best way to CNC the quill z axis

[matridium]

Ok I have a knee mill and I have figured out most of the things I need to do for the x and y. However I don't want to move the knee, I think it is crazy to move so much metal to drill, and mill, IMO, but I think moving the quill is a great idea. however I'm stuck, how does one calculate for ipm, and everything else you need in order to get the z axis to work in the same measuring frame as the x and y if If the quill uses a rack and pinion? Can I just measure the distance of teeth and how many revolutions of the pinion is need to move 1 inch? Any ideas?

Mike

[scadvice]

Mike,
Do you have a Machinist Handbook? In the 27th edition it is page 2033 you want to look at. Otherwise lookup the Gearing Section. This will give you the basic's for calculating the Pitch Diameter (PD). Once you have the PD you just muliply it by pi. That will give you the linear distance for one rotation of the gear. If the gear is clean and build correctly you can find the PD by the gears outside diameter times its number of teeth and dividing that by the number of teeth plus 2.
Steve

[Torsten]

Dose your Quill have a scale on it?
The highest number on the scale would tell you how far it advances per rotation of the Handwheel.
If not you can do a rough messurement by marking a position on the handwheel then do one full rotation on the handwheel and meassure how far the Quill has advanced.
You can fine tune this later once you have the motor installed when setting up the Controller Software.

[matridium]

Thank you Steve

This is what I got

I took the quill apart, and the gears outer diameter is: 1.65"

The number of teeth is 26

So if I take the 1.65 x 26 = 42.12

42.12/(26 + 2) = 1.5042857142857142857142857142857 and this is the PD

Now if I take that 1.5042857142857142857142857142857 * 3.1415926535897932384626433832795 = 4.7258529489000746858588049751333 inches per 1 revolution of the gear.

Now what I think I do next is divide that by the pulley I am gearing it down with. That is 2.5... I think you get to this by dividing the driven pulleys teeth by the motor pulleys teeth (in this case 25/10 = 2.5) so I have a 2.5:1 ration...

So 4.7258/2.5 = “this is the distance the gear moves for every rotation of the output pulley? I think

I guess all I need to do next is divide the output distance of 1.925" by my encoder feed back I believe that is 2000 because I have a 500 line encoder with a quadreture (4x)...

So 1.925/2000 = 0.0009625
So knowing this i can figure out the resolution of my Z axis

If I take 1.925/0.000962 = 2001.0395 steppers per inch

I should be able to use the resolution to indicate how many steps in software is used to turn the shaft 1"..

I think this is right...

I also looked at the distance and looked at the revolutions of the gear. And it appears to move about 4.8 inch for a full revolution..

Thank you all for the help

Mike

Mike

[scadvice]

There you go... Torsten makes a good point though, most quill systems do use linear scales on the Z Axis.
The rack will always have some backlash issues and there have been used a number of methods to minumize it, such as splitting the spur gear and loading them on the rack in opposite directions with set screws.
Steve

[olf20]

Here is a picture of a knee mill Z conversion.
Hood over a Mach3 forum sent me several pictures
of his conversion. I hope to do mine some day, but
so far I get along using the rack and pinion.
olf20 / Bob

[M250cnc]

Mike, IMHO you would be crazy to put effort into a conversion unless it had near to zero backlash as per the photo of Hood's Z axis conversion

Phil

[matridium]

Olaf

do you have more pics of the Z axis conversion especially the quill part? i have decided to go with a lead screw... and split nut for now. I will upgrade to ball screws in the future as fund permit.

[jmelson]

Ok I have a knee mill and I have figured out most of the things I need to do for the x and y. However I don't want to move the knee, I think it is crazy to move so much metal to drill, and mill, IMO, but I think moving the quill is a great idea. however I'm stuck, how does one calculate for ipm, and everything else you need in order to get the z axis to work in the same measuring frame as the x and y if If the quill uses a rack and pinion? Can I just measure the distance of teeth and how many revolutions of the pinion is need to move 1 inch? Any ideas?

Mike

The typical rack and pinions have so much slack in them through the multiple gear trains that they are useless for CNC. Assuming a machine similar to a Bridgeport, here's what I did :
http://pico-systems.com/zaxis.html
I get about 0.0015" of backlash with this setup. To see how bad your mechanism is, just take a dial indicator and set it up to read vertical position of the spindle. Then pull the quill down and push up to see how much slack there is. There is generally NO WAY you can machine anything without locking the quill every time, it will creep up and down under cutting forces. Having to lock the quill totally defeats any way to make it work automatically.

Jon