hi guys

OK , i see one familiar name in this forum and some Aussie like me.

but the true run down is im a motor mechanic and have some free time on my hands

I love drag racing and have a few fast cars and i want to go faster , presently building a 1200Bhp plus turbo 6 cylinder 4 litre falcon E series motor. will find out if i make the horse power once i get it on the heenan and froude engine dyno i run at work.

I'm in the process of contemplating ( have started converting , stripped the old lathe and just fixing her now so it looks good at least ) this old copy lathe to a CNC_lathe . I know it will not be good and fast as some of the CNC_machines out there in the market place like the Okuna's , Mazak' , Puma's , Hitachi Seiki's , Mori Seiki's and Pinnacle's and the list is endless.

But if i had the cash i wouldn't hesitate to buy a MIlling center and a lathe in cnc setup. Maybe in a few years time

It's more a hobby at this stage for me , i want to be able to make my own parts at a slow pace .

i see a great deal of knowledgeable people in here and cnczone.com and many other good websites.

so here it goes i have this lathe and have attached several pictures of it , it is a very robust unit so rigidity is not a issue.

Will have to make all the brackets and adapter to couple the stepper motors to the lathe. This is not a issue as i have other lathes and mills at work. But the problem is im new to electronics and CNC is not my strong point. Some sort of guidance would be appreciated as to the approach i take on doing this conversion .

Yes some may say its a big machine but i thought why do a bench top lathe when i can start on the big lathe and be done with it .

Its just a larger version of the small bench top unit and i want to be able to turn some bigger parts than what the limits of the bench top lathe will permit.

I plan to convert a pacific U2 universal milling machine also to a 4-axis cnc setup also which is sitting in a section of my workshop probably tackle them both at the same time

I'm one that does things BIG , now the stepper motors i had planned to use are http://www.kelinginc.net/SMotorstock.html ( NEMA 34 HIGH TORQUE STEPPER MOTOR 1812 oz-in, ½” Single shaft with flat, 180
KL34H160-62-8B ½” Dual shaft with flat: Price: $189 ( in stock)

this crowd has drivers and power supply also am waiting on a email back from them as we speak.

these stepper motors are probably over kill for the lathe and mill setup but for the price there selling them i figure just spend a little more and get some decent horse power steppers.

again im open to criticism and suggestions here guys , what are your thoughts i need all the help i can get am a fast learner , did bachelor of applied science , all math, science and computer studies at university back 20 years ago and kick my self in the head i didn't finish schooling . Did FORTRAN77 , basic , Pascal program writing back in 1984 now i am showing my age


1. stepper motors or servos ? size ? brands ?
2. stepper motor controllers ? servo controllers ?
3. power supply ?
4. spindle speed controller ?
5. Ball screws and nuts/flanges ? size and specifications ? what brands ?
6. what software ? sorry im not being biased here , i am looking into " Mach " ?

any help will be great here guys

and i'm extremely sorry if some of you think i shouldn't have posted this thread in this section , looked all over and thought this would be the best section for this post

there is no chronological order that i have written this post so correct me

cheers

Off the cuff, I would say to avoid using stepper motors on large machinery, unless you can find a closed loop controller system that will detect and correct for lost steps. I do not think Mach will do this yet.

Ajax cnc or Camsoft cnc, that is a couple of the more often mentioned names around here.

The level of precision required for a useful lathe X axis is very high. This requires lots of steps, and to obtain decent rapid rates, you need a controller that can crunch lots of numbers, fast. Camsoft, and the Galil card that it often uses, can count as many as 12 million encoder counts per second.

Ajax can likely give you some advice over the phone about what you've got already that may work, or a complete system.

Camsoft, well there you'd be able to use a bit more of your programming smarts to write some logic. It takes lots of time, but, it is a versatile system. If you are going to do a mill and a lathe, you might get double use out of a Camsoft system and the time required to get it running. As a bit of a tinkerer hobbyist, you might enjoy the battles to make the system work as you envision.

Sorry, but I know of no really cheap way to do a professional quality retro, and on a larger lathe, you don't want any whoopsies coming out of the woodwork that did not originate in your g code. :)

hi

We in Melbourne Australia have a newspaper called the " Melbourne trading post " , this newspaper has categories of different things for sale from cars to house hold goods to tools and machinery. Yes they are listed on the net and i was reading it and come across some guy selling ball screws.

I telephone him and with a few minutes i was on the road to his place.

He lives about a 1 hour drive from me on the freeway so i zoooooooooooooooommmm down to Geelong, to my surprises 3 of the 4 ball screws were NEW never been used or taken out of there wrappers.

The ball screws are German made by Steinmeyer and the purchase order papers come with the ball screws .

The ball screws are 40mm units , this one i have taken pictures of is 600mm long. the others are roughly 2000mm long . Does anyone read German all the documentation is in German ?

One of the longer ball screws and nut/flange was seized due to the fact there was a lot of grit on the shaft and i imagine inside the nut also.
again this size ball screw and nut is over kill for my lathe but they cost me for all four $150 so i couldn't not complain nor resist buying them.
Now that i'm on this forum , not be harsh on me when i ask this question , i'm new to CNC machinery and computer drawing software and software that controls the CNC machinery .

are there any threads a guy designed a tool change for a lathe has anyone got any other ideas of there own , would love to hear if you do and any pictures or information on a CNC_tool changer for a Center lathe.

what is the best softawre to run a CNC_lathe in 2 axis ? and why ?

Is there a demo version of the software on here that i can have a play with ?

cheers

hi HuFlungDung

I/m confused now mmmmmmmmmmm

why is there any difference between a small ( desk top lathe) to a 2 tonne monster and bigger ?

other than driving 100 ozin to 2000 ozin stepper or srevo motors

i had a look at theses two websites Ajax cnc or Camsoft cnc thanks for that HuFlungDung , i will weigh the pro's and con's of all the systems out there

cheers

hi

anyone used this card

http://cnc4pc.com/Multifunction_CNC_Board.htm

what other cnc drivers are there , guys any help would be fine


cheers

I would go gecko's and Servos. Forget the steppers.

Eric

FPT,

Hi, I make the multifunction board that you posted earlier. That board comes with many bundled features that will make your life easier when retrofitting a lathe. I have other goodies like relays and index pulse index that you will find very useful.

There are a large number of options depending on what you need and want. If you are going to need high speed, and budget is not a problem, you should go to centroid, camsoft, galil, or others.

When doing lathe work, you might need speed on your axis is you run a fully automatic machine that needs to travel to each tool position. Geckos and servos should work ok, if you are not very high speeds. Consider steppers only if you are not going to do lots of traveling.

Arturo Duncan
http://cnc4pc.com

The bit about closed loop back to the software as opposed to just closed loop to the card or driver doesn't fly for me. If the machine delivers the product, and an error signal shuts things down when the servos can't keep up, then all is good.

There are options to take a pulse signal to a servo. I have camsoft, but I could have achieved the same results for less money. One thing camsoft has going for it is I/O and logic, but if I was turning bar stock and just needed to turn on a couple of devices, I wouldn't worry.

I think Mach with Gecko servo can keep up.

RE toolchanger, that's getting to be out there in the realm of lots of extra credit. Consider gang tooling first. You can mount a rear tool and get to 2 tools easily. After that you could envision 4 tools. After you've gotten proficient with that, you will have figured out how to do the toolchanger or more likely bought a lathe with one.

I will say that I have heard it is very hard to retrofit a new control to an old CNC lathe. Your project is not that, and I have never heard the details of why it is challenging, but I have heard it more than once.

The last thing I will mention is that lathes are a bit of a stepchild for Mach. I wish it wasn't so, because I am also working on a CNC conversion, but the lathe functionality always significantly lags the mill functionality, and I mean by a lot. I am hopeful that Art will find more time to work on the lathe piece if more people get interested.

Best,

BW

Get the tool changer right outta your head. No need to concern yourself with changing tools automatically until you got the lathe working under cnc control. One challenge at a time man.

E

hi

need some form of a tool change to drill , bore and internal threads and tapers

I saw on a small bench top lathe that i nearly purchased but realized i was going to pay way to much for what i was getting after i did my homework

The bench top lathe is a training lathe a company in Adelaide Australia makes them for tech schools , way over priced i could purchase a second CNC old mazak or okuma for the price they wanted for theres.

But of the topic now the little lathe had arranged on the X- axis cross slide 4 tools in there holders , the way they where orientated mad it possible to machine large objects in the 3 jaw chuck

will try and find there website and upload a picture so you people can see what im am talking about. nifty little setup

At this stage i do have a budget to meet on this project and it seems everyone has divide differences on what software to use and hardware ?

Seems everyone is saying stay away from stepper motors and use servo motors ??

What is a servo motor like on the insides and one have a picture of one pulled apart ?

I have pulled apart small stepper motors so know what there construction is like but can not find anything on the web to see inside a servo motor cause everyone is raving on about servo motors , there the best thing introduced to cnc machinery since the invention of sliced bread


cheers

FPT,

Hi, I make the multifunction board that you posted earlier. That board comes with many bundled features that will make your life easier when retrofitting a lathe. I have other goodies like relays and index pulse index that you will find very useful.

There are a large number of options depending on what you need and want. If you are going to need high speed, and budget is not a problem, you should go to centroid, camsoft, galil, or others.

When doing lathe work, you might need speed on your axis is you run a fully automatic machine that needs to travel to each tool position. Geckos and servos should work ok, if you are not very high speeds. Consider steppers only if you are not going to do lots of traveling.

Arturo Duncan
http://cnc4pc.com

"I retrofitted a Hardinge HC CNC lathe that had an old bandit control. The machine was working OK with the bandit control, but I had the same problem as you have. Dealing with old electronics and an awkward control (compared with what Mach3 can do.) This lathe runs fully automatic with 10 feet power bar feed and an 8 post tool turret.

Here is a brief description of what I did and how much it coasted:

I used a PC with two parallel ports. I used two cnc4pc bidirectional boards (that I make and sell). The second one was set to use pins 2-9 for input, so I had a total of (5+13) 18 pins for input, and (12+4) 16 pins for output. The cost of these cards is $26 each. I also used an A4 Power supply (that I also sell) for providing 5 and 12vdc that is used to power all the circuits. These power supply costs around $29.

I used the original power supply and motors, but, replaced the tach and resolves with E2 2000CPR encoders for x axis and 1000 CPR for y axis. These encoders are about $32 each. I also replaced the old drivers with G302 from Geckodrive, this cost $114 each. I also replaced the stepper driver that runs the motor for the tool turret with a G201, also costs $114.

I replaced the nice 500CPR encoder that was mounted on the spindle with a C3 - Index pulse card (that I also sell). This card costs $16. With this index pulse I am able to thread and run a closed loop cycle on my spindle. I replaced the awkward speed control with a Hitachi VFD that costs $330, and also used the C6 – Variable Speed control board (that I sell). This board costs $27. With this board you configure your spindle just as you would a step and direction driver, and with the index pulse feedback you have 100% control of your spindle’s speed.

