[cadguy247]

Does anyone know the correct way to stack and preload using both a Double Row Angular Contact Bearing and a Thrust Bearing together? This is how the Z axis is done in this kit, which i think is the BD Tools Kit, but i can't find any info on howthey should be designed and how to preload them. This is what i've used on my mill but it seems very fussy to get them right and in the end, i feel like they will loosen up over time

Thanks
Kyle G0704 Mill CNC Conversion Kit Mounts with Ballnut Mounts, B-G0704M | Automation Technology Inc

[Fastest1]

Do you have a set of locking nuts or the nut with the retaining set screw provided by Linearmotion?

[cadguy247]

i'm using 2 jam nuts as i bought the ballscrews from automation technologies and thats what they came with. I'm nut actually using the mount kit, i've made my own, so i'm not even sure if i've designed them in correctly. I'm only getting .0015 backlash in the Z axis, but i just don't think i've done it correctly where it will hold up over time

[Fastest1]

.0015 is a pretty decent outcome for a single nutted ballscrew. This will change with time and will require some adjustment. Actually .0015 is good regardless. Mach 3's backlash comp will deal with that miniscule amount very well. I suggest just warming the machine up with the wizards and get some travel thru all axis. Kind of a shake down time. The warm up wizards will allow you to choose how ever many cycles you desire, choice of single axis or multiple axis movement simultaneously and distance to travel in each axis. Dont forget to lube all of the ways and screws, even if manually. Then maybe review all of your fasteners and adjust accordingly.

Throw up some pictures. You can take them with your phone. The rest of us do.

[cadguy247]

I'm pretty sure that the .0015 backlash is a fluke and I'm almost positive that it will go away in time. My real issue is whether or not I have the bearings stacked correctly. Here is how I have them set up. If I tighten too much they bind, if too loose then huge backlash. Gotta tighten, then back off a hair and it seems to work but I don't think I have them stacked correctly

[Fastest1]

I believe mine are just the opposite. I dont see a difference either way. There could be I just wouldnt know why. Find the sweet spot is a tedious process. Fortunately once accomplished you shoulldnt need to do it often.

[pippin88]

Do the locknuts bear only on the inner race of the double row AC bearings? If they don't they might be rubbing on the seal (I'm guessing it's sealed).

It's not a bad arrangement to avoid having to make a bearing pocket precisely to depth. I hate making bearing blocks. Have to make a couple today...

[cadguy247]

Do the locknuts bear only on the inner race of the double row AC bearings? If they don't they might be rubbing on the seal (I'm guessing it's sealed).

It's not a bad arrangement to avoid having to make a bearing pocket precisely to depth. I hate making bearing blocks. Have to make a couple today...

the lock nuts are only bearing on the inner race. The Double Row Bearings are actually pretty nice, no preload to worry about if used by themselves. My X & Y are just the Double Rows and no thrust bearing so those are easy to set. But i'm also getting .002 & .0025 in those axis. But my Y ballscrew is actually long enough to add a thrust bearing so i may do that sometime soon. Seems to me that the Chinese ballscrews just arent as tight as the Thompson ones that i got from CNC Fusion for my SX2, i have less than .001 on all 3 axis. and that kit uses stock thrust bearings in the X , 2 single row contact bearing in the Z and i believe a Double Row Angular Contact bearing in the Y

It seems like i have them stacked right and that was my biggest worry. I have about 15 hours of machine time on it so far so i'll just keep an eye on things and see how they go..

Thanks

[pippin88]

Yeah, I've moved to double row AC bearings for all my CNC machines.
Maybe they are not quite as good as properly ground or shimmed single rows, but I'm not paying for properly ground single row pairs, and I've thus far failed to find a good source of shim stock, and to make shims well, so double rows work for me.

Why bother with adding a thrust bearing to your Y?

Doesn't really matter how you stack them. It looks to me like the thrust bearing in that design is just there so you can have two bearings tightened against each other which pinches the mounting plate. This means you don't have to make a pocket / journal a precise depth and don't have to have a cover plate to retain the bearing.

[cadguy247]

Attachment 256676

the reason i added the thrust bearing to the Z was because i was orignally going to buy the BD Tools mounts and that's the way his is set up, so i figured i'd give it a try. Reason for adding a thrust bearing to the Y is because i have the extra room on the ballscrew and figured if my Z has the best backlash out of them, so i thought maybe the added thrust bearing was the reason. can't hurt i can always take it out.

As far as bearing pockets that usually doesn't concern me too much as i just make the pockets a little deeper, then measure and then face off until the pocket is about .001-.002 below the bearing outer race and then trap it with 4 flange SHCS like i've done here.

please note that most of my mounts look odd because i made them before i received my mill based off the old PM-25/G0704 design and when i recevied the mill, found out it had changed and now looks to be that Titan made it, so i either threw some mounts away or butchered others to get the build going. Once i have everything worked out i will make new ones

[Fastest1]

The preload should hold the bearing in. There isnt a need for the SHCP's. Why would the outer race on a bearing spin? Maybe if it had failed, pretty unlikely with these rpm's and loads.

[cadguy247]

The preload should hold the bearing in. There isnt a need for the SHCP's. Why would the outer race on a bearing spin? Maybe if it had failed, pretty unlikely with these rpm's and loads.

as a safety to keep the bearing from pushing out if I had some sort of crash.

[pippin88]

The preload should hold the bearing in. There isnt a need for the SHCP's. Why would the outer race on a bearing spin? Maybe if it had failed, pretty unlikely with these rpm's and loads.

I don't think he is using the screws with a double bearing where the bearings are constrained against each other plus the plate (thrust plus double row Ac) but rather a single bearing. If you have a single bearing you must have something to constrain it axially - the bottom of the pocket on one end, and a cover or in this case bolt heads on the other.

[cadguy247]

I don't think he is using the screws with a double bearing where the bearings are constrained against each other plus the plate (thrust plus double row Ac) but rather a single bearing. If you have a single bearing you must have something to constrain it axially - the bottom of the pocket on one end, and a cover or in this case bolt heads on the other.

correct, even though I make my bearing pockets nice and tight, need to freeze the bearing in order to tap it in, without the SHCS they are only constrained in one direction

[Fastest1]

I understand, I was picturing opposing sets.