[rusmannx]

Hello!

I have some 1605 ballscrews from linearmotionbearing2008 (in my CNC G0704) and have had good luck for the first year or so that I've had them. Lately I have been getting some weird backlash and error and I feel confident that I've ruled out everything but the screws themselves. These screws take a .125" ball, so I ordered some .126" (they mic at about .1259") to try and tighten up any slop but I just can't get them to fit. These ball nuts are supposed take 16 balls per circuit, but I can't get more than 13 balls to fit in the first circuit I load. Also, with the 13 balls loaded the screw is so damn tight I can barely turn it a couple of turns, then it binds.

My question is, has anyone else used these screws and successfully loaded the nuts with oversized balls? Can you tell me the details? I've unloaded and loaded ballnuts before so I know how to do it, I just can't seem to get this combination to work. I know guys are using oversized balls in the Roton screws, but haven't found anyone mention they've done it in the C7s.

Thanks guys!

[hoss2006]

You'll have to try smaller balls if you can't mount the nut with those balls.
You don't say how much lash you have.
The C7's come much more tightly packed than rotons do so it doesn't take a ball that much oversized with them.
My C7's had .1245 balls that I measured.
Hoss

[rusmannx]

The balls I took out average about .1248 and the ones I am putting in are all .1258. I have anywhere from no backlash to .004" depending on where I am at on the screw. Even worse, I have anywhere from 0-.004" in error depending on where I am on the screw (backlash is all over the place) and in some cases I'm off by as much as .008". Something weird is going on and I have taken the whole thing apart multiple times checking and rechecking it.

[hoss2006]

Does the axis move freely from end to end, could be stiction of the gibs giving erroneous readings.
Does the lash occur in the same spots?
If the axis moves smoothly and it gives the same error in the same section repeatedly it could indicate a bad ballscrew, lead errors are supposed to only be .0018 but you never know.
Might be a good time to upgrade to the double nut C7's.
Hoss

[rusmannx]

I don't know if I would say "freely" from side to side, as I run my gibs pretty tight.... but I always have. I considered stiction, but am not really sure how to test for that. I'll clean everything really well and try some different oil (I have more Vactra 2 on the way).

I have considered those double nut C7s, your setup looks pretty slick. I didn't know he was offering them to the public and assumed they were more of a one-off (I haven't checked his ebay store in a while).

Thanks for the ideas Hoss.

[hoss2006]

I haven't looked thru Chai's ebay page to see if he has them on there yet but you can request them from him, I've had many customers already get them.
Hoss

[arizonavideo]

I have used a good amount of the RM-16 ball screws and non have ever had tight spots and .004" loose spots. Something else is the problem.

Test the screw without the seals installed. They are good screws and almost all will let you run positive preload without too much drag with the right sized balls. It is a real pain to get that last .001" of slop out unless you have every size ball.

It can cost more to buy 4 sets of balls than to buy double nuts that will be almost backlash free every time. In the end re packing ball comes down to the difference of 2 or 3 hundreds of a thousands.

[rusmannx]

I've reloaded the screws with the standard .125" balls but haven't put the mill back together yet. If I can't get it sorted out this time (about the 4th time I've take it apart looking for the problem) I'm going to order some of the double nut screws. Plus Hoss just posted stage 3 so it may be time to look at some more upgrades anyway

Thanks guys.