I also replaced the mechanical home switches with the same Index pulse card that I used on the spindle. That way I get a very accurate homing. For the turret home I still used the original mechanical relay that came with the machine. There is no need for precision here, since the turret has a turret lock that holds the turret in place.

I also used the C5 – Solid State Relay Board (that I sell for about $46) to control; turret lock, coolant pump, collect closer and power bar feed, and the cut-off tool.

I also use a C4 – Safety Charge Pump (that I sell for $17) to control the system along with the e-stop switch.

I also installed a 150CPR encoder from us digital (about $32) that acts as a manual pulse generator.

The only thing that I miss from the old control was the speed of the servos. In terms of control I lost nothing and gained lots of features. I currently plan on installing proximity switches to program tool checks o other things. The total investment was extremely low compared to the benefits that I achieved.

Arturo Duncan
http://cnc4pc.com "

Hi Arturo Duncan

now i read this quote you posted in this thread link http://www.cnczone.com/forums/showthread.php?t=20895&highlight=big+lathe page 2 of the thread u posted what i have highlighted in bold blue and visited your website and you have all the goods i need to convert this manual lathe to CNC_lathe , i want to use either a laptop or a PC which there are heaps around the place . I want to be able to basic turning , boring , taper_turning internal and external, thread cutting internal and external .

Arturo Duncan i'm new to all this and ask you would you be interested in giving me guidance in a complete conversion on my old_copy_lathe to a cnc_lathe , I will purchase what is need from you and assemble it at work my end.

I ask the same question like NC Cams
Master Machinist , it is just a lathe with a X and a Y axis a 2-axis , even on OLD CNC machines why cant you just bypass the old fanuc controller and just use the stepper motors/servos/ballscrews and yes some of the hydraulics to operate the chuck(s) , tailstock . coolant pump and any other things that need to be switched on or off , lights , door safety open etc ????

I have spoken to a few people my end of the world in Australia and some of the responses i get is how big is your bank balance , and i look at them and say none of ur F@#Ken business. Everyone wants to get rich quick.

And some of the sales pitchers i get when i tell them what i want to do with my manual lathe they all say its impossible to do ??? and when i say why can people convert small desktop lathes and mills with succes they just ramble on all this bulls@#t to me . I PMSFL

lets see what happens here

cheers

hi

Bought another center lathe with gap bed and a 3 meter bed , come with some bits , a 3 jaw chuck , 4 jaw chuck , 3 jaw soft chuck , 5MT live center , a between center bung , 4 chuck keys , tool post spanner, travel steady and fix steady. Missing a face plate

i need to adapt the chucks to the head as it had some camlock chuck arrangement on it , will be easy job and just go over the machine and a coat of paint and she's is ready for work

I have some big rollers i need to machine up , i want to make a sheet metal roller and a few other machines that i need.

and no i don't want to convert her to CNC LOL

I saved her from the metal scrap heap and i have a few jobs i need to do on her , she looks good and I will get her running right

cheers

woooooooooo

had a look a MAZAK, MULTIPLEX 6200, LATHE, MULTIPLEX, OPPOSING SPINDLE beautiful piece of machinery

if ii had the money would not wait one minute more would buy one straight away

tomorrow is saturday here in Australia and the big tattslotto draw is on first prize is $20,000,000 ;) will buy one ticket for that draw

just in case i win ;) and forget about the red ferrari

cheers

hi

has anyone had a look at doing something like what this guy did ??? Converting an Ordinary DC Motor to a Servomotor ??? have a look at his website http://truetex.com/servomod.htm

cheers

When I say use gecko's and servos, that is what I mean. Gecko 320 drives www.geckodrive.com and ordinary DC motors with encoders attached.

BTW I am not blowing wind. I have converted a milling machine using Geckos and servos (DC motors with encoders attached) and Mach 3 for software. Here are just a few parts the mill has made. I don't sell any of these products.

Beware of taking advice from those who are trying to sell you something....

E

When I say use gecko's and servos, that is what I mean. Gecko 320 drives www.geckodrive.com and ordinary DC motors with encoders attached.

BTW I am not blowing wind. I have converted a milling machine using Geckos and servos (DC motors with encoders attached) and Mach 3 for software. Here are just a few parts the mill has made. I don't sell any of these products.

Beware of taking advice from those who are trying to sell you something....

E

Hi balsaman

had a look at the gecko drivers and many other drivers and stepper and servo motors. Seems the way to go but there are many other options out there.

Balsaman what software are you using ?

what are the advantages of servo motors as opposed to stepper motors ?

I want to make fuel injected parts , trumpets , fuel rails , fuel bungs , fuel injected manifolds , throttle bodies , webber style throttle bodies , holley carburetor style throttle bodies with a square bore base 4 venture in different sizes .

Also for a particular vehicle want to make and modify front stub axles for larger brake conversions.

finally found the link to the australian guys in Adelaide that make small cnc lathes and mills for training tech schools , not cheap link is http://www.axmell.com.au/Hercus/CNC%20PAGE/CNC_Index_Page.htm

they make some sort of four position TOOLPOST changer "A choice of Tool post and holders are available. The NOVIM is fitted with a four position manually operated toolpost. An eight position Automatic turret is an option. The turret is activated by a PC command and is bi-directional. A quick change style toolpost is also an option. "

i spotted a different setup on a lathe some guy was selling out my way where on the X cross slide there was 4 different tools holders positioned on the table where the software of the lathe selected which tool to use

The lathe that i want to convert to CNC-Lathe i will be mainly doing turning around the chuck area but at times will be turning ford 9 inch axles to the specs that i need as i do a lot of 9 inch differential modifications for different Australian cars have even done one for a 240 Volvo and some other weird cars.

So i have to machine the back of the flange and flange diameter bearing and seal diameter and the axle shaft length where i cut the spline , so also machine the spline diameter.

Had 5 differentials to do in the last week , manually machining theses axles is very boring work , this way if i had a cnc_lathe i could program the different axle lengths for all the different cars

cheers

Servo motors don't have resonance issues like steppers might/can/do.

Closed loop feedback to the control helps assure that your machine is in correct position, and is not slowly getting closer and closer to the chuck with each part, until you start to wonder "what is that ticking sound?" :D

Steppers are good to run stuff that doesn't have highly variable loads, or are non-critical in nature. A cnc lathe under control is still a dangerous piece of machinery, and the bigger they get, the more damage can be done. You want the most reliable control you can get, not something that may or may not move exactly where you told it.

FPV,

What I did was a bit simpler because I started with a lathe that had all the mechanical stuff for cnc already in. What I did was to replace the old control with a PC and a box with all the electronics that I make. This lathe is where I test all my stuff. I know I should post some pics of it. It is 100% functional. The only thing I miss is the very fast servos it had (the servos it currently has are the same, but I changed the drivers to G320s). The high speed was nice, but I really do not need it. You explain it very good with the Ferrari example. In my opinion the axes speed really hurts while tool changes. These are my opinions on the most relevant aspects of the lathe conversion:

1. Axes motion control. You want to use steppers. Check out some of the videos for CNC machines using stepper, if you can live with that, go ahead. For a couple of bucks more you could probably go with servos and probably do 50% faster. Geckos are great, but if you’re in Australia, you might consider rutex. I have no experience with them.

2. Tool Turret. Some people just gang up the tools on the table. The good thing is that it is simple, the bad thing is that you need a long x axis to be able to fit all the tools, you are also limited in space; you probably always take the risk of crashing a tool. It probably also has to do with how you do your programming. Another option is to use a 4th axes cnc indexer, John Stevenson did http://machsupport.com/forum/index.php?topic=970.0 . Keep in mind you MUST lock the turret while cutting. You MUST use some kind of index sensor to make sure the control does not loose control of the turret. My machine has a pneumatic lock and I use the index pulse card for homing the tool turret. Works fine for me.

3. Power Bar Feeder, Pneumatic Chuck, Cut-Off Tool. I have no mechanical experience with this, they came with the machine I have. I have seen many plans on the internet that will help build your own. I just replaced the old relays with one of my 4 relay boards. I also control the coolant and turret tool lock with these relays. It is all very simple, you would connect the relay just as you would a regular switch.

4. Control Software. I use Mach3, so far it has everything I need; tool tables, etc. I have written simple visual basic macros to control the turret and tool changes automatically. I created M commands for controlling bar feed, cut-off tool, pneumatic chuck, etc... Most are pretty simple, just using activate and deactivate commands, and sensors (index pulse card, and switches that already built into the machine when I got it). For example, if I did a bar feed cycle, the machine would stop until the chuck was locked and the sensor activated.

5. Variable Speed Control. This is too easy; I just bought a $300 VFD and connected with one of my variable speed control boards. This comes built in the Multifunction board.

6. Safety Charge Pump. A lathe is an EXTREMLY DANGEROUS machine. I learned this when once my PC froze (you know, windows xp). The chuck opened automatically and the bar feed pushed the ½” x 3 feet rod at 2000 RPMs across the shop. I always tell this story to people that are attempting to retrofit a lathe so that they take appropriate persuasion. Now days all the relays, locking cylinders lock all the devices in case of a power or PC lose. A safety charge pump (this feature is also built in the multifunction board.)

7. Threading. I use one of my index pulse cards; Mach does the rest, too easy.

8. I/O Many people will tell you that you cannot get enough I/O from the parallel ports to convert one of these machines. Keep in mind that you can always use a modbus device for additional I/O. I have a modbus device, but it is just for testing purposes. The machine runs with it. I have a two parallel ports, the second port has pins 2- 9 for use for input. This is how my I/O looks like.
OUTPUT INPUT TOTAL
LPT1 12 5 17
LPT2 4 13 17
TOTAL 16 18 34

The features I have are:
a. I control 3 axes using G320 with US Digital encoders on the motors that came with the machine. I use a G210 to run the stepper motor that runs the tool turret.
b. The machine has homing switches for all axes, including the tool turret, so the machine homes all the axes automatically.
c. I have encoder feedback from the axes fed back into mach. I do nothing with this info at this time, but I write a simple program that could compare the axes position in the control software and compare it to the feedback position, and take some action. I am not sure if some people will call this true closed loop, but for all practical reasons I guess it could work as closed loop.
d. I have an MPG, which is a 250 cpr encoder that helps me position the axes. Works very nice.
e. I have an index pulse sensor for threading. It also reads the actual spindle position.
f. I have VFD with one of my Variable Speed Control Boards that control the spindle speed. I have it configured as a closed loop system with the index reading from the spindle, the spindle speed is very accurate.
g. I mentioned earlier, I control the bar feeder, pneumatic chuck, coolant pump, turret lock, and cut-off tool.
h. The E-Stop is just an NC switch connected to one of the input of the cards. My Cards have an external enable pin that gets disabled by the safety charge pump and lock everything in case of emergency.

I believe this covers most relevant aspects of my experience. Please do not hesitate to contact me if you need further advice.

Arturo Duncan
http://cnc4pc.com

Paragraph #6 in post 19 is EXACTLY why I will NOT use any form of Windoze to control a CNC. Any O/S that lets the processor take control of what's going on so it can do what it wants/needs to do is a dangerous entity.

