[auburnCNC]

I own a Wholesale Tools ZX45 RF clone. I am looking into CNC’ing the machine. I am just starting to look into this, so I appreciate the advice. I have a number of questions that I have not been able to easily find the answer to on the forums.

I noticed that my X and Y have 1 inch or ~ 25 mm screws. Looking at the Z-axis and it looks like it is 1 1/8" or 1 1/4", I didn't put a caliper on it yet. I noticed some guys on here are going with 5/8" ballscrews on the X&Y and 3/4" on the Z. Given these things will be spinning a lot faster, does the smaller size pose any problems? Are there any clearance issues with the larger ballnuts? I have not done any calculations yet, just trying to get the heads up on pros and cons.

I also noticed that my X axis is completely floating on one end, no bearing. I assume you guys are putting a floating bearing on that end? Can it be left floating (no bearing) using a ballscrew?

Also, has anyone tried turning their screw ends? I would like to keep the original bearing supports—looks like a pretty simple lathe job on the ends. I have a WT 12x24 belt drive lathe that should handle it. I was reading that you can place the screw in water and heat the exposed end to soften it up for machining preventing damage to the rest of the screw.

Thanks

[arizonavideo]

For the shorter tables a 5/8" is fine. For the Z 3/4" is the smallest but 1" might be better.

On my IH I'm running a 1" on the Z a 30mm on the table and a 16mm for the Y.

The table is a lot bigger than a standard RF-45 so the bigger screw is a safer bet on the X and I got a good deal on it or I might have done a 3/4"

On a RF-45 you rarely do rapids over 100 IPM so screw whip is not much of a problem and the Y is short. It is more of a load/ preload question and if the ballnut has good preload it will work fine.

[auburnCNC]

For the shorter tables a 5/8" is fine. For the Z 3/4" is the smallest but 1" might be better.

On my IH I'm running a 1" on the Z a 30mm on the table and a 16mm for the Y.

The table is a lot bigger than a standard RF-45 so the bigger screw is a safer bet on the X and I got a good deal on it or I might have done a 3/4"

On a RF-45 you rarely do rapids over 100 IPM so screw whip is not much of a problem and the Y is short. It is more of a load/ preload question and if the ballnut has good preload it will work fine.

Thanks for the quick reply. My table is about 31" and I think the y screw is about 40" overall. The X axis screw is 20"overall and free floats on one end. I want to use the old bearing mounts on the X and Y, so I need to stick with those lengths. It looks like the z will be more trouble. I am planning on going with steppers and ebay screws most likley.

Does moving that heaving mass “head” rapidly cause deflection during machining? I understand it gives you a lot more travel, but it seems like moving the quill would be more accurate and then you would not have to worry about supporting the head during power off with ball screws. Has anyone done a comparison? I guess you could motorize both?

[michaelthomas]

Moving the head is fine. I use "gas springs" on mine inside the column to provide counterweight force. It's hard for me to remember, but I think I used 2 100 pounders.

The quill has so much backlash in its stock form, that it would be difficult to overcome without a bunch of work......then you would still be very short on travel.

Generally speaking, the x is the length of your table, and the y is the short way.

I used 5/8 screws on the x and y, and a 3/4 on the z. Works great.

I had to lower the Y ballscrew to make some room for the ballnut......I milled the base a bit , too.

It's not too bad of a project......but if you are planning on using your ZX to make its own parts, you should be prepared to disassemble and reassemble it a whole bunch of times.

If you anneal the ends of the ballscrews, you will have no trouble machining the ends. I used a oxy/acetalene rosebud to heat mine to a very light glow, then placed the end in a bucket of sand to let it cool slowly. Turned easy after that.

Mike

[Runner4404spd]

forget the quill its got too much play when extended. the quill was only designed to drill and to use an endmill. granted if your manual machining you can get away with it since your constantly measuring your work, but with cnc you want repeatability and hence you want everything as ridgid as it can be. that means no quill slop.

[auburnCNC]

Thanks guys, great advice. Yeah, I will be in the take apart and put back together camp. Hardest part will be convincing my wife:cheers:

I am trying to save on cost and plan on going with stepper motors. Can you get by with 425 oz-in motors on the x and y with 5/8 C7 ballscrews? Maybe 1600 oz-in or less for the Z or can you go smaller if you counter balance? What size steppers are you guys using?

