[jhowelb]

Well, I bought an E-machine from Costco for the purpose of running my equipment. There is no network out there and no anti-virus nor has there ever been an up date. I take programming to it on a thumb drive relative to the parts I'm cutting or when I update Mach3.
With Photo V carve, Sheetcam and Deskengrave I'm set for most anything I need to do. No mess, no fuss, no interference and no crashes. I wish I could demonstrate my system for you.

But then, I wouldn't want to intrude on a love (big grin) affair!

[acondit]

Well, I bought an E-machine from Costco for the purpose of running my equipment. There is no network out there and no anti-virus nor has there ever been an up date. I take programming to it on a thumb drive relative to the parts I'm cutting or when I update Mach3.
With Photo V carve, Sheetcam and Deskengrave I'm set for most anything I need to do. No mess, no fuss, no interference and no crashes. I wish I could demonstrate my system for you.

But then, I wouldn't want to intrude on a love (big grin) affair!

Thanks,
Alan

[acondit]

Well, I found out how to zero the x and z axii and I made my first cnc cuts on the lathe. The problem is that my camera battery went dead just as I started. I cut a radius on the end of my cam-lock lever for my tailstock. I will post a picture as soon as my camera battery is recharged. I should have changed the feed on the final cut to get a smoother cut, but live and learn.

Alan

[jhowelb]

Don't get too deep into those tailstock mods till you get a chance to see what I'm doing next! Here's some sketches.

[acondit]

John,

I already have mine done (third and fourth pictures). When I first did it, I just had a bevel or round-over on the end of the handle (first picture). So today I took the handle loose and turned a radius on it (second picture). Before I put it back in the camlock shaft, I polished the end a little on the buffer.

Alan

[jhowelb]

Awright! Good show!

[acondit]

I think the next step is to put on some kind of cover for the ballscrew.

John,

I have written some directions to myself on building the camlock modification. I got the basic idea from Steve Bedair's website. I think it is a lot simpler construction than your drawing. If you would like I would be happy to send what I have to you. I am going to put the directions up on my website but I haven't gotten around to it yet.

Alan

[acondit]

On the emc list I got the idea for an spindle encoder. The original implementation was done by Lawrence Glaister. It will use some of the optical switches that I built earlier. Two switches positioned so their signals are 90 degrees out of phase will provide the A and B quadrature signals and a third switch positioned slightly farther out will read the index hole. I used part of the gcode that he wrote but modified it for my purposes.

I cut the encoder wheel out of 1/8" tempered masonite. I will have to turn the edge slightly thinner on the lathe (I could have done it in the router when I made it, but it is easier to put it in the lathe than to try and recenter it and hold it down on the router.)

I am planning to mount it on the spindle where the drive gear currently is located. The drive gear is 12mm thick so I will make a mounting sleeve that will support the encoder wheel and with it occupy the same 12mm of space. I will also have to cut a keyway in it to go over the spindle key.

Alan

[jhowelb]

Not so sure about the masonite taking the RPM. I built a spindle disk from a cd. It turned into shrapnel scattered all over the shop. Ultimately a thin sheet of aluminum did the job. Be careful, my friend, protect those eyes!
jb

[acondit]

Not so sure about the masonite taking the RPM. I built a spindle disk from a cd. It turned into shrapnel scattered all over the shop. Ultimately a thin sheet of aluminum did the job. Be careful, my friend, protect those eyes!
jb

John,

It was a little bit thick for the opening in the optical sensor, so I chucked it with the 3-jaws through the center hole pressing out. I ran it at top speed on the spindle while I took about 0.010" off at a time for about the outer 1/2" and it seems fine. It tracks true and is actually a lot sturdier than a cdrom.

I think that the diameter may end up being to small though. I think I may have to make it about an inch or maybe 2 inches larger in diameter for the switch boards to be able to clear the pulleys. The larger the diameter the more likely the masonite is to have problems, so I may end up switching to a piece of aluminum. I have never machined aluminum on my router so, I may just machine it on the mill using my dividing head. Or maybe I'll be gutsy and try machining it on the router. I am a little worried about my router trying to cut aluminum at 20K rpm. It is not a variable speed. I am not sure what feeds (depth and speed) to use with a 1/16" carbide bit and I don't know if it will gum up.

Alan

[jhowelb]

A two flute bit and a lubricant (I use WD-40) will help with the alum.

I haven't yet been able to decipher exactly what your up on this little project.

Good luck, tho.....I'll be lookin over your shoulder!

[acondit]

Well I haven't finished the spindle encoder yet, so I can't thread on my cnc lathe. However, I have cut a couple parts with it and it works great.