I can not be convinced that you need a multitasking machine for a lathe or mill controller. If Bridgeport could get a 386, 486 or Pentium 133 to run a mill, lathe, VMC (4 axis plus spindle speed) and surface grinder with the same interface card and DOS software in 1996 or thereabouts, why do you need Redmond's latest experiment in incopletely developed O/S to try to reinvent the wheel today???

PC104 back planes abound out there - can't say there is no more support for ISA. Pricey, yes. Not available/supported??? Hardly.

The way I see it, steppers are low cost and efficient controllers. However, they're trying to use fancy digital electronics to simulate truly analog motion. Adding encoders to warn of errors is too much too little too late. Take the encoder and use it up front, not after the fact.

Besides, sooner or later, the speed factor you're trying to achieve will result in resonance with a stepper - resonance that can't necessarily be tuned out. See the circle cutting problem thread elsewheres on the site.

Servo's on the other hand are essentially analog devices. If you've ever "driven" an RC car with digital servos versus analog ones, you can literally "FEEL" the difference. The digital servo moves in steps and shudders, the analog one moves in linear proportion and smooth fashion.

Need instant proof??? Take your mouse and make a perfectly diagonal line across the screen - it can't be done because it is a step fuction device. You get closer to perfect as you add steps to make smaller and smaller errors occur but you're still dealing with a step function device.

Servos have their own problems but torque and speed are not usually part of those problems. For a hobby type lathe, you could deal with sterpper issues. For industrial iron like you're working with, you're going to need a BIG stepper. At what point will you be approaching the cost of a servo and will it perform when it has to MOVE the iron???

My Fanuc powered lathes have 90vdc motors, 12 amp continuous. Rated speed is around 3000rpm. Torque output is huge due to the pulsed DC drives they use. If you plan on moving the carriage quickly, you need OOMPH. An equivalent DC drive today would need comparable voltage but probably 20 amp continuous/50amp surge rating due to the current surge. At that point, the only stuff close to being able to drive it would be one of Advance Motion Control's DC servo gems. Don't think Gecko or Rutex make anything that robust

DON'T buy anything until you decide/figure out what you need to have to move what you're trying to move. Look at similar commercially made CNC lathes and see what they used for power/drives/etc. Perhaps it might be expensive but, in an industrial setting, properly engineered stuff simply runs.

As simple as you'd think a lathe would be to CNC retrofit, the separate lathe retrofit thread shows that the task is frought with peril and unfulfilled promises. The reason lathes don't get the attention is simple economics - far fewer lathe retros so why bother???.

Too bad Bridgeport never did a Ezpath lathe retrofit kit. At least it worked and was well devloped. More than what can be said for some of the other lathe retrofit systems out there.

What is your budget for the conversion?


hi guys

OK , i see one familiar name in this forum and some Aussie like me.

but the true run down is im a motor mechanic and have some free time on my hands

I love drag racing and have a few fast cars and i want to go faster , presently building a 1200Bhp plus turbo 6 cylinder 4 litre falcon E series motor. will find out if i make the horse power once i get it on the heenan and froude engine dyno i run at work.

I'm in the process of contemplating ( have started converting , stripped the old lathe and just fixing her now so it looks good at least ) this old copy lathe to a CNC_lathe . I know it will not be good and fast as some of the CNC_machines out there in the market place like the Okuna's , Mazak' , Puma's , Hitachi Seiki's , Mori Seiki's and Pinnacle's and the list is endless.

But if i had the cash i wouldn't hesitate to buy a MIlling center and a lathe in cnc setup. Maybe in a few years time

It's more a hobby at this stage for me , i want to be able to make my own parts at a slow pace .

i see a great deal of knowledgeable people in here and cnczone.com and many other good websites.

so here it goes i have this lathe and have attached several pictures of it , it is a very robust unit so rigidity is not a issue.

Will have to make all the brackets and adapter to couple the stepper motors to the lathe. This is not a issue as i have other lathes and mills at work. But the problem is im new to electronics and CNC is not my strong point. Some sort of guidance would be appreciated as to the approach i take on doing this conversion .

Yes some may say its a big machine but i thought why do a bench top lathe when i can start on the big lathe and be done with it .

Its just a larger version of the small bench top unit and i want to be able to turn some bigger parts than what the limits of the bench top lathe will permit.

I plan to convert a pacific U2 universal milling machine also to a 4-axis cnc setup also which is sitting in a section of my workshop probably tackle them both at the same time

I'm one that does things BIG , now the stepper motors i had planned to use are http://www.kelinginc.net/SMotorstock.html ( NEMA 34 HIGH TORQUE STEPPER MOTOR 1812 oz-in, ½” Single shaft with flat, 180
KL34H160-62-8B ½” Dual shaft with flat: Price: $189 ( in stock)

this crowd has drivers and power supply also am waiting on a email back from them as we speak.

these stepper motors are probably over kill for the lathe and mill setup but for the price there selling them i figure just spend a little more and get some decent horse power steppers.

again im open to criticism and suggestions here guys , what are your thoughts i need all the help i can get am a fast learner , did bachelor of applied science , all math, science and computer studies at university back 20 years ago and kick my self in the head i didn't finish schooling . Did FORTRAN77 , basic , Pascal program writing back in 1984 now i am showing my age


1. stepper motors or servos ? size ? brands ?
2. stepper motor controllers ? servo controllers ?
3. power supply ?
4. spindle speed controller ?
5. Ball screws and nuts/flanges ? size and specifications ? what brands ?
6. what software ? sorry im not being biased here , i am looking into " Mach " ?

any help will be great here guys

and i'm extremely sorry if some of you think i shouldn't have posted this thread in this section , looked all over and thought this would be the best section for this post

there is no chronological order that i have written this post so correct me

cheers

Don't even consider the info on Ajax's site as valid for a budgetary estimate. Yes, they'll offer a $2500 USD or so package, UNTIL they find out what you want to drive. It will easily balloon to over $4k USD

Then, none of your stuff will work, buy ours, Or, they'll contend it is outside the realm of DIY and send you to the Centroid site - that gets into the real high 4/low 5 figure range in USD.

Fadal and Fanuc are in the $8K to $10K range but their lathes work.

A lathe is NOT a mill and you can't use mill hardware as a budgetary estimate.

Caveat emptor...

hi balsaman , between $4k and $5k hopefully ???

do not want to pay commercial prices , im looking at how these guys have converted little bench top cnc lathes

i know things will cost more for the bigger lathe , which is normal.

as far as software i have seen some cheap as $200 to $300 right up to $3000 to $4000 USD and i imagine there is even more expensive software around.
same applies to the hardware of the electronics.

ball screws and single and or double nut preload ball screws , i have seen a lot f variations again in prices

a PC or a laptop there are heaps of them around for next to nothing in price
old 486 will run most of the programs out there considering the old very early CNC lathes and mills had very old PC boards in there and functioned OK to a point until you overload the memory space on them LOL

machining and making brackets is not a issue for me i have lathes and mills at work and a friend has 2 cnc lathes and i cnc milling center at work so if i need some things machined it can be handled also.

lets see what happens , it is just good to see some of the peoples views and inputs thanks guys.

cheers

The ONLY and I do mean ONLY system (believe me I looked!!!!) that will run a lathe on "free" 386, 486 or 586/Pentium boxes was sold by Bridgeport as their Ezpath lathe with DX32 system (sold in mid to late 90's). THe same system ran CNC mills, lathes, 4 xis VMC's and surface grinders.

The Ezpath 2 or 2SD lathes would do all the trick stuff you'd ever want to do with tool changing, bar feeding, etc. This would be a DIY'ers DREAM, if ONLY it were offered on the aftermarket. The system involved a plug in ISA PLC board plus some external break out boards. These PLC based boards interfaced to industry standard servos and/or spindle drives. With simple plug in harnesses that any decent DIY 'wirer' could duplicate from a simple kit, it was darn near a plug and play system and very well integrated system.

IF (big IF) you could still buy the proprietary pieces and software from Bridgeport (don't think you can), you probably could do a retrofit.

BUT you'd have to do a lot of parameter engineering on your own to somehow 'tell' the system that it had miraculously been transformed from what it was to what it now would be. The parameters are NOT intuitively obvious to start hacking.

You'd get NOT a single word of help from Hardinge - they are clueless about support. I got lucky and the guy or two who did know gave me the name of field guys who knew the system and these guys were very helpful. They'd hardly be able to help you, however due to your location with respect to the USA.

Any factory trained Bridgeport service guy could EASILY tune the Ezpath system but don't hold out lots of hope. Even my service guy who respects what I can do with a machine and humors me with special support if I prove to him that it's viable, winced when I asked if it were possible to do a retrofit with this system.

Trouble is, there is/was too much that went sour when Bridgeport folded and the folks who know the workings of the system moved on. Sadly, they put it all behind them and do other things now. Hardinge is not doing much if anything to make parts availble for anything outside of a Bridgeport mill. Heck, they even use someone elses controller for their current "eztrak".

Cost if you want to do it anyway???

Who knows. All the parts are HUGELY explensive as they're no longer made by Bridgeport/Hardinge. The service industry is controlling the availability of remaining good hardware.

How??? By selling the boards on on an exchange basis (to keep control of the cores). Only EMI can really be trusted as they are staffed by some of the guys who developed the system and they service it exclusively pursuant to an agreeement with Hardinge.

I thought EXACTLY the same as you re; using a PC to do a retrofit of an existing CNC lathe with an antiquated controller. The "lathe retrofit thread" clearly outlines the challenges you/we face in the attianment of our goals.

Sadly, I've come to the conclusion that asside from Mach (NOTE: some of the early versions (especially the DOS versions) still in use are NO LONGER supported by the developers. Thus, Mach or EMC is/are the only inexpensive system out there. And Mach needs XP which is something I refuse to trust to run a CNC anything.

In my estimation step/dirction controls don't do real time "driving". They're always and at best, steps behind the parallel port even with servo feedback. The limited bidirectional data stream offered by the parallep port (even 2) is ultimately a communication bottleneck.

The machine will stop IF things get too many steps behind but the PC is still only working in step/direction mode as opposed to true direct control of a servo via a PLC ala the direct control path used by the PLC in the defunct Bridgeport system.

Good luck in your endeavors. I know what you seek, been there done that and still looking......

Mach 3 on XP (recommend a minimum of 1 Ghz. comp) is your best bet running Gecko's (or rutex) and servos on ball screws with a gear reduction using timing belts. You should be able to do this for under $3,000 easily, prolly less.) This would get you the best setup FOR THE MONEY.

E

Mach 3 on XP (recommend a minimum of 1 Ghz. comp) is your best bet running Gecko's (or rutex) and servos on ball screws with a gear reduction using timing belts. You should be able to do this for under $3,000 easily, prolly less.) This would get you the best setup FOR THE MONEY.

E

hi balsaman

there are still heaps of laptops with Xp very cheap and greater than 1 Ghz speeds

I plan to use large stepper motors or large servo motors havent yet decided

ball screws will be min 25mm shaft diameter and possible 30mm on the y axis

maybe.

