[LongRat]

It is possible that you are running the motors below their rated torque - that is if they are rated for more than 2.5A each. I'm guessing they are 5V nominal voltage, but I would be VERY surprised if you are running them at 5V. Normally you'd run at about 20 times the rated voltage, to maintain torque at RPM. On 5V 2.5A you may have holding torque but all torque would vanish at any reasonable rotation speed.
For example, my motors are 3V rated, 6A. I run them at 6A on a 39V power supply. That's 13 times over voltage but not above the rated current. I still have plenty of headroom to go to a higher voltage supply, but if the holding torque is not sufficient I need to look at bigger motors.

[LongRat]

A warning about BK15 bearing blocks...

It's just another thing to look out for and maybe because they are cheaply made. But this isn't good.
I assembled my Z axis and noticed a large amount of play. It was immediately clear this was end-float in the bearing. The whole screw was sliding up and down in the angular contact bearings. On closer inspection, the bearings themselves were moving with the screw, in the block! It turns out that the end cap that retains the bearing races in the block was not machined deep enough. This meant the bearings could float by 0.2mm axially in the block. I didn't notice this when I was rebuilding the blocks because the bearings are a pretty close fit in the bore and didn't slide very easily.

I'm going to have to machine 0.3mm off the face of the block to allow the end plate to properly contact the bearings. I don't think my BK12s are like this but need to check anyway.

Chalky, something tells me the Zapp Automation blocks probably wouldn't have this problem.

[tef9]

Lol As I said, the electronics stuff makes my brain hurt...

Still its osmthing I should learn as Motion Control is a part time obsession, yet again, so is building the mill

Did you have to machine your ballscrews? If so how did you do it?

Andy

[digits]

A warning about BK15 bearing blocks...

It's just another thing to look out for and maybe because they are cheaply made. But this isn't good.
I assembled my Z axis and noticed a large amount of play. It was immediately clear this was end-float in the bearing. The whole screw was sliding up and down in the angular contact bearings. On closer inspection, the bearings themselves were moving with the screw, in the block! It turns out that the end cap that retains the bearing races in the block was not machined deep enough. This meant the bearings could float by 0.2mm axially in the block. I didn't notice this when I was rebuilding the blocks because the bearings are a pretty close fit in the bore and didn't slide very easily.

I'm going to have to machine 0.3mm off the face of the block to allow the end plate to properly contact the bearings. I don't think my BK12s are like this but need to check anyway.

Chalky, something tells me the Zapp Automation blocks probably wouldn't have this problem.

Hmm, that's really not good my BK12's are solid as a rock! IMHO sometimes it does pay to shop locally even if it hurts the wallet at the time - being able to send faulty stuff back or get next day replacements does have value to me.

Hope you can sort it quickly and painlessly - I want to see that kitty roar!

[ihavenofish]

Chalky, something tells me the Zapp Automation blocks probably wouldn't have this problem.

i bet they would. i think they all come from the same source. in theory you could just return it for a new one, but i imagine shipping may cost more than the unit.

i dont know if you can machine the body on the x2 though, it is probably hardened steel no? you could probably use a washer instead.

[LongRat]

It isn't that hard, if my file serves me right. I can try it at least, if not I will make a spacer washer as you suggest. Looking at the pics on the Zapp site, the finish looks MUCH nicer than the blocks I have here. For the money I would hope they are better.

I ordered the screws to my specs including end machining. I have machined ballscrews before and succeeded, but it took a couple of carbide tips and wasn't worth the hassle.

[Hans_G]

My BK/BF 12 blocks weren't hardened. I fly cut one set w/ my sherline without a problem.

Longrat, I found your enclosure thread with the 70L/min air pump mentioned... do you still use the air pump with your X2? Will it move to this new mill? Considering doing the same for my mill. thanks.

[LongRat]

Hans,
yes I still use the same pump. Plan to use it with the new mill too. I might add another pump and line, to hit the tool from 2 sides.

[LongRat]

I skimmed 0.5mm off the face of the block with a 32mm indexable face mill. Now I can reassemble the whole thing all over again!
I'm almost certain I will have to do the same thing to the Y axis bearing block too.

[digits]

Hi LongRat - did you manage to get it all back together yet?

[LongRat]

These blocks have turned out to be a bit of a nightmare. After checking the Y it turned out to be exactly the same. But I still wasn't happy with the Z either, even after the skimming.
A couple of days of investigation have yielded the following information. The blocks were machined to the correct depth all along, but there should be a washer or shim separating the outer races of the two bearings in the bore. My understanding is that angular contact bearing pairs are made such that there's a small gap between preloaded races, so when you compress them together this gap is closed up and gives you your defined preload. I am certainly no expert though.
In my case the races to be preloaded were touching all the time, meaning that no external force could ever apply an axial load from inner to ball to outer race. Placing a turned washer, 1mm thick between bearings solved this issue, and it meant that the facing cut I made across the block was unnecessary. This leads me to believe that they should have been supplied with some sort of spacer washer. Or maybe they ripped off the design from elsewhere and neglected that part.
Another clue to the nature of these bearing blocks comes from the faint scored lines crossed where the mounting holes are. They have been marked out - the blocks are drilled manually!