I am looking at ebay’s wantmotors. They seem to have a good ebay rep, so I am guessing their stuff can’t be too bad. Are their drivers any good? They look nice on paper, but who knows???

Thanks

[TiagoSantos]

I used 425oz/in nema 23 motors on my RF40 for the X and Y with the stock leadscrews. It wasn't great, but it worked. With the added efficiency of ballscrews, I think they'll work fine - that's what I'm hoping anyway. I'm not expecting 200IPM rapids or anything like that. I was OK at 50IPM with the stock leadscrews, so I'm hoping for 100IPM with the added efficiency and better (who am I kidding.. not better, some!) oiling to the ways.

[TiagoSantos]

As for motor drivers.. I've seen enough people have never ending trouble with random chinese drivers that I went straight to what I was sure would work out of the box. I'm running a gecko g540 that I had purchased for a smaller machine, but if I knew what I know now, I would have gone for the separate higher rated geckos (so I could eventually run big arse nema 32 steppers :P)

[auburnCNC]

I will be happy with slow as well—it is a cheap machine toy after all. My guess is the faster you go the more inaccurate these machines are—just my guess…

Yeah, I worry about buying ebay drivers. Anyone have a good ebay driver experiences and what company? I am sure it is hit and miss.

Has anyone measured the minimum torque required to move the x,y, and z axis using 5/8" and ¾" C7 ebay ballscrews. I understand it will vary, but having an idea will help in motor selection—at least make me feel better…..

[TiagoSantos]

Going faster shouldn't affect accuracy as long as the stepper motors aren't skipping steps - as in, at some point, if you're trying to accelerate or deccelerate the table too fast, for instance when changing directions, the inertia/weight of the table might overcome the torque of the motor and it'll skip steps. The controller won't know about any of this and your parts will be ruined With the right motors though, you can go to town on speed and the accuracy will be just the same.

Anyway, for drivers and motors, I'd try to figure out what your budget is first.. Figure out how much you can afford or are willing to spend, then see if you can get something acceptable for that much money.. Keep in mind this is an expensive hobby/toy. You'll be spending lots of money on software, stock, misc parts and connectors and wires, etc. I'd say the minimum you can get away with for X and Y is the 425oz/in motors you mentioned. At $50 each, plus just under $300 for a gecko g540, that isn't a lot of money.. But if you can spare the extra couple hundred bucks, it would definitely be worth going for the separate gecko drivers and some bigger motors.

[109jb]

If you want to see a RF45 size machine running on NEMA 23 motors, check the link below.

The videos show that it can be made to work, but take note of a few things in the video and the build. He is using Gecko G250's pulled from a G540 with 620 oz-in NEMA 23's on x and y, and a Gecko G203V and 1290 oz-in NEMA34 on the Z. I look at the videos and I notice that the acceleration seems pretty slow on all axes. I'm not sure if he later tweaked those settings to get better accel or not. Also note that he has belt drives on all axes and I'm not sure what ratio he is running. I tried to contact the builder in that thread via PM but got no response. In a nutshell, it will work, but may not satisfy. If you are willing to live with somewhat slow speeds it may be an option though.

http://www.cnczone.com/forums/bencht...45_g9972z.html

[auburnCNC]

Thanks for the info and the video. Once I gather all the information I can, good and bad, I will set down and hammer out some calculations and then make a decision on screws and motors. I would like to figure out the min torque first to move each axis. Just thinking, if I determine the minimum torque on my ACME screw assuming say 30% efficiency and assuming 90% for the ballscrews, I could estimate the torque requirements for the ballscrew--at least give myself a starting point on paper.

[TiagoSantos]

If you google hard enough, there is a simple formula to calculate torque needed to move a known weight with a given screw. I can't remember what it is or I'd tell you! I'd guess the force needed to move the table with tight gibs is probably around 30lbs? That doesn't account for inertia when changing directions though, which would depend on how much crap is bolted to the table 50lb vise? heavy fixture plate?

[auburnCNC]

If you google hard enough, there is a simple formula to calculate torque needed to move a known weight with a given screw. I can't remember what it is or I'd tell you! I'd guess the force needed to move the table with tight gibs is probably around 30lbs? That doesn't account for inertia when changing directions though, which would depend on how much crap is bolted to the table 50lb vise? heavy fixture plate?