Do you honestly think i will have to use some form of gearing ? using timing belts and toothed pullies and a reduction of 2:1 ratio gearing will required with such large torque stepper or servo motors ?

I just planned to direct couple the high torque stepper or servo motors to the ball screws.

But if i need the extra speed for finer cuts or speeding the cuts thats a different story.

Since this is my first CNC converions to a manual lathe , just a question

What ball screw pitch would be the best for the lathe X-axis and for the Y-axis ?? there are different pitchs in ball screws also coarse threads and fine threads so which is better for the lathe feeds.

I assume from my own experience that fine thread ball screws will give finer cutting per turn and the coarse thread ball screws will give more a cut per turn

All will depend on the speed the motors are turned and i don't know what adjustments or fine tuning features there are in the software , I plan to use mach3 for lathes.

What other lathe software is there avalaible at a reasonable prices for amateur cnc-apprentice-machinest ? ;)

cheers

FPV,

One last advice; DO NOT USE A LAPTOP FOR RUNNING MACH. I know they look very sexy on the machine, but there are many issues about the hardware, and how battery saving issues messing around with the processor’s speed. Please watch NC CAM’s comments. I am a very happy user of mach and winxp, and I do recommend it, but please do not take it for granted. Follow Art's recommendations on how to set it up, and please, use a dedicated machine. If you find that machine is not reliable with winxp and mach, please do not try to fix it, change the PC. I understand you must balance your budget with the final product you need or want. I think none of us can tell you if you need to buy a compact or a pickup truck. The only thing that you can not play around is safety. Do not attempt to play around pneumatic chucks, tool turrets, and bar feeders unless you know what you are doing. I know the way I say it make it look like wiring all the gadgets is pretty simple, and it is, but if make an error estimating how a relay, solenoid, or valve will behave on a certain circumstance, then you run into safety issues.

Arturo Duncan
http://cnc4pc.com

I dunno if the "Please watch NC CAM’s comments" in Arturo's prior post were accolades or criticism. At this point, a dose of reality wouldn't hurt to be considered.

Retrofitting a lathe is NOT easy. It becomes ESPECIALLY hard if you're trying to convert a manual one to CNC. I simply wouldn't even THINK of doing it an I've done some agressive machine conversions to my CNC lathes and mills.

I have existing CNC lathes. Simply replacing the controller with a PC based system has not been possible - what it would take to engineer and fit ball screws and drives to a manual lathe ain't easy even if one has an engineering background.

Integrating the new stuff with all the electronics is ANYTHING but plug and play. Finding and figuring out the encoders alone is a nightmare and I'm pretty good at finding PDF's for industrial equipment. WIthout a proven wiring and/or ladder diagram to go by??? Recipe for a disaster.

There are a plethora of PC's out there. Some run fine with XP and some don't. Some guys have trouble with parallel port driving and some don't. USB to parallel converters have had their share of problems. Like Arturo says, DON'T USE A LAPTOP!!!! I simply won't use XP. You make the choice.

In this case, the member expressed an interest in using legacy PC's. That pretty much precludes the use of XP based Mach or XP'd ANYTHING - PERIOD.

Moreover, the earlier versions of Mach' s that run on the type of DOS boxes he was planning got use are NOT SUPPORTED ANYMORE BY MACH or they DON'T RUN LATHES. Who do you call when an unsuported software pukes or won't run at all????

The last thing this guy needs, especially with the path he's expressed interest in traveling, is to get involved with legacy software that's NOT supported. We already know that the low dollar hardware he want's to use won't run XP.

I'm not trashing Mach. Many people use it and it satisfies their needs just beautifully.

However, I got the impression that some of the more crafty users of it know how to write code or are MUCH more computer savy than I with regard to integrating it with their particular machines.

When I contacted Mach about my intentions on a lathe retrofit, the system was ruled out as being viable when the words "you'll have to write some code to do that" were found in the reply. Sorry, I don't write code as my youth was apparenlty misspent in not learning to do so. Custom code writing is not something the Mach creator does and that's understandable.

I'd contend that my needs and those of the member from Australia coincide. We're techincal people who are not affraid of DIY projects. We probably can both load software and run it. We an also wire, rewire and troubleshoot wiring in machinery. Mechanical component development is something we can readily do/understand.

We want/need a CNC whatever to make stuff for us (in my case to support a business). Hence a hobby type system is not quite suited for my needs. I dunno if a hobby type system will fit our Australian friend's - he'll have to answer that.

Howver, my interest in and ability to write code to make a machine work is simply NOT going to happen. Where do guys like us to to get what is essentially a "component based" DIY system for CNC retrofitting???.

Mach, Ajax/Centroid, MDSI and/or Camsoft have already been investigaed and found to be incompatible with the equipment and or skill levels available to me. They are ouside the realm of affordability of my struggling little business or too much DIY which I simply can't do as a part of day to day business operation.

If "watch" my comments is meant to emphasize the insight I"m trying to provide. I've done what I chose to do - keep someone from making a horrible mistake.

If "watch" means to ignore what I say because I've misspoken, I'd contend I was misheard.

My particular needs are simply not met by what's offered by Mach or remain unsatisfied by the "write the code" issues with any number of softwares including MDSI and/or Camsoft.

The issues outlined by the respondants of the separate lathe retrofit thread with the other commercial packages mentioned speaks VOLUMES. The run around I personally got and others reportedly got from Ajax was beyond insulting.

At this point, it would seem that our friend from down under needs to revisit and/or rescale his priorities.

He's simply not going to get old DOS boxes to do what he's trying to do via a retrofit - the stuff to do it doesn't exist or isn't supported anymore.

Moreover, I'd contend that retrofitting an old manual lathe with CNC control, ball screws and all will cost MORE that what he could buy a fuctioning Expath lathe from here in the states.

BTW, the Expath WOULD run on the DOS boxes he's interested in using - that's what they used in them. Finding one of these gems is easy - the shipping to Australia would be a whole 'nother issue.

hi balsaman

there are still heaps of laptops with Xp very cheap and greater than 1 Ghz speeds

I plan to use large stepper motors or large servo motors havent yet decided

ball screws will be min 25mm shaft diameter and possible 30mm on the y axis

maybe.

Do you honestly think i will have to use some form of gearing ? using timing belts and toothed pullies and a reduction of 2:1 ratio gearing will required with such large torque stepper or servo motors ?

I just planned to direct couple the high torque stepper or servo motors to the ball screws.

But if i need the extra speed for finer cuts or speeding the cuts thats a different story.

Since this is my first CNC converions to a manual lathe , just a question

What ball screw pitch would be the best for the lathe X-axis and for the Y-axis ?? there are different pitchs in ball screws also coarse threads and fine threads so which is better for the lathe feeds.

I assume from my own experience that fine thread ball screws will give finer cutting per turn and the coarse thread ball screws will give more a cut per turn

All will depend on the speed the motors are turned and i don't know what adjustments or fine tuning features there are in the software , I plan to use mach3 for lathes.

What other lathe software is there avalaible at a reasonable prices for amateur cnc-apprentice-machinest ? ;)

cheers


About the only other software out there is Turbocnc www.dakeng.com, and I have tried (and bought) both. Mach is better but requires a faster PC. Tcnc runs in DOS. BTW I only have experience with mills.

Steppers are a low speed, high torque device, and can be connected direct to the ballscrew via coupling. Servos are lower torque, higher speed and almost always need to be geared down. Go with 2:1 or 3:1 for ball screws with a 5 TPI pitch. Higher ratios may be needed for lower pitch screws.

Do not go with steppers on this lathe. Servo's are way cheaper for this size a machine (ever priced the high torque stepper motors?), and more reliable because the drive can detect if there is a positioning problem and shut you down so you are not making bad parts and also possible prevent a catastophic crash.

If going Geckos search the online surplus places for DC permanent magnet motors in the 50-70 volt range that draw 7-10 amps and RPM in the 1000-2000 range (geckos are good for up to 80 volts). Modify these these for 250 CPR encoders from www.usdigital.com. Rutex makes higher voltage servo drives so the DC motor voltage can be higher if using theses drives.

E

Balsaman: Lathes are a TOTALLY different animal than mills.

For a mill program to even come close to doing a lathe, it would have to be capable of doing SIMULTANEOUS motion in Z and X axes - such motion is needed for thread and/or taper turning.

Essentially, only program that would accept an encoded signal from the spindle (ala a VMC or sophisticated lathe control) would even have a hope of working for our Australian friend.

In the "lathe retrofit" thread, Shizzlmah outlined the problems he had with an industrial grade, commercially sourced lathe retrofit kit. Seems the "lathe" version hadn't made a good/adequate transition from being a mill controller.

To summarize:

1. our Australian member preferred to use legacy PC's. Such devices preclude the use of XP needed to run Mach 4.

2. Mach 3 is no longer supported by the creator - they've moved on to XP. Support would have to come from the user group. Whether or not that would be adequate for FVP is something only he could answere - it is not adequate for my needs/operations.

3. Earlier versions of Mach that run on the preferred DOS legacy computers will run mills but will NOT run lathes. That is right from the folks at Mach.

4. Dind't see where TurboCNC (DAKENG.COM) would run a lathe but I didn't shop the whole site. See reasons why this isn't feasible at beginning of this post.

After literally WEEKS of searching, the only proven DOS based CNC system that I could EVER find that would run on legacy mills, lathes and 4 axis VMC's was sold by Bridgeport in the late 90's as part of their DX32 system. It was never offered in the aftermarket in any form.

Unless the member is willing to modify his intended component integration list, he can't do what he wants to do with what he has.

I too would contend that for a big machine like our Australian member has, a servo system would be the better way to go - it would have the speed and torque he will inevitably need.

Balsaman's comments in post #29 regarding the use of servos vs steppers on a big INDUSTRIAL SIZE lathe this size speaks volumes!!! Far cry from what you an get away with on a small/smaller, home size lathe.

All versions of Mach software have always only ever worked on Win XP or 2000. There has never been dos versions of Mach software. The original version was called Master 5 and ran on win 95/98 only, never DOS.

Both Mach 3 and Turbocnc can be used on lathes The author of Tcnc originally wrote the software to run his converted lathe. Both have built in single point threading capabilities by using a simple spindle 1 CPR RPM sensor. Both can run multiple axes simultaniously for tapers or whatever (up to 8 axes for Tcnc and 6 for Mach). Mach 3 runs my 3 axis mill all axis simultaniously for 3D contouring. Mach 3 lathe version is called Mach Turn. I have no experience with it but I know it has improved leaps and bounds as of late. I am sure it is not up to the mill level feature wise but I know Art (the author) keeps improving it. Mach 2, 3, and 4 (which is still a beta version) are all currently supported by the author. Master 5 and Mach 1 are no longer supported.

I am not saying the conversion is easy. It's not. But it can (and has) been done. The fellow who started the thread asked some questions about how to do it. I am trying to answer the questions to help him accomplish it in a realistic, cost effective way. Answering the questions by telling him it can't be done does not help. I would have never done anything cool in life had I listened to the nay sayers, and there are plenty of them.

