special designs. The concave cut stones will go fast.

There are to many things that can go wrong with real rough stones, not to talk about the prices for the good graded ones. Imagine having the machine damage a 5-10k stone.
That's why expensive stones are cut by hand.

 

[Nena]

Still have several hundret stones laying by my dad, many tourmalines, ruby's (4kg), emeralds, hauyn... etc. But lab grown stones are oke to but the large stones are more quartz stones.

Ill see what happens, even when i cut 5000 stones and sell those for 1 or 2 euro's each i can get my machine costs back I just hope its going to work as i expect. But not only that, i also want to cut microscope slides. But if it's going to work to do large stones it should be easely to get my machine costs back. First prior lol hehehe

But theres always possibilities to get it all working perfectly. The first prior is to get it all working. How i going to do it, i dont know it all yet but i can come far if i just do the things i have in mind.

I wont do stones with a value of 5 to 10k only make them lol hehehe dont know how far i should go. The chance it goes wrong by machining is small with care but the stone itself is the problem. I dont know where im going and where im coming. If some one give me a bunge of $ mayby i sell the complete machine. (Already got an offer, to low but i got a offer from 60k but 40k on parts... wich i wont get anymore). So i will make my own stones dont forget the tax authorities are also looking then (52%) lol!

All the fun on one stick...

My plates are hopefully done next week!!! Already asked another friend of mine for the bearing house and he will do it... So i can continue soon with the machine!

To get an idea of the size of the head, this is my linear guide with one car on it! +/- 6kg! Heavy lol!

 

[vonG]

Damn thats big, and heavy.
Wouldnt the frame be better off made out of steel instead of aluminum profile, I mean the head is so heavy and when time passes the weight will have its tear on the frame. Then again the frame isnt that long so it should work.

Hope you have buyers for all those stones, always easier that way.

A guy I know told me once that you can cut the best stones in the world, but if you dont have any customers for them they will not be sold. At least not for the right price.
Its easier to cut stones after demand and then do special cuts on the side to show.

Im guessing that you want an extra 0 on the offer for the machine when its done.