[cncuser1]

Does anyone know of any proven designs where a Z-axis( for a benchtop mill) is driven by a Rotating Ballnut( incorporatinng drivebelt and appropriate bearings) instead of a rotating ballscrew.

I'd love to see some schematics

[Willbird]

I'm going to do it on a full sized Bport, have not sketeched it out yet. it makes sense to me to do it that way when your machine used bevel gears to originally drive the knee leadscrew like a lot of knee mills do.

[cncuser1]

I'm going to do it on a full sized Bport, have not sketeched it out yet. it makes sense to me to do it that way when your machine used bevel gears to originally drive the knee leadscrew like a lot of knee mills do.

I have a tabletop knee mill, I want to mount the stepper motor and ball nut on the base of the machine, and have only bearinngs installed on the knee.

It'd be nice to have some ideas to kick start my design.

[Willbird]

Well the way I see it, the ballscrew would just clamp to the knee, no bearings at all needed on the knee end. The ball nuts need to be held on bearings so they can rotate driven by the stepper or servo motor.

Simple, all you need to do is start working it all out :-).

Bill

[Zumba]

I don't see any benefit of doing it on a benchtop mill. Slapping a stepper on the top of the column to drive the screw is so much easier.

A router with a 10-foot X-axis is a different story.

[cncuser1]

I don't see any benefit of doing it on a benchtop mill. Slapping a stepper on the top of the column to drive the screw is so much easier.

A router with a 10-foot X-axis is a different story.

My mill has a knee already on it. It moves on the column.
the spindle is stationary ( except for a small z-axis mechanism that i want to eliminate, hence my post)

[Zumba]

I hear you. Is this the machine you have?

http://www.grizzly.com/products/G3102

[bill south]

Zumba;
My reason for looking for a driven nut design (on-going for a few weeks now with very little luck) is so I don't have to move all of that weight on my benchtop lathemaster ZAY series milling machine. I've installed a 3/4 inch ball screw in the column and driving with a 600 oz continuous servo. The balls wear out very fast and I have to replace them on every once in awhile. I have a second nut on order and will install a dual nut to help with the wear problem if I can't find a good, simple quill drive design.
Also by powering the head on a square column design, the head movement up and down the column is not very predictable. It jerks from time to time like it is sticking. I've tried every adjustment on the thing and it still sticks upon occasion. Also, I have both ends of the ballscrew achored to try to help this problem. However, the problem (whatever it is) is still there and rears it's ugly head at the most inconvenient time ususlly screwing up a good piece of metal.
Yes, driven quill is for me and I want to use a fixed ballscrew design. I saw a great setup somewhere but can't find it any longer. Somewhere out there in cyber space.
Any help will be appreciated guys. Lathmaster works great otherwise!
Good luck.
Bill

[bill south]

Hey Guys;
After looking at almost 1000 photos on my hard drive, look at what I found. Can we reverse engineer this thing??? Looks like a good design to me.
Wat ya thunk?????
Bill
:cheers: :cheers:

[Zumba]

Cool photo. I like the design. Looks pretty simple. Here's how I'd do it if I were copying it... take my advice with a grain of salt.

1.) 3/4" ballscrew and nut. Ballscrew is turned down to 5/8" on one end. Ballnut typically has a 1-1/8 x 18tpi male thread.

2.) Blue cast iron mounted bearing: 1.5" ID.

3.) Steel cylinder: 1.5" OD at the bottom, steps up to 2.0" OD in the center, then steps back down to 1.5" OD at the top. 1" ID bore all the way through. 1.5" diameter outside threads cut at the bottom. Top of cylinder is bored out a bit more with threads cut or tapped to 1-1/8" x 18.

4.) Flanged Timing Pulley, maybe a 42 tooth .200" XL, bored out to 1.5".

5.) Servo Pulley, maybe a 14 tooth .200"XL for 3:1 drive ratio.