It can be done. It's tough, it takes perseverance, patience, know how, mechanical abilities, imagination, blood, sweat, maybe tears, and some money. But it can be done. I could do it.

Eric

Pursaunt to FVP's post #1

"...But the problem is im (sic) new to electronics and CNC is not my strong point...."

Pursuant to FVP's post #23:

"...a PC or a laptop there are heaps of them around for next to nothing in price
old 486 will run most of the programs out there considering the old very early CNC lathes and mills had very old PC boards in there and functioned OK to a point until you overload the memory space on them LOL..."

Maybe old stuff down under but not Mach

With these givens, and the fact that retrofitting a lathe is clearly NOT A PLUG AND PLAY sor of projet, I'm only trying to point out some realities that the member is going to have to face/deal with

The type of PC's he's talking about simply won't run the XP version of Mach no matter how much as you say "... perseverance, patience, know how, mechanical abilities, imagination, blood, sweat, maybe tears, and some money" that you throw at it.

I dare say it would be nye on impossible to do for a CNC challenged indiviual and if the member is electronically challenged as he admits to being, the retrofit would be even MORE difficult to pull off successfutlly.

In point blank terms:

DON'T use a laptop - per ArturoD's experience.

Do use EXACTLY WHAT MACH recommends with respect to PC, processor, processor speed and memory requirements (forget the salvaged legacy PC idea).

DO USE XP because each and every verssion of MACH that is currently supported runs on that O/S

If you use MACH 3, use whatever Artspoft recomments. Per the Mach FAQ page:

http://www.machsupport.com/artsoft/faq/faq.htm

Looks like you have a choice between XP and (tah dah) XP.

I hope our Australian member finally understands by now that his seemingly enthusiastic plans to bolt together legacy stuff to do CNC machining on the cheap is NOT going to work the way he'd initially hoped to do it.

I am trying to help. I'd contend that telling him in no uncertain terms that it CAN'T be done the way he'd hoped to do it is good, straight forward information that he need to know and recognize before he buys more stuff that simply won't integrate.

If when the member finally realizes that perhaps Mach Turn is what he needs, the Mach experts can support his project as I'm sure they have a want to do.

IMO, I think he'd be better off getting an already converted CNC lathe with a dead controller. The mechanicals would be DONE and the only challenge (AND IT WOULD BE A CHALLENGE) would be the CNC side retrofit. That's enough for ANY newbie, again IMO. The complete retrofitting of a manual lathe to CNC is an extreme case of complecity in search of a need.

A 1 Ghz 'puter is a legacy machine.

Then 233mhx Pentiums must be pre-stone age.


IF that's the case, then what are 133's and, God forbid 386 or 486's???? (which still will do a lot of machine controlling and/or spreadsheet or wordprocessing tasks.

Pursaunt to FVP's post #1
IMO, I think he'd be better off getting an already converted CNC lathe with a dead controller. The mechanicals would be DONE and the only challenge (AND IT WOULD BE A CHALLENGE) would be the CNC side retrofit. That's enough for ANY newbie, again IMO. The complete retrofitting of a manual lathe to CNC is an extreme case of complecity in search of a need.

On this point you are indeed correct. This is the best way to do it.

E

Hi

It is ok guys , as i said before whether it be a bench top lathe or a bench top lathe that will break my back trying to lift it on the bench top what is the difference ?

there both lathes but different size ???

My lathe conversion i don't expect it to be as good as a commercially bought unit but it will definitely be able to perform cnc tasks

"quoting someone's words here I have existing CNC lathes. Simply replacing the controller with a PC based system has not been possible - what it would take to engineer and fit ball screws and drives to a manual lathe aren't easy even if one has an engineering background. "

Whether you mount a ball screw on a small bench lathe or a large lathe like i intend to do , what is the complications there ? I understand its not going to be a straight forward fit .

Will involve some thinking and tricky machine work which is a challenge but like anything else in life i will think hard and smart how to solve the task.

I have machined i think more complicated things in my time. and had great challengers in life.

I said electronics is not my strong point but i'm sure i can learn as i go along.

I know everyone has to start some where and we all have to learn from someone but one thing i have found people that hold a education tittles seem to think there better than everyone else.

A title like electronics engineer and many spin of engineering titles in Australia , most of these guys look down on you and think your no one , they have the attitude of I'm better than you and talk in there own language.

Guys that may be offended here , but that is the truth

I have more than just a 386 laptop LOL , and remember guys PC's are not expensive even new late model units .

Let me have a look on the mach3 site and see what the minimum software and hardware requirements to run mach3.

Like i said im open to any guidance and feedback from anyone , whether i use that advise is another thing.

Yes i am new to CNC and i'm sure i can learn it just like everyone else has.

What is this forum all about and many other forums a processes of sharing information and discussing ideas. A means of communication and a lot of advertising from business.

cheers

hi

Just went to artsofts sites and the minimum requirements are as follows

" All of Artsoft Software has been developed around the ArtSoft pulsing driver for Windows. This driver was created for reliable pulsing under Windows environments to eliminate the problems which previously plagued Windows CNC software. It can reliably pulse to 45000 pulses per second in XP and Win2000 configurations. Its list of features grows every month from user requests. Join the support group to add your voice to the growing list of Mach2 and Mach3 users. "


there are a lot of laptops and PC's in XP and Win2000 around that are a year old and very cheap , www.ebay.com.au is fantastic site for buying computers.

even new laptops are cheap with the minimum requirements to run mach3


cheers

Regarding the use of a laptop:

FOR GOODNES SAKE, PLEASE REREAD POST #27 ABOVE.

If/when a laptop causes problems, you won't be able to say you weren't warned by folks who tried to help...

FPV,

One last advice; DO NOT USE A LAPTOP FOR RUNNING MACH. I know they look very sexy on the machine, but there are many issues about the hardware, and how battery saving issues messing around with the processor’s speed. Please watch NC CAM’s comments. I am a very happy user of mach and winxp, and I do recommend it, but please do not take it for granted. Follow Art's recommendations on how to set it up, and please, use a dedicated machine. If you find that machine is not reliable with winxp and mach, please do not try to fix it, change the PC. I understand you must balance your budget with the final product you need or want. I think none of us can tell you if you need to buy a compact or a pickup truck. The only thing that you can not play around is safety. Do not attempt to play around pneumatic chucks, tool turrets, and bar feeders unless you know what you are doing. I know the way I say it make it look like wiring all the gadgets is pretty simple, and it is, but if make an error estimating how a relay, solenoid, or valve will behave on a certain circumstance, then you run into safety issues.

Arturo Duncan
http://cnc4pc.com

hi

Ok ,read and understood , Brian cheque is leaving monday to you for mach3

OK laptop idea discarded nothing a nice tool box on wheels and draws painted in a nice machine colour scheme with a big 19 or 20 inch monitor will not fix , get the BIG SEXY LOOK ;)

cheers

Who is Brian? I would wait with the purchase and concentrate on the conversion first. The demo can be used to get the machine up and running. The demo is fully functional for up to a 1000 line Gcode file.

E

hi

Brian is from artsoft mach3

mach3 is very reasonably priced , its not a big investment at all but seems the go by a lot of the forum uses in here and mach3's forum pages.

"The demo is fully functional for up to a 1000 line Gcode file " understand where you are coming from but this gives me a fully licenced copy and it's mine. I can play around with it , i know i can do it with the demo copy also but why waste time buying it when it seems the software to go with , read there website and features there software has as opposed to more expensive ones arond.

No one has suggested any other reasonably prices softwares that i can afford at this stage , I see even nickname " stevie " in this forum is using it on his gantry milling machine and i imagine even on his lathe.

So this will not break the camels back LOL

Had a look at some nifty tool changes just pondering with the idea so guys don't shoot me yet LOL , that are commercial tool changers here are some pictures , i'm sure some of you guys will recognize the brands.


several questions guys

q1. what is the best ball screw thread pitch for the X-axis ?

q2. what is the best ball screw thread pitch for the Y-axis ?

in thread diameters from 25mm and say 30mm ball screws ??


Another reason i want to use very large stepper motors is i don't need to gear the drives , will be plenty grunt in the stepper motors

Large servo motors are very expensive anyone know of a good cheap source for servo motors ? Found a few sites and still waiting on email replies back.


cheers

You are going to need a BIG any kind of motor to run a lathe of that size. Big steppers are NOT that much if they in fact are cheaper than servos when you get that big.

The servos on my Fanuc 5T's are the smallest ones Fanuc makes - Model 0. Yet, they are the size and weight of a Chevy V-8 starter motor - and they're geared 2:1 to boot.

The comparably sized stepper that I have on a mechanized infeed for my cam grinder is comparably sized (to drive the same weight or nearly the same) and the price is as much or MORE than the Fanuc servo motor. In fact the stepper is more because BIG steppers are rare and thus more pricey.

If you want any resolution, you simply CAN NOT go direct drive. Figure the resolution potential for a 10 rev per inch ball screw, driven at 1:1 with a 2000ppr encoder versus the same screw at 2:1 with same encoder.

Twice the resolution with 2:1 and twice the torque needed to drive the 1:1. And ultimately torque is what moves the carriage and crossfeed. Also, if you direct drive, your current requirements will be HUGE. When you keep current in the 12-15 amp range, you can get away with and can afford Geckos or Rutex.

Start pulling 25-30 amps, you're now talking AMC's (I don't think the make step/direction drives) and these are easily $560 USD each and they don't turn up on EBAY that often.

The ball screws on my Fanuc 5T's are 35 mm in the ball path area OD for BOTH axis. Your choice ultimately but bigger is better and 30's would be the smallest I'd consider.

In the mean time, you might want to reconsider some of the SWAG"s that you've got on your plans list. Do some looking at what comparably sized machines may be using.

If the results of your investigation don't bear out your plans or theories - believe what's working and invent new plans and/or theories.

hi

"saturday here in Australia and the big tattslotto draw is on first prize is $20,000,000 will buy one ticket for that draw " back to work again didn't win the big draw

what is a good CAD/CAM program ? remembering budget in mind something that's affordable and i don't have to sell my house

Ok understand when you mention increasing the working power of the stepper motor and or servo motor when using a 2:1 ratio pulley setup.

what have other people done to ratio setups , as in 3:1 , 4:1 and so on ??

the lathe that im converting is not a small benchtop , so putting some sort of pulley and timing belt arrangement will not be a issue , just concerned with the load the timing belt will have any issues driving the cutting tool and slide carriage the ball screw back to the stepper/servo motor. As this is not a small lathe cutting deep might be scarey.

Oil , grease , water , grit and sworf all will affect the rubber constructed timing belt will have some major job just sealing the belts from the harsh environment it has to work in . Since most timing belts are around 25mm wide might have to double up the width 50mm to halve the work load of one belt on its own. Just a thought.

I will most likely will use overhead cam engine camshaft and crankshaft toothed sprockets and timing belts are available in different lengths and very readably available . The timing belts are strong driving the engines camshafts and accessories.

