Keep losing my balls.

[thenewguy]

This is about the third or fourth time I had to reapply jb weld to the ball ends. I ran a bunch of prints just fine. Then I changed a few things during a upgrade period.

It could be the heat in the enclosure since I started using the glass window.

I tightened my belts with tensioners.

I re adjusted the pillars.

Installed dampeners

Ever since I did these things I can't print for more than 4 hours without my balls breaking. Since I did it all at the same time it's hard to tell the source.

Of course this happened right after I fell asleep.

Also the red filament I bought from see me cnc was 1.58
How weird is that? I haven't checked my other colors yet since my big filament purchase.

[Cleveralias]

Mine have all held up just fine. Maybe something in your preparation? I can't see how any of the changes you mentioned would be a cause... The only one acting on the balls would be your enclosure, but jb weld is good to 550F so it seems unlikely this would cause a problem.

You definitely want to scuff the balls where the screws are going with some sandpaper so the jb weld has something to grab. When mixing the jb weld, once I think it's mixed I keep mixing for another full minute. Clean the balls and screws well with alcohol or acetone. Be sure to let them cure the full recommended time before putting them in use (I think 18 or 24 hours).

Hope that helps.

[lordbinky]

I'm going to go ahead and reveal I've gone against the common knowledge here because I was having the same issue. I used Locktite Super glue - professional. In my experience with it, it has held better than the epoxies I've mixed, and I use the part within minutes. Same prepping (since I had already done all the cleaning and scouring for the epoxy). If I didn't use what I already had, I would have considered using their ultra gel super glue.

[gestalt73]

Heh, I've had the same problem for the first week after converting to magarms.

Here are some of the things that I've done that have resulted in better ball retention:

• quick rinse and shake of both the bolts and balls in acetone. specifically the bolts are covered in a machine oil.
• take a file and scuff up the top of the bolt a bit
• get some overlap when you jbweld the bolt to the ball, don't just dab the top of the bolt and expect it to stick
• when you do this, jbweld some extras, "just in case"
• let the jbweld sit for at least an hour. I know it says 6 minutes, but the longer you wait the better the bond

I can upload some pics if it would help, but I haven't had a ball drop since I followed all the steps above.

[Cleveralias]

[*]let the jbweld sit for at least an hour. I know it says 6 minutes, but the longer you wait the better the bond[/list]

That's jb kwik. It's faster but not as tough as jb weld. I wonder if others experiencing failures are using this instead of jb weld.

[lordbinky]

I used the highest hold strength full blown cures slower than pain drying two part tube of awesome and it was popping. *shrug* I don't argue with superglue working for me.

[Zerosink]

I don't know so I'm a askin

Why are you "bonding" the balls to screws?
You use a magnet on the arm... why not on the effector.
The ball would be between the 2 magnets and... the "bond" strength
of 2 magnets with it's poles in line would be stronger upon the ball.
==( )== ?

I'll go now

[Batteau62]

I think your right That's genius I haven't done my mag arm mod yet. I think I'll play around with your idea. As long as the ball is supported with cups on both sides, it should be a zero lash joint. Might almost be better to go up to 1/2" balls for more range of motion with the arms. Good one Zerosink Off to the test bench

[Flateric]

I MIG welded mine to the tops of the cap screws. Mine have never come off.

[Eric]

I don't know so I'm a askin

Why are you "bonding" the balls to screws?
You use a magnet on the arm... why not on the effector.
The ball would be between the 2 magnets and... the "bond" strength
of 2 magnets with it's poles in line would be stronger upon the ball.
==( )== ?

I'll go now

You should be able to do basic tests just by using two of the arms (hand-held) and a spare ball bearing. My concern would be how it behaves when the arms are NOT in line with each other.

[thenewguy]

I did use the jb weld quick which is only rated to like 250f. I did everything in the loctite gel super glue this time. Tried snapping the balls in my hand. I think we got a winner!

[Eaglezsoar]

I did use the jb weld quick which is only rated to like 250f. I did everything in the loctite gel super glue this time. Tried snapping the balls in my hand. I think we got a winner!

How long did you leave everything sit before trying to snap the balls?

[thenewguy]

I did use the jb weld quick which is only rated to like 250f. I did everything in the loctite gel super glue this time. Tried snapping the balls in my hand. I think we got a winner!

How long did you leave everything sit before trying to snap the balls?

around 8 hours

[edward]

I MIG welded mine to the tops of the cap screws. Mine have never come off.

What kind of setup and settings did you use? Wire size? Feed rate? Gas or flux core?

The question of MIG welding them was posed to me earlier and I couldn't really answer as I TiG'd mine.

[barry99705]

I MIG welded mine to the tops of the cap screws. Mine have never come off.

What kind of setup and settings did you use? Wire size? Feed rate? Gas or flux core?

The question of MIG welding them was posed to me earlier and I couldn't really answer as I TiG'd mine.

As long as you don't vaporize the screws, just a couple tacks should work. It's not like they're holding any weight.

[edward]

I MIG welded mine to the tops of the cap screws. Mine have never come off.

