Need Help!- Is my X2 spindle fixable?

[sansbury]

Hi All:

I need some help trying to diagnose a problem with my X2 spindle. Here's the story:

I bought an X2 in very lightly used condition off eBay and after less than an hour of running decided to put a belt drive on it. Everything went OK until I noticed that the spindle would heat up very quickly when running above 1000RPM or so.

I fiddled with the belt tension a bit without any change. The heating was definitely coming from the bearings.

Since then I have pulled and replaced the bearings with Nachi rubber-sealed bearings following the Little Machine Shop instructions for doing so (see here). The amount of force needed to budge anything felt pretty heavy to me and if the original bearings were not damaged, I suspect they were getting them out.

After pressing on the new bearings (which also took an enormous amount of force, or what felt like it to me--it was a stick-and-jerk process, not a smooth one--the problem seemed worse in that the spindle ran even hotter faster, though it did sound a little smoother.

Afterwards I pulled everything apart again and polished the spindle journals on my lathe with some fine sandpaper. This made the bearings press on a *little* smoother though still under substantial pressure. I reassmbled everything but no improvement seen.

One thing I am wondering about is whether there is any problem besides the bearing installation themselves that could cause this.

Right now I am kind of at a dead end and wondering if my only good choice is to buy a whole new spindle assembly. I suspect that the process of installing and removing the bearings several times was not kind to them. There are several steps that as best as I could tell had you pushing on the bearings the "wrong" way.

I've measured the bearings and spindle journals and they seem to be right on top of each other size-wise. Measuring with my caliper I get readings of between 35.02 and 34.99 mm on the lower section for instance, mostly right at 35.00. The bearing ID measures at 35.00 or less though to be honest I have a hard time reading consistently. I could polish this down farther but not sure how far I should go.

I can buy another pair of bearings for $20 but right now I have zero confidence it will turn out any better than the first time. Either there is something defective in the spindle assembly or in my technique but it is not clear how I would determine this.

The only other thread I saw about this problem seemed to be resolved with a bearing replacement without anything special. So I am escalating it here in hopes that something new is revealed or confirming that replacing the whole spindle head assembly is the wisest choice. Right now I'm down $30 or so of parts and two days of what was supposed to be productive time and I'm loathe to throw more at it without something new....

THANKS IN ADVANCE!

[John S.]

Common fault on these machines caused by the makers grinding both bearing journals the same.
The lower one can be tight and it's as well that it is, tight that is not super tight.

The upper one need to be a sliding fit, i.e. the bearing needs to be able to be tapped down the shaft but not fall down on it's own.

The reason for this is that when you assemble the spindle and fit the top bearing you press the bearing home with the adjusting nuts to get the preload but when you back the nuts off the bearing is that tight it stays where it is with too much preload on it.

One word of warning when polishing the spindle to get the correct fit. The spindle will heat up and expand when polishing, allow to cool before trying the bearing or you may get the bearing to fit nicely when warm and it's too loose when cool.

Not knowing what equipment or skills you have it's also possible to improve on these spindles in that you have two C spanner type nuts to adjust the bearings, one sets the preload and the second locks the first.
It's very hard to lock the second without moving the first. If you have the skills and equipment think about making one nut as deep as the two.
Then drill into the side, tap for a grub screw and fit a small brass pad to the bottom before fitting the screw so it doesn't mar the threads.

This way you have only one nut to worry about and tightening the grub screw will not affect the preset preload.

John S.

[fatal-exception]

It will eventually cool down after hours of running. As long as the spindle runs smooth, and isnt in danger of catching on fire, I wouldn't worry about it.

I have 3 X2 spindles, and they all have their own quirks. One came with damaged bearings right out of the box. I agree that the press fit on the bearings is a little excessive. I pressed them in on a hydraulic jig with loadcells and measured the press force at 800 lbs! Oh well, they will never fall out.

Paul

[sansbury]

fatal--thanks for the data points. FWIW, when I say "hot," I mean after 3-4 minutes at top speed it was almost too hot to touch. If it kept going I'd be afraid it would carry a lot of heat to the belt--those aluminum pulleys make great heat sinks!

Well, the game is up for now anyway I think. I managed to polish the spindle to exactly 0.01mm too small, giving a perfect slip fit.

You know if I *tried* to do that I couldn't possibly hit it so well. I'm almost tempted to put it all back together just for giggles and see what happens. I figured the risk of exactly this happening was pretty high but since I was looking at the full replacement anyway, I figured it was worth a shot. Oh well, thanks for playing I guess!