This will be experimental time in the conversion


servo motors are not cheap from what i have seen on the websites. What's a cheap supplier ?



cheers

" don't you hate it when someone buys something for $2-00 and sells it for $2000-00 "

Point of order - rule of law in CNC:

1. Big stuff (motors, drives, etc) are ALL costly

2. Lower volumes (big motors are industrial/limited use) mean higher prices and less likelihood of Ebay cheapness - other people with "real" CNC's need service parts and they have to pay dearly for them.

3. Too wierd of a component hodge podge can put you in a position where finding help can be difficult if not impossible.

4. Matched automotive pulleys for OHC are ALL going to be 2:1 ratios in sets - think about it for a while and it will make sense.

5. Yes, various belt lengths but ALL are very long. My CNC lathe belts are about 1.25" wide and the center to center distances is about 8" or so - real tight package AND well shielded from coolant splashes. Automotive belts are good in quality BUT too long in length - they will NOT survive direct coolant exposure. Wanna do it cheap? think go-cart roller chain drives - bullet proof in compared to ANY belt.

2:1 is the defacto standard reduction factor. Odd ( 3's, 5's etc) as opposed to even ratios (2 or 4) are simply NOT done - do the math and you'll see that odd numbers don't divide well into a number system that is traditionally based on a system that is evenly divisible by 2. Besides, 2000ppr encoders don't take kindly to division by 3 either...

EDIT Besides if you run 4:1 the encoder counts per rev can get so high that the computer can't count them fast enough. This situation gets worse when you try to feed the counts thru the parallel port. NUMERIALLY, anything should be possible, TECHNICALLY it ain't always possible to do that much in what becomes too little time.
END EDIT

Question: have you even looked at/considered what's being used and why or is this a TOTAL start from scratch with a preconcieved SWAG devlopment program???

No problem with creative thinking but it ultimately has to integrate with "traditional CNC engineering and design" concepts. It also needs to be technically viable.

Too much out of the ordinary thinking makes getting help for a "too far out there" new scheme difficult. The "boldy go where no man has gone before" Star Trek approach can ultimately be the undoing of a project.

hi

all automotive pulleys are a two to one ratio.

my question was not do engine camshaft and crankshaft pulleys come in different ratios? But has anyone used a 3:1 or 4:1 ratio in pulley arrangement to drive a ball screw via a stepper/servo motor to make the mechanical effort to drive the system easier and higher resolution in steps?

Timing belts come in all types of lengths from 200mm up to 1500mm in linear length , as far as a straight on fit I'm very sure it will be a lucky , but some sort of belt tension arrangement will be required firstly for adjustment and secondly belt stretch maybe a automatic spring loaded tension to keep things nice and firm.

Yes probably a chain and sprocket arrangement would be a better choice. And again you local V8's ford or chevy engines have a nice double row chain and sprocket arrangement " note all engines run a 2:1 ratio " , the chain is very short probably talking about 8 to 9 inches between centers , will measure one at work tomorrow.

I'm very surprised that a PC can be restricted to serial data counting ?

PC's are very fast in number crunching I'm sure there is a limit but we wouldn't be any where near it if use a 4:1 ratio increment increase ???one example is a optical ignition sensor in a nissan VG30ETT V6 engine has a 360 slot wheel and at 7000rpm that's a lot of square waves pulse counting by a 1995 ECU in a vehicle as compared to a PC , and that's not the only signals going into the cars ECU and not to mention data to activate things coming out of the ECU .

Everyone has different ideas other wise all commercial CNC machines would be the same ????

cheers

Hi

I failed to mention i seen someone in this forum using 1.8:1 raito connection between the ball screw and stepper motors

I haven't had a look at what is involved with the software to run the cnc setup

I imagine there is fields one can tell the software in the thread pitch and gear ratios and steppers it has to count to control the X and Y axis

lets see how i go here , will be fun

cheers

"AMETEK 38VDC 900 RPM MOTOR" about 3/4 down the page in the center would be good candidates for your lathe.

http://www.73.com/a/0082.shtml

Add some 250 CPR encoders and you are good to go. These are very similar to the motors on my mill. Lots of torque. Go 2:1 on 5 TPI screws.

40-50 volt DC powersupply coupled to Gecko 320 or rutex drives.

E

hi

Balsaman having a look at that website now , thank you for that info balsaman .

NC Cams do you have a thread or pictures to show on you CNC_LAthe that i can have a look at ?

Ok guys anyone else have a thread on a lathe or mill setup , just post a link in here or even your own website , would love to have a look .

If for some reason you don't want the whole world to look just email me a link or picture or PM via www.cnczone.com

Pictures are worth a 1000 words so guys picture of you latest CNC toy with a brief specification description on the machine.

Another Question for the senior members in here where does one make a donation ? to this site since im using the cyber space time i thought i better contribute some funds for ONE OF THE GREATEST CNC FORUMS i have ever come across.

So much information in here i will be sifting for hours does someone have a direct link to the conation section ?????

cheers

hi

Yes you are right , a very nice motors

motor1 -AMETEK 38VDC 900 RPM MOTOR - 900 RPM AMETEK, #965922-101. Permanent magnet DC motor rated 38 VDC nominal. Reversible, ball bearing. Max. rated rpm is 1300 max. Armature current before permanent demagnetization is 12 amps. Torque constant is 50 oz-in per amp. Terminal resistance 20 ohms +-20%. Voltage constant is 37 VDC per 1000 rpm. When tested on 38 VDC, we recorded the following data: No-load spedd 1100 rpm @ 0.260 amp. With a 100 oz-in load the shaft speed was 900 rpm @ 2.3 amps. When tested as a generator at 1000 rpm, the output was 5.8 amps @ 12 VDC. Shaft: 5/8" dia. X 1-5/16" long. The opposite end of the shaft has a turned down diameter and a flat mounting area with two sets of tapped holes. This was probably originally intended for an encoder or tachometer mounting. This shaft has an outside dia. of 3/16" and is approx. 3/8" long. We do not have the encoders whtat were mounted on this motor, but suitable encoders are probably available from encoder suppliers if required. The motor has five #10x32 tpi tapped mounting holes on the shaft end of the motor body. Electrical connections are via 3-1/2" long flexible wires. Dimensions: 4" dia. X 4-1/2" long (excluding shafts). $74.50 Ea. MD548

motor2 - AMETEK 38VDC 900 RPM MOTOR WITH 5/8" DIA. OUTPUT SHAFT - 900 RPM MOTOR AMETEK, #965922-102. Same specifications as our MD548 except the main output shaft is 5/8" dia. And 1-7/8" long. The encoder shaft at the other end of the motor is identical to the one described in the write-up of MD548. Approximately 1500 units available. Weight 7 pounds. $74.50 Ea. MD549

now is just a case of finding some 250 CPR encoders ? any ideas where guys ? will do a yahoo search lets see what i find.

this what i found amongst a lot of other encoders , here is a link

http://s120220635.onlinehome.us/servo-motors.asp

http://www.usdigital.com/products/s1s2/ found a little bit of info EDIVIDE
Integer (Whole Number) Resolutions Available http://www.usdigital.com/products/edivide/integer-resolutions.shtml

converting shaft rotation into electronic pulses, encoders are used to electronically this is a PDF file

www.ab.com/sensors/sensorstoday/pdfs/july03/encoder_sg.pdf#search='250%20CPR%20encoders

and more encoders http://news.thomasnet.com/fullstory/486545



cheers

http://www.usdigital.com/products/e2/ These are the ones you want. 250 CPR. Get the 6' cables with integrated connectors with them at the same time.

You will need to dissasemble the motor to add a stub staft to mount the encoder disk.

E

http://www.cnczone.com/forums/view.php?pg=donations

Re: lathe photos = I don't own a digital camera.

At one time, I did a "Cadillac NC100 lathe" search on google (my exact machine) and a foto of one with something other than a Fanuc control came up. They are not that popular now as they were a 1979 era machine - I got mine out of "mothballed" close out sale from a shop that had 3 of them.

You might try googling a "Bridgeport Romi 2 or 2S" lathe. These are decent size full CNC lathes made in late 90's with a proprietary Bridgeport DX32 control system - much more common and easy to find.

It will give you a representative picture of a CNC lathe albeit with electric "feed handles" to electrically drive the servos ala what you'd do with a mechanical lathe.

Re: what a NC/CNC lathe looks like: picture a mechanical lathe with a ball screw in place of the lead screw to drive Z.

Instead of a quick change gear box hooked to the lead screw, a servo is mounted in its place. Endoders on opposite end of belt drive. There are NO shift handles and NO manual handles at all on my lathe - it is all servo driven. The "lead screw/ball screw nut is fixed mounted to the carriage for both Z and X..

The cross slided has NO handles nor clutch either. No mechanical infeed at all. Servo is out the back side of the cross slide and is 2:1 connected to the X axis.

The device is "connected" between the spindle, X and Y axis via "wire" ala the servo mounted encoders and a spindle mounted encoder with a 1 pulse per reve "timer pulse" for threading.

If you mess up the encoder count with respect to screw pitch with respect to gearing, you'll perpetually make/cut junk.

Lathes of the size and type you're working with are NOT regularly retrofited to true NC. This is why I suggest you get an NC lathe with a dead control. if you get one that hasn't been picked over, your mechanical side issues may be done for you already.

Some math to consider:

5 revs/inch on screw. 2:1 gearing 250 ppr of motor mounted encoder. For each tick the motor emits, the axis will move 0.000400".

10 revs/inch screw, 2:1 gearing, 2000 ppr of motor mounted encoder. Each tick now only generates only 0.000025" theoretical travel of the axis.

Have you REALLY set down to figure out what you need or are you just buying stuff on price and availability on E-bay??? Is it still a great price if it ultimately doesn't work properly????

Sooner or later, all this stuff needs to be integrated. If you're not careful, you could end up with great deals on stuff that you can't use or ultimately won't work or will be dead slow or gutless.

From a previous poste, you expressed a comment about number crunching in a computer. Here's why PC's sometimes SUCK for CNC as composed to a true, dedicated control thread:

Why parallel ports may sometimes suck at doing CNC control (See post #5 in the following thread):

http://www.cnczone.com/forums/showthread.php?p=181852#post181852

Develop a wrong perception of what a computer does and/or how it does it and you could become extremely disappointed.

Actually 250 CPR is 1000 pulses per rev so on a 5 TPI lead screw @ 2:1 ratio you get a resolution of .0001 or 1 thenth of a thou. I guarantee you this is much better resolution that the mechanical parts of the mill will get you, so no need for higher CPR encoders. You don't want 1000 CPR encoders as it takes a very fast PC to make the machine go fast. The Servo/ratio/leadscrew combo I recommend above will get you roughly 100 IPM rapids which is not lightning fast but good for a hobby type lathe. If you want faster rapids you need different servos. I don't recommend going faster than that. Makes crashes even scarier. My mill will do 100 IPM but I set it up for 50 max as it's safer.

E

One other thought if you like those motors--read this post by Mariss Friemanis and the associated report he has written:

http://finance.groups.yahoo.com/group/geckodrive/message/9088

That's very similar performance to the motor you are looking at claims, verified by someone who knows and is an independent 3rd party, it is brand new, and there is a known formula for what controller will work with it. You'll still want the same encoder Balsaman has mentioned.