Note that mounted bearings are typically just simple radial bearings. I would much prefer a dual angular contact bearing mounted in my own housing. But I guess those mounted bearings are cheap enough that you can just replace them every once in a awhile and save some labor.

[handlewanker]

Hi Bill, looking at the design, in post #9, gives me a bit of doubt as to whether the offset drive would be strong enough to remain true without flexing under load.
I would be more inclined to look at a way to have two ball screws, one on either side of the spindle to balance the load.
Using a rotating ball nut is going to be an expensive way to do it, apart from being more complicated.
Looking at the attached photo, could you fit a stepping motor in place of that fine down feed handle?
In this case I would make sure that there was positive up pressure on the spindle all the time, ( like a bench drill) so that backlash didn't come into it, probably by spring loading or counterweight.
The knee drive mechanism is probably only suited for hand winding, due to the weight, and if a stepper and ball nut were contemplated for this situation then some form of relief to get the load off of the nut would help, probably a hydraulic cylinder that assists lifting but still keeps some down pressure on the nut.
At the end of the day which way do you want to go? Table lift or spindle drive?
Both methods require different designs. Spindle drive has short travel and table lift has long travel but large loads.
I could not imagine profile milling a job by lowering and lifting the table to be very accurate, as opposed to spindle or Z axis drive.
Ian.

[Zumba]

Hi Bill, looking at the design, in post #9, gives me a bit of doubt as to whether the offset drive would be strong enough to remain true without flexing under load.

I don't think the offset is that big of a problem.

The problem is that in the photo, the bracket that attaches to the quill is only 1/4" thick. Adding another ballscrew to the other side would help, but no more than thickening the bracket and using a larger ballscrew.

The real issue to consider is the quality of the quill. Most bench mills have pretty crappy quills. If your quill deflects as it goes through its travel, even the sturdiest drive setup will not be good enough. It would be like using precision ground ballscrews with crappy linear motion guides.

[cncuser1]

I hear you. Is this the machine you have?

http://www.grizzly.com/products/G3102

No, mine is home made, about half the size of this one.

But the basic configuration is the same.
For the readers of this thread: How could the knee be driven, I don't want to cnc my quil.

[cncuser1]

Hey Guys;
After looking at almost 1000 photos on my hard drive, look at what I found. Can we reverse engineer this thing??? Looks like a good design to me.
Wat ya thunk?????
Bill
:cheers: :cheers:

That is what I am looking for!!!
do you know where it is sold, ?

What is the blue thing, it look like it has a lube nipple.?

I guess thhe big question is what bearings are used and in what arrangement?

[Willbird]

On the accuracy issue, it is not good practice in manual milling to mill with your quill down unless you HAVE to for some reason :-)...the quill is for drilling holes...and now and then to move the cutter into a place where it cannot reach with the quill fully retracted. The tools motion is not as well controlled with the quill as it is with the knee way slide which was designed for that purpose. The quill after all is simply a round part in a round hole with a farily crude method avail to clamp it in pkace, that most cnc designs do not even use...so we are then milling holding our tool with a round shaft in a round hole with slop present...not good at all :-).


In looking at mounted bearings like the blue one in the photo I have wondered if there are some designed to handle thrust.

I do think the pictured unit would benefit from some gusseting to the plate attached to the quill, looks kind of floppy to me :-).

[handlewanker]