[Stepper Monkey]

Even if the journals are ground to the right size for the bearings, whether they are aligned perfectly is another issue as well. This can also cause excessive heat and binding problems from the uneven loading on the bearings. They have to be perfectly concentric to each other as well as sit exactly parallel. Any minor tilt is going to load the bearings unevenly and cause a crapload of heat. From what we have seen of Chinese QC it would be prudent not to assume one or both of those things to always be properly trued on every machine as shipped, so you might want to check that as well before you throw in the towel!

BTW, if the spindle is only .01mm undersize, there are several easy ways to 'grow' the shaft back up larger again for that small amount. So don't give up on it quite yet either!

[sansbury]

Parallelism and concentricity were on my mind. Any suggestions on how to check for those? That said, it would seem like those would be even harder problems to fix than the bearings.

Oh, and about putting .01mm of metal back on, what would you suggest? I know about knurling but was doubtful that would work here. I also thought about something like a little loctite but that seemed to be asking for trouble. My hope is that a replacement assembly from LMS will, at worst, come with an exchange warranty.

Also one poster on a Yahoo board suggested making sure the lower edge of the pulley wasn't fouling the bearing shield--he said this happened to him and it just needed a little relieving of the pulley and all was well again. I'm well past that now (and I doubt that was the problem) but figured I'd post it here for the benefit of future readers.

[Speedygonzales]

im not sure what part you need to grow back, but doesnt heating it up red hot help?

[Stepper Monkey]

Originally Posted by sansbury Oh, and about putting .01mm of metal back on, what would you suggest? I know about knurling but was doubtful that would work here. I also thought about something like a little loctite but that seemed to be asking for trouble. My hope is that a replacement assembly from LMS will, at worst, come with an exchange warranty.

Knurling is certainly one of the options, a time-tested old way to do it. Depending on the type of knurl used, once it is turned back down to normal diameter after knurling, you can often get back to the vast majority of the surface area of the shaft at the right diameter. It often looks just like a bit of cross-hatching that doesn't lose more than a small percentage of the original surface area, and in this case it isn't necessary to have all of it there anyway. It should work just fine for this.
Another simple way to do it is to electroform. I have seen worn parts brought back up to spec by plating even up to 5 thou or so of Nickle onto the parts and resurfacing. On a round shaft that growth would of course be doubled as it would be applied evenly to all sides. In fact, when hard nickle plating new gun parts the dimensional growth can actually keep them from fitting properly. It is cheap, quick, and easily done, but then again I have that equipment around, so its easy for me to suggest. I don't know how easy it is for you to find a local shop for that if you don't know someone already with the right gear.

The part is probably too cheaply replaced for these other options to make sense, but in restoring old cars and vintage aircraft with irreplaceable parts there are other options I use too. Quick and dirty, you can turn the shaft down much smaller, swage a bushing on, and then turn it to correct size. Just make sure there is enough wall thickness for the bushing not to be too thin.
If the shaft is hollow or you can't get at the ends of it you can't do that, so you weld a spiral fillet of metal winding around it (like a screw thread) to build the surface back up, then turn it back down and re-heat treat it. That is the way we redo old crankshaft bearing journals.
You can fix anything with enough time, though in this case maybe not worth the effort, but its nice to know you can always 'grow' a part back up.

[bilinghm]

If you have the spindle torn down again, why not mount it between centers and measure the total run out. Both bearing journals can be measured in relationship to each other. Mark the highest spot on the journals. If they are high on different sides, but not too far out, perhaps you can take it to an auto machine shop for regrinding. Once reground, you can knurl, or plate your way back into spec. Actually you can add a few thou with DIY electroless nickel plating in your own shop. I do electroless nickel frequently. It is cheap and extremely easy to do.

[sergizmo]

A new spindle assembly is $100. Honestly, is hours of work and ordering a few dollars (or more) worth of parts several times better than dropping another $100? To me (and most people), no.

[sansbury]

Guys,

Thanks for the ideas and help. As Stepper Monkey alluded to, $100 for a full semi-guaranteed replacement is going to be the cheapest option here if I place any value on my time as well as the downtime which is preventing me from moving on to the project which got this whole ball rolling in the first place. Order placed this morning and with any luck in hand by the weekend or so.

Still, I appreciate the suggestions as they may someday come in handy when a replacement is less economically feasible. The electroless nickel plating is a new technique for me and may come in handy for other purposes. Somewhere down the road I might try to make this spindle work again for the heck of it--somebody suggested building a small horizontal mill around it which I thought might be neat as I do have a spare X-Y table around. Besides, there's at least three square feet of basement space yet to be filled with junk, er, I mean future projects....

[sergizmo]

Good stuff. Be sure to update with the quality of the new spindle. Hopefully you aren't unlucky enough to get another lemon. If the new one is bad, LMS may be able to exchange it for you.