RF-31 Round Column Mill CNC Conversion Build Thread

Hello Gents,

I've been doing a lot of searching and reading here on this site. I've had an RF-31 round column mill in storage for a long time and finally have a place to put it after moving around a bit between places. I'm planning a CNC conversion starting with the X&Y first and then moving along to the Z. In a perfect world I would have known more 10 years ago when I got the RF-31 and instead purchased an RF-45. However, I think this mill will be suitable for my needs as I'll be using it to build fixtures for bicycle frames. All of the parts will have limited z-axis movement as they are pretty small. I figure I'll learn the ropes getting this mill converted and optimized then moving along to an RF-45 should I feel the need and sell the 31.

I'm planning to document the conversion in detail and also offer up any of the CAD files and/or plans to the public under the free GNU license. Hopefully the information will be useful for future people doing the conversion. This will be my first CNC conversion so documenting it will likely help me the most since I'll be able to find my details for the next machine!

In the mean time I'm curious to start the build with the following bits from Keling. They look like a good source of hardware from what I have read here. I want to do a direct drive of the screws from the motor to keep it simple. I thought about a belt drive from the motor to the ball screw but it seems a little elaborate for a 1st conversion.

Here are the parts from Keling I'm considering

(3) 570 oz-in stepping motors, NEMA 23
(3) KL-5056 Drivers (5.6Amp 24-50 Volts, 1-1/256 microstepping)
(1) 48VDC /12 .5A Power Supply 110VAC/220VAC
(1) C10 Breakout board
(1) 5V power supply

I'm also looking to get some ball screws from linearmotion2008 on eBay or from McMaster. McMaster has a warehouse local to me here in LA which is great for a quick will call pickup. I have a 12x36 lathe to machine the ends. The ball screw is one of the areas where I need to do more research as I haven't worked with them in the past. Does anybody have good recommendations for ball screws and/or the nut specifically?

Any feedback on this would be appreciated, let me know what you would or have already done!

I'll be following your conversion with interest to see how someone else would do it. I converted mine back about 2000 and have enjoyed using it. Before conversion, I fitted a 3ph motor and VFD drive. That part was very worthwhile and I really like the ability to change speeds with minimal belt changes. Also be aware if you find any plans to be sure you double check the various bolt locations on your machine. It seems that no two machines are the same! Mine is a genuine Rong Fu and when shared plans came from other sources who also had the same machine, we found discrepancies between the machines.
As for ball nuts, I used a pair and spring loaded them to give zero backlash. I definitely recommend doing a zero backlash or you will run into problems when doing cnc machining. It will also help making round hole machining possible. Backlash is NOT your friend:{(

Good luck with your build and If I can help, let me know.

Hey Bubba, I was actually hoping you would comment since you seem to be the resident RF31 CNC expert. Mine is a genuine RF machine as well and hopefully I get as much use from it as you have from yours.

Agreed about the ball screws. I've been doing some googling and have found some of the following information.

5/8" screws seem to be about the right size in diameter. The guy from this link RF-31 Mill CNC — Baka Mecha looks like he used 25mm size screws and they were a little big for the amount of space in the X-axis. The nut needed to be ground off a little as shown in the picture below.

Attachment 258452

I'm not familiar with the type of flange style ball screw nut he is using. Is this style of nut preloaded as a self contained unit? I've also seen the double nut with a spring in the middle style shown below in this next picture.

Attachment 258454

Then there is this idea to re-use the original acme threaded part. I'm assuming the nut he is using is already preloaded as well since there is no spring to be seen.

https://www.youtube.com/watch?v=TDEoj5HSKXU

I agree with the use of a 5/8" ball screw. Based on the loading charts that I have seen, it is plenty stiff enough for our machines. The ball screws and nuts that I got came from Roton and are rolled. This means that you may have a bit of a "drunken" thread and need to spring load any double nut that you have or it will either be loose or may bind up at certain locations on the screw. Based on what I know (which ain't much) it appears the flanged nuts you show are single circuit and "may" have considerable backlash. It is best to get the specs from the supplier at to what backlash can be expected from any particular product. As I remember (if I haven't completely lost it) I used a spring pressure of about 270 lbs to preload the nuts and this was in accordance with info I found on one of the manufacturers sights.
As for the nut mount, using the Y axis one might be fine, but on the X axis, the first one I made was similar in nature to the original one in that it was fastened down by two bolts in the original holes. Well, in short order, I found by backlash was increasing drastically. Pulled the table and found the two bolts had loosened enough to allow the mount to move under the stresses of acceleration. To solve this problem, I made a new mount that was wider and in addition to the two bolts, I used two dowel pins to secure the mount. Haven't had a problem with that since.