Hi will, a good point. To mill with the quill right up would mean that the knee drive becomes a prime factor in Z axis movement and so I would contemplate gearing the stepper down a lot, for more precise incremental movement, as you only want to move to get depth of cut increases once the cutter is near the job, as opposed to X axis drive which really has to cover some ground, and requires a free running slide set-up.
In the end, retro fitting drives to an existing machine, not intended for the new purpose, is always a compromise instead of starting off with the requirements for milling and drilling and designing to these.
Take a Bridgeport for example, A very large and heavy machine, but I don't think it would be practical to drive the knee up and down as opposed to using the quill to half it's out travel for milling.
As far as quill lock is concerned, the ball leadscrew will hold it in position if it's braked and mounted firmly enough.
CNC milling is usually taking small cuts at a fast travel rate and the ploughing cuts normally asscociated with manual milling arent suited to CNC due to the weight involved.
When I worked in UK we had a bank of Cincinnatti vertical mills. The heads on these were/are massive, but they have a huge counterweight that allows the head to be moved with ease with a small hand wheel and power Z axis drive.
My ideal mill for CNC would have a table that moved as freely as a cylindrical grinder's, and they run on balls.
Second the X and Y slides would be firmly attached to the base, no knee, and the head, no quill, would slide on linear ball bearings.
Cast iron or steel fabrication throughout, no alluminium.
Last but not least a variable frequency drive for the main head spindle and side high speed spindle.
I would go for linear bearing slides throughout and ball screws on all axis.
Small cuts and many of them as opposed to ploughing, and spray mist cooling with total enclosed cabinet with extractor.
Table size would be about 30" X 8"
Z axis travel would be 12".
To enable a tool changer the main spindle taper would be 30 Int, with pneumatic draw bar.
Tool changer would be a swinging arm 180 deg type with vertical drum magazine and 10 tool stations.
I think alluminium could be used for the base and column and probably the head.
Ian.

[WayneHill]

http://www.homecnc.info/cncparts.htm

[Willbird]

Wayne, It looks as though they only used ONE ballenut there, I have in mind ot use two so they may be pre-loaded to take out backlash.

I am going to counterbalance my Knee with two 2" bore air cylinders, plumbed into a common tank, I call the tank the "boyles LAW tank" because Boyles law describes what air pressure changes occur inside the cylinder as it is moved throughout it's travel without an egress for the air. With a tank 12" in dia and 48" tall my rough calculations say I can balance the knee within 15 lbs for the 16" length of it's travel. So the machine thus setup I think will work decent with 1150 oz/in steppers running 1:2 reduction, we'll see :-). The quill having only 5" of travel would greatly limit the machine's range of motion.

The Z axis will have more mass to move, peck drilling is one thing I can see would not be optimum, I probably will just use a feed pause to break chips rather than drag the drill clear out of the hole each peck. Running one manual thats what most folks do anyway :-). I can always cook up a cnc quill if I get into lots of drilling, with it up all the way and locked milling will still be very solid.

I have ideas of a whole new head in my noggin anyway...there is lots of "extra" stuff on the head of a cnc convertted bport that has no function, and it is not as rigid as a purpose built head bolted to the same location would be....and said purpose built head could employ a 2 speed belt drive deal like FADAL uses, and air cyl alternately loosens and tensions two seperate belts to give a high and low range, and a vfd takes it from there. The head I speak of could go on the OTHER end of the ram as well, leaving the normal head stored on the back side of the machine when not in use :-)...I have an entirealy differant machine that I can use when I need to set the head at an angle.

Bil

[svenakela]

Wayne, It looks as though they only used ONE ballenut there, I have in mind ot use two so they may be pre-loaded to take out backlash.

Bill,
There are two ways to avoid backlash in a ballscrew system.
Either you have two nuts (as you said) pressing against each other, OR the nut comes with slightly bigger balls that forces the nut to roll tight. The groove in the nut and screw are not circular, rather "V-ditched", so a bigger ball makes both sides of the groove to have contact with the ball.

Regards,
Sven

[Zumba]

Wayne, It looks as though they only used ONE ballenut there, I have in mind ot use two so they may be pre-loaded to take out backlash.

Two nuts would help, but it's not always necessary depending on your application. For typical machining operations, backlash doesn't matter on the Z-axis. Just make sure the axis is zeroed after a downstroke of the quill and that your retract height isn't something ridiculously small like .002 above the workpiece.

If you're making 3D stuff it's a different story of course.