Thanks

Yeah, I know there are lots of different variables to consider. I actually have an mech engineering background. I have some books I can dust off to figure it out and get a ball park figure.

[Runner4404spd]

it really depends on how fast you want it to go. i think that the 5/8 screws will be too small. i would go with 3/4 screws. the 5/8 screws won't be able to have enough preload on them for zero backlash or at least it may be an issue in heavier cuts.

on mine i run 400ozin servo motors with a 4:1 gear reduction netting me 1600ozin of torque for cutting not counting the screw multiplication as well. the steppers are really going to limit how fast you can push it. and if you run direct drive steppers you will most likely not like the cheap 5/8 ball screws as the pitch is too fine.

[auburnCNC]

it really depends on how fast you want it to go. i think that the 5/8 screws will be too small. i would go with 3/4 screws. the 5/8 screws won't be able to have enough preload on them for zero backlash or at least it may be an issue in heavier cuts.

on mine i run 400ozin servo motors with a 4:1 gear reduction netting me 1600ozin of torque for cutting not counting the screw multiplication as well. the steppers are really going to limit how fast you can push it. and if you run direct drive steppers you will most likely not like the cheap 5/8 ball screws as the pitch is too fine.

Thanks, more good info for thought. I don’t think I can afford servos at this time. I guess if I went with larger steppers, not much more money, I should be able to run a belt drive—have to look at the torque curves. I’d rather not have them sticking out and direct drives would limit my ability to tune the system using different pulley configurations.

If it works out on paper, I might stay with 25mm screws. It gives me enough meat to turn them down; I can reuse the old bearing mounts, and use some larger angular contact bearings.

[arizonavideo]

This is one of the best steppers out there. Buy three.

High-Torque Stepper Motor, Stepper Motor, Driver, Stepper Motor kit, DC Servo Motor, DC Servo Motor kit, Stepper Motor Power Supply, CNC Router, Spindle, and other Components. Automation Technology Inc

For the Z run a three to 1 reduction and direct drive the X and Y.

You should run full power drivers so you can't go with the chip drivers. You could look at the Gecko 201X

G201X 10 Microstep Digital Drive

You could save some money and get the China Digital drivers. They look fine from what other have said. I have tried some of the older non digital and did not care much for them.

Stepper Motor Driver Automation Technology Inc

As you can see the drivers do add up. Don't go crazy on the power supply if you make a transformer one they will deliver 2 to 3 time the instantaneous current you might need so a 56V 5.4A should be OK.

Antek - PS-3N56

If you buy smaller stepper or smaller drivers you will pay a large price in performance.

You could go with the G250 drivers and the smaller steppers but you might want to add belt reduction and then the cost is about the same as a direct drive setup.

Good luck.

[auburnCNC]

This is one of the best steppers out there. Buy three.

High-Torque Stepper Motor, Stepper Motor, Driver, Stepper Motor kit, DC Servo Motor, DC Servo Motor kit, Stepper Motor Power Supply, CNC Router, Spindle, and other Components. Automation Technology Inc

For the Z run a three to 1 reduction and direct drive the X and Y.

You should run full power drivers so you can't go with the chip drivers. You could look at the Gecko 201X

G201X 10 Microstep Digital Drive

You could save some money and get the China Digital drivers. They look fine from what other have said. I have tried some of the older non digital and did not care much for them.

Stepper Motor Driver Automation Technology Inc

As you can see the drivers do add up. Don't go crazy on the power supply if you make a transformer one they will deliver 2 to 3 time the instantaneous current you might need so a 56V 5.4A should be OK.

Antek - PS-3N56

If you buy smaller stepper or smaller drivers you will pay a large price in performance.

You could go with the G250 drivers and the smaller steppers but you might want to add belt reduction and then the cost is about the same as a direct drive setup.

Good luck.

Thanks again, great links and information. I am looking at this as an educational experience not just a CNC build.

[cjsamples]

If you want to see a RF45 size machine running on NEMA 23 motors, check the link below.

Also note that he has belt drives on all axes and I'm not sure what ratio he is running.



http://www.cnczone.com/forums/bencht...45_g9972z.html

He mentioned in one of his post all of the belt drive is 1:1 ratio if that helps. Maybe I should have read the rest of the post on here to see if someone else picked up on it.