[SCzEngrgGroup]

These blocks have turned out to be a bit of a nightmare. After checking the Y it turned out to be exactly the same. But I still wasn't happy with the Z either, even after the skimming.
A couple of days of investigation have yielded the following information. The blocks were machined to the correct depth all along, but there should be a washer or shim separating the outer races of the two bearings in the bore. My understanding is that angular contact bearing pairs are made such that there's a small gap between preloaded races, so when you compress them together this gap is closed up and gives you your defined preload. I am certainly no expert though.
In my case the races to be preloaded were touching all the time, meaning that no external force could ever apply an axial load from inner to ball to outer race. Placing a turned washer, 1mm thick between bearings solved this issue, and it meant that the facing cut I made across the block was unnecessary. This leads me to believe that they should have been supplied with some sort of spacer washer. Or maybe they ripped off the design from elsewhere and neglected that part.
Another clue to the nature of these bearing blocks comes from the faint scored lines crossed where the mounting holes are. They have been marked out - the blocks are drilled manually!

Take a look at the bearing inner races. You will see the inner race is wider on one side than the other. The narrower sides *must* be facing each other. When mounted this way, *no* spacers should be required, and there should be no backlash when everything is clamped up tight. If this is not the case, then you got some truly crappy bearings. A/C bearings are precision ground so that when mounted this way, with the outer races securely clamped together, and the inner races securely clamped together, correct pre-load will be applied automatically. But, if you mount the bearings in any other configuration, you will have either a great deal of backlash, or excessive preload.

Regards,
Ray L.

[LongRat]

That's exactly what I thought. These bearings have inner and outer races absolutely flush with each other, on both sides. Mounted back to back or front to front makes no difference, each way I have a lot of backlash.
Only by fitting a thin washer between the outer races of the 2 bearings was I able to get a good preload and no backlash. I have no doubt these are low quality bearings.

[digits]

Not good news about your bearings

So how bad are the blocks? Can you just drop in a pair of decent A/C's or do you think you'll have to make your own blocks or buy some complete ones from somewhere reputable?

[SCzEngrgGroup]

That's exactly what I thought. These bearings have inner and outer races absolutely flush with each other, on both sides. Mounted back to back or front to front makes no difference, each way I have a lot of backlash.
Only by fitting a thin washer between the outer races of the 2 bearings was I able to get a good preload and no backlash. I have no doubt these are low quality bearings.

Perhaps they're not even angular contact bearings? Sure doesn't sound like it. I'd go to www.vxb.com, and order new bearings. They won't cost you much, and I think you'll find they work much better, and you can toss the spacers.

Regards,
Ray L.

[ellik]

I experienced exactly same with the bearings I got with the ballscrew for my X2 from linearmotionbearings2008 on ebay. I ended putting a thin spacer between the bearings to get some preload on them. Then I also had an issue with the cap that holds the bearings in place. The pockets for the four small screws that tighten it down where so deep that there was hardly any material left to tighten down on. I fixed that with longer screws and shims.

Bottom line is I will not get these end supports and bearings for any of my future projects. I will however buy another ballscrew from them.

More here : http://www.cnczone.com/forums/showthread.php?t=74631

[H.O]

Hi,
In my experience there are three types of "standard" single row A/C bearings:

** General purpose or "basic design" (as SKF calls them) angular contact beraings.
** Angular contact bearing for universal pairing.
** Matched pair or sets of angular contact bearings.

For example:
A "standard" SKF 7200BEP is a general purpose or "basic design" bearing. These are not designed or meant to be mounted flush to each other as the faces are not ground to provide the correct preload. This does not make them crappy bearings they are just not designed to be mounted flush to each other while providing preload. However, mounting them back-to-back with a spacer between the outer races allows for adjustable preload by "pushing" the inner races together.

The same bearing but for universal matching may be called 7200BEGAP. Two of these bearings mounted flush to each other will provide a certain amount of preload, how much that is depends in the 4th letter in the bearing designation, in this case the A means a light preload. The actual numbers can be looked up on SKF's website of course.

Then we have the matched sets of angular contact bearings. These are basically the same as the universally matchable bearing but they have been paired up by the manufacturer to give a more precise amount of preload when mounted flush to each other. These are obviosuly not sold individually but as sets, see here for example.

Then of course there are tons of different "special purpose" A/C bearings specifically made for things like ballscrew supports (these normally have a different contact angle than "standard" A/C bearings) and of course for high speed spindles etc etc.

/Henrik.

[LongRat]

Wow, that's a good chunk of useful information. Thanks!

Ellik - sorry to hear you had the same experience, but I am almost happy that somebody else has verified my analysis of the problem, on exactly the same parts. I'm not sure I wouldn't buy these again, now that I've gone through the hassle to fix them. Not sure.

Digits - I have fixed the issue with the 1mm shims between bearings. No need to replace bearings or remake the block, fortunately. The mill is now complete, just got to add my axis motors and all the other little things, like limit switches, lighting, etc etc etc!

Ray, H.O. - Yes, these are definitely AC bearings - they also have the letters AC (along with a load of other numbers I forget) laser engraved onto the faces. Thanks for the info, it makes total sense that these bearings are probably not designed to be run paired up touching each other.

Remember how I had to skim the faces of the blocks to get the end cap to correctly retain the outer races? If I had made the spacer shim first, I wouldn't have needed to skim the blocks. One shim per block would have fixed all these issues!

[digits]

I'm glad you've got a work around LongRat - it's just seems rather shocking that those block aren't correct right out of the box.

Good luck with the final assembly - I'm really looking forward to hearing how this thing cuts it as a CNC!

[LongRat]

Yeah, if I'd paid a lot of money for them I would be annoyed. As it is, I put it down as a learning experience.
Another interesting point to note is that with these small blocks, the saddle can now run significantly off the front of the base casting. I think we are talking about a realistic Y travel of at least 250mm, still with 90% or so of gib-dovetail engagement. That gives this machine a seriously huge cutting envelope!