One of the projects that I have in the works is a cnc conversion on my X2 mill. I will start a thread on it later. I took some of the ideas for the z-axis from Dan Kemp (Hoss26) and some from Ron Stirling.

I just couldn't resist trying to cut some of the parts on the cnc lathe. I cut the internal hole for the adjusting nut with a little cnc program and turned down the external diameter with another little program. I just faced it by typing in the gcode commands to the MDI interface (I think that I will add a facing routine to my software collection). I had to do the internal threading on my other lathe (can't wait till I get the spindle encoder done.). And I cut the tightening slots on my mill (no live tooling). The threads on the spindle are 1.156 x 32 and the nut was internally threaded to match that.

The first picture shows the part mounted in a 3" 4jaw on my lathe ready to run the facing operation. The second picture shows the finished nut on the rotary nut spindle. I still have to take it back apart and finish the corners on the base piece.

Alan

[wlwoodall]

I previously posted this info in another thread. However, since I am now working on my conversion in earnest (I got my router running). I thought I would go ahead an start my on lathe thread.

Here are a few pictures of my variable speed conversion.

I used a Pacific Scientific DC motor that had a shaft the same size as the OEM Emerson motor (5/8"). This allows the use of the original pulleys and belts.

remainder deleted

Alan

Hello Alan:

I have a metal lathe and am trying to upgrade it to CNC and want to start at the main spindle motor and it sounds like you've already done some work in that direction already. It is a AC motor rated at the typical 1725RPM at 3/4 HP. What would be the ratings for a good typical servo motor (with encoder) to replace this? Servos are rated in oz./in. and I'd like to run this down to 200 RPM if possible.
Thanks for any help you can provide.

Bill

[acondit]

Hello Alan:

I have a metal lathe and am trying to upgrade it to CNC and want to start at the main spindle motor and it sounds like you've already done some work in that direction already. It is a AC motor rated at the typical 1725RPM at 3/4 HP. What would be the ratings for a good typical servo motor (with encoder) to replace this? Servos are rated in oz./in. and I'd like to run this down to 200 RPM if possible.
Thanks for any help you can provide.

Bill

Bill,

I used a 1 1/2 hp DC motor because I felt like the lathe was underpowered at low speeds anyway.

746 Watts equals one horsepower. Power (Watts) = RPM times in-oz / 1351. Figure out how many hp you need at what RPM. Then (HP * 746 Watts *1351)/RPM = in-oz torque. So (1.5 hp * 746 *1351)/3000rpm = 503 in-oz torque. I think. There are a lot of formulas posted at various places on the web. I got this from a document that Mariss Freimann wrote. I honestly don't know if I used it correctly though.

Alan

[acondit]

Bill,

I have seen a formula related to hp required for metal removal rates. That is how many horsepower it takes to remove so many cubic feet of metal per minute at a certain rpm but I don't know where to find it right now. Some of the other factors that come in to play are: what gearing do you intend to use? how deep of cuts do you want to make? and what kind of material are you going to be machining?

If you are running the servo at 200 rpm it will take a larger servo than if you are running the servo at 5000 rpm and gearing it down 25:1.

Alan

[wlwoodall]

I want to stick with the stock gearing. My needs are not challenging. Mostly removal of aluminun and brass.

I have a motor attached to a 90 degree angle plate on the milling table (kind of like a poor's man's horizontal mill). By moving the milling table left or right I can mill longitudinal slots and grooves into the material held in the lathe spindle (after being machined to the proper diameter which is why it's on the lathe to start with). Currently if I want to mill four slots 90 degrees apart along the shaft I have to manually rotate the 3 jaw chuck 90 degrees, lock the spindle, make my cut, rotate another 90 degrees, make my cut, etc... I currently use a makeshift indicator aligned to marks printed on the jaw chuck. This is a manual process prone to cumulative errors.

Why not take the material out of the lathe and attach it to the table and mill slots using the vertical mill, you might ask? The problem is that I make round tapered and square tapered shafts and trying to hold tapered items in a vise is impractical. Also, as long as I don't move the tailstock taper, the depth of all slots and grooves will be consistent.

I would like to use cnc to rotate the main spindle 90 degrees, have some decent holding torque, have the cnc move the x and y axis to make the cut (x axis cutting the length of the spindle, y axis cutting the depth). Have the g code turn the spindle 90 degrees, repeat until all slots are cut. The depth of cuts are very shallow, less that .125" and multiple passes are allowable so the holding torque doesn't have to be tremendous.