What kind of setup and settings did you use? Wire size? Feed rate? Gas or flux core?

The question of MIG welding them was posed to me earlier and I couldn't really answer as I TiG'd mine.

As long as you don't vaporize the screws, just a couple tacks should work. It's not like they're holding any weight.

Uh....thanks?

This doesn't really answer any of my question. I would have guessed that keeping the screws in some sort of physical manifestation of their original selves would be a given.

I was asking for informational/educational purposes.

[Flateric]

My trick to getting them all seated and consistent was to use a large magnet or a magnetic grounding attachment for the welder. Place this in a vise, let the cap bolt magnetize to the magnet and then let the ball magnetize to the cap bolt.

Then with extremely low feed and at the lowest setting on my welder a selectable 4 power levels a tiny and quick tack was all that was needed. Any more sand the tiny cap bolt will warp or burn off. Does not need strength so much as just being held in position.

Flux based, being messier and not as precise would be more difficult but not impossible.

For one of my effectors I currently use a hd platter from a hard disk that has small holes drilled for each ball to sit into and then I use jeweled stick from the bottom through the hole. I rub he uncared but mixed stick over the surface of one side of the ball (slightly roughened with 400 grit sandpaper) until I feel the jeweled stick I am rubbing against it start to get tacky and adhering then place the ball and squish the stick through the hole leaving a small amount on the opposite side of the hole to act as a lip as well. This is a zero chance for failure mount technic if you do not want to or have the means to weld.

The disc of the hard drive is the perfect diameter for a effector platform and has a decent sized hole perfectly centered for the hotend. It is also stiff and light in weight and can be easily drilled for the standard hole pattern for mounting options.

It also will not melt or deform from the heat of the hotend. And best of all can be had for free since everyone can get there hands on an old or bad hard drive to disassemble.

[cambo3d]

In my prior testing, I could not get the balls to adhere enough. They kept coming off even with careful prep so I never installed this modification to my rostock.

Welding is the probably the best most reliable way of doing it. Unless you can find magnetic balls with studs already screwed to them.

[barry99705]

My trick to getting them all seated and consistent was to use a large magnet or a magnetic grounding attachment for the welder. Place this in a vise, let the cap bolt magnetize to the magnet and then let the ball magnetize to the cap bolt.

Then with extremely low feed and at the lowest setting on my welder a selectable 4 power levels a tiny and quick tack was all that was needed. Any more sand the tiny cap bolt will warp or burn off. Does not need strength so much as just being held in position.

Flux based, being messier and not as precise would be more difficult but not impossible.

For one of my effectors I currently use a hd platter from a hard disk that has small holes drilled for each ball to sit into and then I use jeweled stick from the bottom through the hole. I rub he uncared but mixed stick over the surface of one side of the ball (slightly roughened with 400 grit sandpaper) until I feel the jeweled stick I am rubbing against it start to get tacky and adhering then place the ball and squish the stick through the hole leaving a small amount on the opposite side of the hole to act as a lip as well. This is a zero chance for failure mount technic if you do not want to or have the means to weld.

The disc of the hard drive is the perfect diameter for a effector platform and has a decent sized hole perfectly centered for the hotend. It is also stiff and light in weight and can be easily drilled for the standard hole pattern for mounting options.

It also will not melt or deform from the heat of the hotend. And best of all can be had for free since everyone can get there hands on an old or bad hard drive to disassemble.

Careful with using hard drive platters. A lot of the newer ones are glass. I've got one around here somewhere that had a head crash and it wore the magnetic gold colored layer off the center.
Found it!
[img]https://lh4.googleusercontent.com/-T0zo ... 212512.jpg[/img]

[lordbinky]

As long as they keep using ferrofluid they junk hard drives will always be awesome.

[Flateric]

I had no clue some of them were glass now. Neat.

Do you know the make and or brand that use glass platters like that?

Also, nice keyboard, what type is that.

[Eric]

Someone who is proficient at welding could probably make a few bucks selling sets of balls already welded to bolts. Or in the case of a machinist with the right equipment, balls already drilled and tapped for bolts.

[Flateric]

If someone has found a fast easy and cheap way of drilling and tapping steel ball bearings, PLEASE speak up and tel me how you are doing this. Because I cannot see this being, fast, easy, cheap that I can think of.

[barry99705]

I had no clue some of them were glass now. Neat.

Do you know the make and or brand that use glass platters like that?

Also, nice keyboard, what type is that.

It's a cheapo keyboard that came with my wife's office's Alienware desktops. I find it funny that she uses gaming desktops for her bookkeeping business. They're the small formfactor models. As for which brand? I think this came out of a seagate, but I'm not 100% sure. Easiest way to tell, other than trying to break them is use one of those obscenely bright led flashlights under the platter. Might need a dark room to see it, but you will be able to see the led through the platter if it's glass.

[edward]

Thanks for the response, flateric. I'm always trying to learn new tricks and ideas in order to expand my welding abilities. I worked in the fab shop at my university machine shop and the range of projects we received was great for pushing and expanding my abilities.