It's worth consideration.

Best,

BW

PS Here is a similar sized lathe, BTW, that is basically a manual lathe professionally converted for CNC:

http://www.microkinetics.com/lathe1236/index.htm

It can be done! The specs on that lathe may be of interest in terms of "what worked" in another application.

curtisturner - Chip Sweeper - This might be a dumb question
Link - http://www.cnczone.com/forums/showthread.php?p=181852#post181852
This might be a dumb question but here it goes.

What is the real difference between the following 2 (other than price)
PC and a couple of Gecko Drives some Servo's and Mach 3

And

A NC Controller like Fanuc or Allen Bradley

Thanks
Curtis
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

post # 4 NC Cams Master Machinist -
Why parallel ports may sometimes suck at doing CNC control (See post #5 in the following thread):


Mach is a step/direction system and it doesn't close the loop between what the steppers ultiametely do and what they're told to do.

Sort of like driving a car down the street via radio control - do you feel the acceleration in the seat of your pants if you pull the throttel trigger harde??? The radion (computer) tells the motor to speed up and slow down but can't never feel if it did it or not. That's the easiest way to look at it. Even if you use a servo card with Mach, you don't get closed loop feedback to the PC.

Yes the PC sent out a directive via the LPT port but until/unless the error gets too bad, it continues sending out pulses fat dumb and happy until the servo card sees a sufficient error. ONLY then, will the card notify the PC, again that will be recognzed when the next LPT port call occurs. Although the encoder feeds back to the servo card,the card doesn't feed an error back to the PC until/unless it gets too far behind.

In my servo system, the moment the drive faults (over current, no feedback signal, lag singal, whatever) IT STOPS. It doesn't do so many steps, gets so far behind then says, "we have problem, better stop because the problem finally got TOO bad".

Look ahead functions and interpolation (which any computer time so as to look ahead and calculations to happen for it to work), will bring thruput to a crawl. This is especially true depending on how your processor does background calculations and the priority it gives to them.

Instead of interpolation, try doing point to point milling in very, VERY small steps. WE do it on cam profiles and you can't tell that we've got teeny tiny flats on what should be a full radiused surface.If you make 2880 flat cuts to create a circle,you can't tell it after you buff it with 320 then 600 then scotchbrite.

At 360 cuts, the thing looks like a diamond disco ball with all the facets. Sounds like a machine gun when a roller runs along the surface.

We have the ability to do constant surface speed tool path cutting via S speed changes for every single cut point. The RAMDRIVE program in DOS does the look ahead, the controller sends out the code, the servo amp responds and if it does so properly, the cutter keeps mozying right allong. We run real slow (inches per minute and real high spindle speeds cause we're looking for mirror like finish).

Why an it move so slow and smooth at CSS?

Because a 12 bit ANALOG voltage is being sent to the servo amp (in increments of 0.004vdc, +/-10vdc full scale). Between the possibiltity of the stepper hittiing a resonance frequency and or the step function not being sufficient to provide adequate control at small incremental changes, it is possible for the system to hang up or default to a safe mode or the motor starts to "shiver" and that isn't what you want/need.

In your case, you're still dealing with a step funtion. You may have reached the limitation of your stepper motors and/or the limit for a particular step/direction system. Steppers do amazing things. But, they can't be turned into servos no matter how hard you may wish or

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

post #5 - fkaCarel Apprentice

For starters: it's not a dumb question, but you must have the opportunity to look on both sides of the fence. Fanuc etc. are dedicated and embedded controllers. This means that you cannot do much more with it than controlling coordinates on a machine. A PC is an office machine which gives you, in the end, on screen, the sum of 1 + 1.

The advantage of a PC, fast execution of a part of a program or a whole program, disappears if it has to execute the result of a small program. This means: sending pulses to the parallel port in timeslices is crippling. The design is made to send all the data in a stream to the port. A printed page!

A real time processor derives the time from a timer. When the timer is zero or overflows, depending on the type of processor, a new task can be started. For example the generation of one step. When the task is done, the program waits for the next timer interrupt.

I cannot look in the Mach3 software, but it looks like it's pulse delays are generated by a loop. Load a number, decrement by one and exit if the number is zero.

Where the embedded controller can do initialising of new instructions, calculate interpolation and maintain constant velocity in its free time, for the PC everything is overhead. The to decrement number should be, but cannot be adjusted for the specific software action of the moment.

So the easy solution is: buy the fastest PC you can get. The number in the delay loop grows, the overhead is executed faster. This is noise reduction by numbers.

The costly solution is: buy a Fanuc and have a specialised controller.

The crazy solution is: do it yourself and invent new G-codes and new approaches
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@


thanks for that NC Cams Master Machinist

I haven't gone on a wild rampage yet buying parts other the two lathes which were cheap and the 4 ball screw which are a steal at the price i paid.
Just picking peoples brains and asking questions and there experiences when they started to contemplate doing a CNC_lathe conversion

I know where you are coming from and i thank you for your guidance and helpful information.

I have at work PC's that are on automotive test equipment and PC's in the office and have several early to the latest laptops already in my possession . I was just thinking of using a dedicated PC and or laptop just for the lathe itself with no added programs working in the back ground.

That is why i said there are cheap laptops and PC's floating around with wind2000 and winXP , this way the machine will have its own PC not to share with anything else in the workshop

The actual components steppers motors or servo motors , encoder , stepper motor control , servo controller , power supply(s) , software ( mach3 , desktopcnc etc ) and so on, I haven't yet purchased.

I'm hoping i get some guidance here from you experienced people who have already done a conversion lend a hand with your knowledge. The likes of you NC Cams and balsam and BobWarfield and others that have contributed a lot of information to me. Also some of the postings in other threads i have searched in here , there is just a vast amount of information in here it is like a library in here.

The biggest problem in here is find the information, there is a lot to read and that's why i posted this thread , just to bypass some of the information that's out there as it will take me another year to read , i want to commence on the actual lathe conversion.

I followed some threads and looks at how these people tackled the lathe conversions and mill conversions and what i have noticed they started from knowledge factor zero of CNC machines from the hardware to the software component of CNC machining. I have noticed there either using it as a hobby thing , or there in the trade of fitting and turning ( machinists ) which is a trade in Australia , different countries call the trade a different name.

NC Cams your suggest buy a old CNC_lathe is a great idea but most of the bigger lathes are snatched up well before i hear about them being up on the market place . At auctions it just attracts a lot of potential buyers and the machines end up escalating in price when the WAR ON BIDDING COMMENCES . Great for the seller but in the end the buy has paid way to much for the machine they have just out bided 40 other guys.

I have followed several machine when they have come up on auction and i set a reserved price that i would pay and always the prices exceeds my expectations . That is why i have chosen this path a old very ridged copy_lathe to convert , you could call them a CNC-lathe of the very early 1970's era.

So guys any information is appreciated in this thread . Even if there are threads buried away in the archives of this forum and you know the link just cut and paste a direct link to it in this thread please. I have use the search facility in here and tried all sorts of key words and read a lot of threads in the past few months . Like I said this forum is huge and a wealth of information in here.

Possibly one of the worlds best CNC forums , this is my opinion though.

I still am interested in making a donation to the forum for there services , who is the moderator to speak to ? so i can make the contribution.
and guys thanks for your patients with me .

cheers

For donations to the zone see post #51 above.

hi

thank you balasan , i missed that post my error and thanks

oops :violin: :mad: :violin: :mad: (nuts) :violin: (nuts) :violin:

cheers

OK

donation details have been requested , i would prefer to use a bank check ( cheque ) method

so if anyone reading this thread would like to make a donation to this forum the link is as balasan posted in here

donation link http://www.cnczone.com/forums/view.php?pg=donations

cheers

If stuff is so expensive down under, what about importing something in from the USA???

There are tons of machines looking for homes that can be purchased quite readily. I dunno about shipping but maybe all in all, it might be a better go.

As in get something DONE, iron wise, that only needs a controller replant. That was my intent with my lathe retrofit and learned that lathes retros are NOT easy. Knowing the "challenges" alone offered by the PC/controller side, I wouldn't even THINK of retrofiting a manual 1970's gem. BTW, I have a fully functional Fanuc 5T under power you could buy quite reasonably.

My posts may seem hard core and tough but the task at hand is just that. When I was a training engineer, I was that way - hard a$$ed to the students. One day, months later, a former student came back for a sales conference and thanked me for being such a hard a$$. I asked why as I was beyond horrible to them.

His reply, "... yeah you were, but the customers are worse....". And he thanked me for it.... Go figure.

You've chosen a particularly hard task - and one that is not in a stage of mature develoment (I'd contend). Just, don't get overwhelmed. We'll try to help as best we can.

hi
LOL NC Cams na you ok , i have to admit in my time i had some hardass teachers but they were all good teachers , some of the students can get out of hand

And your ex pupil is right customers are harder to deal with LOL

Just of the topic from this thread , what is a better controller the maz_control or fanuc_controller ??

I'm under the impression the maz_controller is made by Mitsubishi and is a bastard of a controller when it comes to software ? It is a all in house type of system not PC friendly with other software ???

Does anyone have a site that has information on both controllers ??

what are all the fanuc-controllers that come out from day one right upto now ? anyone have a listing ? I see fanuc is also in china ?

Same as the mazcotroller , what are all the mazcontroller that come out from day one right upto now ?? anyone have a list of them ??

cheers

hi

Just surfing the net again ;) come across this software called SURFCAM and one use it and what do they think of it ?

www.surfware.com

cheers

Here is a must read thread that is quite pertinent to your efforts:

Great NC controller thread - NC vs PC’s

http://www.cnczone.com/forums/showthread.php?p=182852#post182852

Speaks volumes about the controller options.

A lot of the dedicated controls (IE: Fanuc, Mazak/Mazatrol, Mitsubishi, etc) are NOT geared to the one-off retrofitters. Fanuc service sucks and what service they do offer is expensive - can't speak for the others. Al the Man has spoken well of Mitsubishi I think but the controller alone is $10K USD or such.

You can buy a turnkey Ezpath PC based CNC lathe on Ebay for numbers like that. Folks who have them speak well of them.

The previously posted thread says a lot about steppers. Seems that the technology is NOT that user friendly for to the DIY'er with regard to servos- HOWEVER that depends on the technology level you're working with.

The once common, still available analog controls (at this point in my level of understanding) are closed loop capable but how "closed loop" is something that I"m learning is NOT quite perfect. Digital drives like those used by Ajax supposedly have neat 'self tune' features but the other aspects of the Ajax have been discussed ad nauseum on the other lathe thread. Halfnutz experiences with Camsoft disuaded me from even considering them.

Servos too have an "error level". This is why a servo has position and velocity feedback. THe PC tells it to move so far and fast and the servo does so albeit a bit slower than the instantaneous instructions - fast but not instantaneously identical. It seems that the faster you ask it to move, the more error it can develop.