Also, make sure your stepper mounts will allow the table to "over travel". If you run the table all the way to the stops, you will have dovetails exposed on one end of the table. I made the mistake of mounting my servo to the end of the table and that end will not "over travel" and I lost about an inch of X movement. You also indicate that you want to use direct drive. Make sure the stepper does not come above the top of the table or you may have problems with stock that extends beyond the end of the table.

HTH

tybeede,

My double nut design is the one you show in your 2nd picture. I will say that the concept worked, but the spring pressure was too low on that design. I eventually reworked it to use wave washers that provided much more preload. I think I wound up with about 175 lbs of prload and that worked well onthe RF-25ish size machine I had. Sorry, but I didn't get pictures of that one, but the design was similar to my G0704 which you can see in the link below

http://www.cnczone.com/forums/bencht...67662-cnc.html

Thanks for the link to your RF45 build. I spent quite a bit of time reading through your conversion of the round column machine. I'm hoping to get one that works as well as your conversion! How have you liked the keling stuff so far?

When you did the z-axis how did you take out all of the rotational movement and/or play when the quill is not locked? I noticed that without the lock engaged that I can grab the end and it will rotate about the axis of the quill a little. When the lock is on that isn't a problem. It looks like the rotational degree of freedom is restricted by a key way in the quill but mine seems to have a little slop between the key and the slot or something.

I tuned up the mill last night. Started by replacing the belts with the Fenner Drive brand stuff ordered through McMaster. This should help with the vibration a little. I think it might be worth taking apart the spindle and having a look as well to get it cleaned up. I wonder if the RF brand comes with better bearings than the other clone machines too or if I should swap them to something of high quality from McMaster.

Attachment 258572

Then I had a look at tramming the machine which went pretty smooth. Started by pulling off the head and round column using an engine hoist. I used a gasket scraper to scrape away all the putty stuff at the edge and cleaned and oiled the flat surfaces. That way I could slip in a bit of shim stock and know I wasn't getting a bunch of gunk in with it.

Attachment 258574
Attachment 258576

It only needed about 0.011" worth of shim stock on the right hand side to get it squared up. When I put it back together I used some grade 8 lock washers I had laying around as they were a little more beefy. Torqued the four bolts to 90 foot pounds which felt about right. From what I could tell the tramming operation was successful and the X and Y measurements were as square as they were going to get for the night. It looked to be about .0005" in X and nearly dead on zero for Y over about 8". I'll have a better look at it the next time I'm down there and see if I can't measure it with a finer indicator after doing a little stone job on the table to clean off any gunk and high spots. The picture below shows the shim stock before I was finished, at the end after everything was torqued I used a razor knife to score it and then flexed it back and forth a bit to get it to fail at the score mark. It ended up being nice and flush with no sharp spots all around.

Attachment 258578

I also made a replacement for the cracked and brittle old plastic guard for the ways and y-axis screw. The original one died so I used the old metal hardware from it with a bit of buna-n 1/16" sheet from McMaster. Got a 12" x 14" hunk of it for about 9 bucks and it worked well.

Attachment 258580

Next step is to take some measurements and figure out how the ball screw conversion is going to work out.

I'll be watching your progress. My RF-30 has been in the conversion process for years now. Hopefully I will get it done soon. I have X and Y axes moving, still working on Z. Like you, if I had know better I would have gotten a square column mill. Live and learn.

Like Bubba, I converted mine to VFD drive and would never go back. I got lucky and got the motor very cheap as well as a very cheap enclosure. Even if I had needed to pay more for the motor it would have been worth it. Mine is only 2hp, not 3hp like Bubba.

I used ball screws from McMaster. I got them years ago, Thompson brand. They are the 5/8" size which I think is fine. Today I would just get screws from China, maybe with double nuts. I attempted to make backlash free setup with two nuts. It was never very smooth and getting the preload right was fiddly. I ended up using a single nut with oversize balls. Backlash now is about .002". I may fiddle with ball sizes a bit more but I doubt it. With the cheap prices for China screws I don't think it's worth it to roll your own now days.

I had a hard time getting the Y axis yoke stabilized. I'm still getting a little backlash on Y from what I think is the yoke rocking back and forth. I'll be interested to see if you come up with something better than I did.