Help with BP J-head CNC quill drive

[titchener]

I have a Birmingham BP J-head clone that was retrofitted by an unknown LA company (might have been CNCMasters.com) with 3 axises in around 2000. I bought it used around 3 years ago, put Mach on it as a controller and its been a great machine for me.

However recently the Z quill drive has developed some excess backlash. The quill drive is pretty simple, it has a machined aluminum block attached where the original BP quill stop nut used to be, and that block has a ball screw nut in it, driven by a ball screw and motor that sits offset to the front above the screw.

This block that attaches to the quill appears to be a little loose, and is probably where the backlash is coming from. However, I can't figure out how to tighten it.

The picture below is what it looks like.

The aluminum block has the ball nut in it but no visible fasteners at all.

Does anybody know how these units were assembled and how I can tighten it?

Its not possible it was bolted from inside of the quill is it?

Thanks,

Paul T.

[titchener]

Well I was able to figure out part of the quill drive, by partially dis-assembling it I was able to see that there is a milled down socket head bolt holding the aluminum block to the quill. But its covered up by the ball nut cylinder.

So the next mystery is what is holding the ball screw nut in place. No visible bolts or set screws, and it also feels a little sloppy, I can rotate it around one quarter of a turn.

Anybody every take on of these fully apart?

Thanks,

Paul T.

[HimyKabibble]

Paul,

The ballnut would normally be screwed into its mount, and locked in place with a setscrew, roll pin, or some kind of clamp bolt. There really isn't any other way to do it.

Regards,
Ray L.

[titchener]

Ray L., you were right about the ball nut, by poking around on the unit I eventually figured it out last night, the ball screw nut is threaded on the end attached to the aluminum block, it was screwed all the way in so no threads were visible, but it was also a little loose and after I realized it would rotate a little bit I thought maybe it was threaded and it was.

The block design is actually not that bad, although there is only one bolt attaching it to the quill, its pretty hefty at 3/8", and the block has a big machined surface cut to match the quill curvature, so when correctly tightened its pretty secure and I don't think it will loosen in the future.

Unforunately they didn't do any of the things you described to retain the ball nut (no set screw, pin, etc.) and the way the thing is designed its not easy to tighten it that much, so I'll have to sort out something to keep the nut from loosening again.

Thanks for the help Ray.

Paul T.

[HimyKabibble]

Paul,

You might get away with backing the nut out a bit, and putting some Loctite on the threads, then tightening it back down securely.
If you have it opened up, would you mind posting a few pictures of the ballscrew assembly? I'm going to be CNC'ing my quill one of these days, and I'm looking to see how others have done it.

Regards,
Ray L.

[titchener]

Originally Posted by HimyKabibble Paul, I'm going to be CNC'ing my quill one of these days, and I'm looking to see how others have done it. Regards, Ray L.

Geez, sorry Ray, I actually meant to take some pictures while I had it apart but it was bugging me not to have it running and last night I thrashed on it until it was back together, and I forgot to take the pictures.

But its pretty simple. They used some 1.5" aluminum square to machine a block that fits flush against the quill through the slot where the original stop assembly normally mounts. They machined it nicely so it matches the curve of the quill. This has a through hole for a 3/8" socket head bolt, and the block matches the quill well, so when tightened its very solid, I torqued it to 25 ft-lbs with some red locktite.

As you can see in the picture in my first post, the block is also machined on the sides to fit in the slot. I guess there may be a little loss of Z travel because of the 1.5" height of the block, but not much, I have close to 5" total travel.

They then cut the block down to make room for the ball screw nut, its centered 2-3/4" out from the quill surface, and as mentioned earlier, its threaded into the block. They machined it down to a .400" "shelve" for the nut that it screws in from the top, if I was making one over I would make this thicker as even with the quill all the way up there is plenty of room for the nut, they could have make the shelve .75" thick with no problems.

They then made a housing that bolts to the front of the head to hold the bearings for the lead screw on either end. The housing has bearing caps on either end, one containing a belville washer for preload, and the other shims. You can see the top bearing cap pretty clearly in the picture, its pretty simple, just a square with the bearing recess machined in it.

They just used standard 6202 bearings for the thrust bearings, not the best choice for thrust usage, there's a bearing that is the same size but is designed for thrust usage that I'll probably upgrade them to (can't remember part number off the top of my head but its pretty easy to find).

The housing uses 1/4-20 bolts to attach it to the head on the flanges normally used for the manual stop screw ends. They drilled and tapped the holes in these flanges and used shims to square the housing up (well actually they hacked it with stacked washers, I made shims instead).

The pulley mounts on the top of the lead screw and there is a second housing that bolts on top of the first one that offsets the motor, up above the first housing.

Here's a shot of the assembly with the motor not mounted, so you can see how the lead screw mounts. The housing is the black steel, the aluminum strips you see on the side are just for mounting limit switches.

Its a pretty simple setup, probably not as rigid as the more common ones that bolt to the bigger head retention bolts on the BP head, but until this recent backlash issue came up its been working really well for me. I've made a lot of parts and get pretty good surface finish on both 6061 and 1018 steel.

It uses a 16mm lead screw, and probably the trickiest part is machining the lead screw just right as it has to have bearing shoulders machined at just the right length to allow the thrust preset to have the right range, but I guess that's something you can sneak up on.

This was a commercially produced drive (made out of formed steel), although I don't know where it came from, although it looks similar to the Z drive used by these guys - www.cncmasters.com , although instead of offsetting the motor with a belt drive there current is direct drive, which I'm not so warm on.

Let me know if you have any more questions.

Paul T.

[HimyKabibble]

Paul,

Thanks, that's very helpful. It's very similar to what I have in mind for mine.

Regards,
Ray L.

[titchener]

After completing the rework of the axis I'm pretty happy with the results, I'm now seeing way less than 0.001" total backlash on the axis, best I can measure with my indicator looks like 0.0002" or so, right at the limit of the ballscrew.

If I push and pull hard up and down on the spindle I can make a total "forced" (returns after removing force) deflection +/- .001", which could either be the bellville spring compressing, slop in the 6202 bearings (which are known to not be so hot in thrust) or just the overall flexibility of the system.

I was getting +/-.005" before the rework and it wasn't returning to zero, so I'm sure my chatter problems are going to go away now, but I might try upgrading the bearings to see if the deflection under force improves.

Paul T.