Error to a a servo is analagous to lost steps in a stepper - so much is manageable/tolearable. This is why after a multithousand step tool path, you need to check if the servo will find/keep the same 0,0 (not home) origin.

The way to check this is to mill a circle or cut a taper. If the taper isn't skewed and the circle is round, things are going pretty well. IF the taper is skewed and/or the circle is out of round or doesn't meet up without a step, you have problems.

Whereas a Gecko or like stepper driver will stop if the error steps get say 128 behind, the servo amps, too have the same sort of error logic that they respond to. In my Bridgeport system, the speed/direction signal is in the form of a +/-10 vdc full scale analog signal. The motors send back 2 signals - position to the PC and tach to the servo amp.

While monitoring current and voltage, the amp checks to see it the motor is running in proper direction and at proper speed via the tach signal. If not, shutdown. Similar to what steppers do but without the "issues" of having to go thrue parallel port which is also discussed in the previous thread.

THe following thread helps outline the goods and bads of steppers and servos:

http://www.cnczone.com/forums/showthread.php?t=17419

At this point, I'm not convinced that a CNC lathe will cut threads as reliably as a mechanical lathe. You're still "flying by wire" and an electronic gear is NOT as 100% positive as a mechanical one. Yet, they do so very well, usually.

I'll assure you this much, the more you look, the more disillusioned you may become. You'll learn that this CNC lathe thing is no where's near the level of development of a mill or VMC. Ultimately you'll have to decide as to which compromises you're willing to live with. IF you can, go see a system - surely someone in your neck of the woods has done a lathe retrofit. THey can tell you MORE than you'll imagine. Forget what the salesman says, talk to a user.

When I bought my bench lathe from Grizzly, I asked if they had customers in the area. They did and gave me some names. Called one guy and he said "come on over". Spent the evening with him and he went thru the whole deal with me. I bought the same lathe based on the experience both with the seller and this buyer. A person who has a system is the system's best critic and/or spokesman.

RE: my NC lathes:: Mach is out because of the fact I'd have to go to XP and I don't have the funds to buy that much stuff to support this O/S. I too was hoping to use legacy boxes.

The problem there is there are no legacy systems that have a control offering anymore. DOS/PLC's that worked on legacy boxes are gone. Moreover, analog servo drives that were used with these systems are pricely as well. AMC's are about the only ones that will drive my servos and they are nearly $600 each - far cry from the Rutex and even these are not cheap ($325 each if I recall).

Keep researching.

You need to feel comfortable with your decisions. I can probably provide some servo and drive P/N's that I know will work as a matched assy (they are from BOM of a production late 90's cnc lathe).

CAUTION: these will involve either 3 phase AC and/or DC drives. I still haven't learned enough about the AC's. Since I grew up with DC trains and slot cars, I have a natural inclination to go DC servos - God are they expensive. For that matter, so are the comparable sized steppers on some equipment I have. AC's in that size frame ain't cheap either.

Finally FORGET 100" per minute jogs. They are flat dangerous and downright scarey - besides the're useless. 30"-50"per min is plenty and besides, you don't need a power supply that you can arc weld with to drive them. True cutting speeds and spindle power under load are much more critical, especially if you're looking for a good surface finish....

There is no "better control". They all have features and detractions. Some meet your needs and others don't. Guys who were taught on Fagors swear by them. Guys who learned on Fanuc's, ditto. My neighbor has only Haas and he's a 1 man shop with 2 lathes and 3 VMC's. Says a lot for Haas. The service guy was there once in the 4 years I've been in the complex.

Hope the info helps

Re: surfcam lathe: didn't see anything about threading or taper turning.

Looks like all they did was treat a lathe as a VMC with a 4th axis. A lathe is NOT that.

Error to a a servo is analagous to lost steps in a stepper - so much is manageable/tolearable. This is why after a multithousand step tool path, you need to check if the servo will find/keep the same 0,0 (not home) origin.

This is incorrect. A servo will have a small error while running but will always catch up when the axis get to it's final destination and stops turning. A stepper that losses steps will never catch up again.

The only was to "loose steps" on a servo system is a bad encoder, or noise in the encoder cables or in the parallel cable. This can be prevented by doing proper shielding. What can also happen is the load on the servo is so high that the "following error" gets beyond a fixed amount and the servo drive faults.

E

Let's assume that for a mill the servo position errors form a ball around the command position. For every command position, at rest, the mill is somewhere in the ball.

This is incorrect. A servo will have a small error while running but will always catch up when the axis get to it's final destination and stops turning. A stepper that losses steps will never catch up again.

The only was to "loose steps" on a servo system is a bad encoder, or noise in the encoder cables or in the parallel cable. This can be prevented by doing proper shielding. What can also happen is the load on the servo is so high that the "following error" gets beyond a fixed amount and the servo drive faults.

E

This is absolutely accurate, insofar as it goes, but note that servo following error can lead to innaccuracies in the finished part even though the servo caught up without faulting. The curve that gets cut in the part will be slightly different than what was intended. In this sense, the other poster is also correct to say that following error can be just like lost steps. If the cutter didn't follow the exact path commanded, but it caught up later, you still have an error. There are many examples of this even on high end machines.

For example, consider the "waves" described by Geof in this thread:

http://www.cnczone.com/forums/showthread.php?t=22589

Somewhere I came across a picture on a GE-Fanuc page of a part that had a curve on it that was off by almost 0.1". There was a ruler for scale and a line drawn in to show what it should have been. I think it was part of the sales pitch for the lookahead and interpolation options. I wish I had the link, but their point was that the machine may have to look outside even of the g-code being processed and plan for what's coming next or the sheer inertia and following error that results can lead to some pretty dramatic inaccuracy while the CNC machine still thinks it did the right thing, closed loop and all.

That's on a real nice has VMC with servos. CNC involves MANY approximations to reality.

Best,

BW

Let's assume that for a mill the servo position errors form a ball around the command position. For every command position, at rest, the mill is somewhere in the ball.

That is pretty much how it works. If outside the ball is a servo fault, then hopefully you are staying within the ball while running a part. The heavier the load (or bigger the following error for whatever reason) the closer you are to being towards the outside the ball. As the system comes to a rest tho, you should return to very close to the center, since the closer you are to the edge of the ball, the harder (more current) the servo system tries to get back to the center.

The ball for a gecko servo drive setup done properly will have a radius of 10 thou or so. On a mill I imagine 99% of the time you are within 1 thou of the commanded position unless you are making a rapid move (where it doesn't matter). Cutting forces are generally very light on a ball screw fitted milling machine. I imagine cutting forces on a lathe x axis can be quite high. Possibly a higer ratio from servo to ball screw is required here. Since the travel is quite short, the slower rapids won't matter.

E

I wish this thread was in play last summer.

That's when we were going nuts trying to get our Extrak to cut a perfect circle. We ultimately learned that by running R E A L slow feeds, and doing point to point milling at 1/8 deg intervals and a BIG diameter finish cutter, we got good roundness and minimal error. A 0.010 "window" is way too primitive for what we do.

Any way, the whole time we figured that the computer would recall where the 0,0 origin was for the center of the hole we were miling. Close but no cigar. When we got the code working on the 'Trak, we took it to a Haas for speed and potential accuracy reasons. Essentially we were milling an OD profile that had to be concentric to the ID mounting hole.

The Trak could get within a thou or two which was GOOD but not good enough - we were looking for half thou. This is why we tricked out the ballscrews to the level we did. Got circles round to within 0.0003" - as milled!!!.

After porting the code to Haas (Identical to DX32 save for a couple of insignifigant comment lines and tool change codes), we ran it (with real slow feeds which we learned were critical to accuracy on the BPT). The Haas could do half thou to 8 tenths TIR which we could just barely live with.

A CNC ground part is within a few tenths but triple the price and two/three weeks delivery as opposed to 48 hours or so.

If we get a viable lathe retrofit, we're going to try to rough turn a lobe shape into a cam blank. Don't have to be perfect and don't have to be smooth, just a reasonable lobe approximation. We'll tune it up as need be in finish grind. THe trick will be to point the 16 or so lobes in 16 different angular positions. That will be the tough part as will moving X in and out to generate a decent lobe shape.

Essentially we're already doing it in X and Y axis on a mill, should be doable in solely X if we slow down the spindle and can get the X servo to respond quick enough...

Servos too have an "error level". This is why a servo has position and velocity feedback. THe PC tells it to move so far and fast and the servo does so albeit a bit slower than the instantaneous instructions - fast but not instantaneously identical.
In my Bridgeport system, the speed/direction signal is in the form of a +/-10 vdc full scale analog signal. The motors send back 2 signals - position to the PC and tach to the servo amp.

At this point, I'm not convinced that a CNC lathe will cut threads as reliably as a mechanical lathe. You're still "flying by wire" and an electronic gear is NOT as 100% positive as a mechanical one.


Now Modern servo's do not use velocity feedback, this was used on earlier systems that did not have the processing power to control the loop by encoder alone, If replacing drives on older systems many like AMC still include a tach option input if needed, but the favoured Torque or current mode is prefered, with the loop closed back to the control via encoder only.

One of the problems of too high a following error is if the servo's are not sized correctly to enable the system to keep within an accepted tolerance, this is part due to them not being sized in order to keep the motor to load inertia within the industry recommended maximum of below 5:1 (10:1 tops).
If I program an installation to keep withing a following error of 200 encoder counts which represents 10 microns, then following error is not usually an issue,
Actually the threading routine on a Lathe, (which is the equivalent to rigid tapping on a mill) , is more efficient because it is usually much faster, I remember the first time I programed a Lathe retrofit and tested the threading at 250"/min. :cool:
This was for the Aerospace industry so I don't think they would have stood for anything less than 100%.
Al.

hi

just searching on ebay.com , wooooo there is alot of stuff on there

come across some fanuc servo-motors

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=330013577155

If we get a viable lathe retrofit, we're going to try to rough turn a lobe shape into a cam blank. Don't have to be perfect and don't have to be smooth, just a reasonable lobe approximation. We'll tune it up as need be in finish grind. THe trick will be to point the 16 or so lobes in 16 different angular positions. That will be the tough part as will moving X in and out to generate a decent lobe shape.
Ever considered a mechanical solution for this? As for ultimate: a 1 on 1 driven, splined shaft drive with a multitude of 8 splines, desmo actuated tool. It's only for removing a lot of metal, so there must be a common denominator.

hi

just searching on ebay.com , wooooo there is alot of stuff on there

come across some fanuc servo-motors

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=330013577155

Can't be driven by Gecko's. Gotta watch what ya get to make sure it's compatible with step and direction type drives.

Eric

Hi

Guys some sad news , had just learnt Stevie a cnczone.com forum member has sadly passed away some time this week , spare a thought for Stevie and his family members.

I'm sure his work inspired a lot of members in here . The best machine work i have seen in a long time.

I know i was inspired by his work and i always read his threads , Stevie rest in peace .

My sincere condolence to Stevie's family and friends , Stevie rest in peace mate




Had a look at a small colchester triumph lathe that was converted by a crowd downunder yesterday , nice conversion the lathe is brand new and so is the Bosch