Then when a new part is loaded in, I use the serve to rotate the stock at 600rpm, use a bit in the tool post to reduce the diameter, then command the servo to stop at the 0, 90, 180, 270 positions while the horizontal mill cuts the slots. The main spindle servo would be used for two different functions.

This is why I was so interested in your modification of the lathe spindle motor. I need a cnc controlled spindle motor which can command the 3 jaw chuck to rotate through a prescribed number of degrees and hold with some reasonable torque.

Does it seem like your 1 1/2hp DC motor with an encoder will do the trick in a cnc system?

Thanks for any suggestions and sorry for writing a book.

Bill

[acondit]

I want to stick with the stock gearing. My needs are not challenging. Mostly removal of aluminun and brass.

I have a motor attached to a 90 degree angle plate on the milling table (kind of like a poor's man's horizontal mill). By moving the milling table left or right I can mill longitudinal slots and grooves into the material held in the lathe spindle (after being machined to the proper diameter which is why it's on the lathe to start with). Currently if I want to mill four slots 90 degrees apart along the shaft I have to manually rotate the 3 jaw chuck 90 degrees, lock the spindle, make my cut, rotate another 90 degrees, make my cut, etc... I currently use a makeshift indicator aligned to marks printed on the jaw chuck. This is a manual process prone to cumulative errors.

Why not take the material out of the lathe and attach it to the table and mill slots using the vertical mill, you might ask? The problem is that I make round tapered and square tapered shafts and trying to hold tapered items in a vise is impractical. Also, as long as I don't move the tailstock taper, the depth of all slots and grooves will be consistent.

I would like to use cnc to rotate the main spindle 90 degrees, have some decent holding torque, have the cnc move the x and y axis to make the cut (x axis cutting the length of the spindle, y axis cutting the depth). Have the g code turn the spindle 90 degrees, repeat until all slots are cut. The depth of cuts are very shallow, less that .125" and multiple passes are allowable so the holding torque doesn't have to be tremendous.

Then when a new part is loaded in, I use the serve to rotate the stock at 600rpm, use a bit in the tool post to reduce the diameter, then command the servo to stop at the 0, 90, 180, 270 positions while the horizontal mill cuts the slots. The main spindle servo would be used for two different functions.

This is why I was so interested in your modification of the lathe spindle motor. I need a cnc controlled spindle motor which can command the 3 jaw chuck to rotate through a prescribed number of degrees and hold with some reasonable torque.

Does it seem like your 1 1/2hp DC motor with an encoder will do the trick in a cnc system?

Thanks for any suggestions and sorry for writing a book.

Bill

I am not trying to convert my DC motor to a servo. I am working on an encoder that will mount on the spindle, not on the motor. However, I am thinking about adding spindle speed support (considerably different from full servo support). I am not sure that the motor and the stock drive train would have enough hold for what you are wanting to do. Even if the motor has enough power, I don't think the stock drive train could hold accurately as a "4th axis".

You said it isn't a very challenging thing that you are trying to do, but powering a spindle to turn is significantly different from powering a spindle to hold position against the forces of milling an item held in the spindle. I really question trying to hold position like a 4th axis with a simple belt drive.

You might want to think about a mill with a geared 4th axis. If you can hold the part in a lathe chuck, then you could hold the part on a mill in a lathe chuck mounted to a stepper motor driven rotary table.

Alan

[acondit]

Well I ran into a problem earlier with my spindle encoder effort. My optical switch was too big to mount between the back of the headstock and the spindle pulley.. So, I re-did my optical switch pcb to shrink it a little bit. I redesigned it to use surface mount technology. Then I got sick for a couple of weeks

Finally, back at it today. Wow, are those parts small. When I was putting some boards together today, I lost one of the resistors. I looked all over for it. Finally, found it stuck to the palm of my hand.

I have to wear a magnifying visor to see what I am doing. I built 5 of them in about an hour this afternoon. The first one I messed up and when I tried to fix it, I spoiled the pcb. The next four when together fine and all worked.

Now I have to figure out the mounting and wiring.

Alan

[Runner4404spd]

i have a question about your sensor? i'm basically running an optical switch to give me a feed back pulse as a spindle timer for threading. when i look at mach 3 with the spindle off i'm reading about 200 RPM and its fluctuating. i am wondering what you guys are seeing on mach 3 when the spindle is off? also any ideas on how to correct this would be great.

thanks
Jerry

[jhowelb]

You are probably picking up an errant signal.
Try moving your cable about in it's routing from spindle to connection. Maybe shielding for BOB/driver/power supply/A.C. inlet?