My little 7x14 is now ready to be cleaned, fettled and the rest

Do you have any general advice about how best to get a preload that'll be in the right general area?

*I suspect it's this that's making me a bit averse to installing the new bearings and so the idea of 'putting off' this job feels more attractive but really, in addition to the fact that avoiding this feels like I'm cutting corners a bit (something I endeavour to avoid in my professional life), I do enjoy this kinds of thing.


First let me say that hand setting bearings is not that bad. When I type it out it's gonna sound long and drawn out, and it can be. Especially maybe the first time. It is a little bit fiddly, so you can expect it to take some iterations, and some time, but it's very simple steps, very approachable, and a lot less daunting if you know what the end goal is. Don't let this throw you off. In an application like this, they need to be good, plenty good even, but you don't need anywhere near theoretical perfection. The bearings won't ever see any load that's going to challenge them. And I'm going to throw some numbers at you, and while this is a "feel" thing in the end, these (rough) numbers give you some reasonable targets as to what's enough, too much, or too little during process.

If you're able, if the bearing manufacturer has a website, they probably will have some advice on how much preload those bearings can take. That might be useful. In general though, it's based on an installation that has zero flex, and the installer/builder has to add that into the calculation. In practice, if a bearing that size is preloaded to half a thousandth of an inch, maybe to one thousandth of an inch, it'll be in a very happy place. But I don't know at all how much flex is going to be in that headstock casting. Nor do you probably. But since I'm gessing you don't have or want the tools to measure that in order to throw it into any calculations..... Well, that's probably why the manufacturer stuck radial bearings in there in the first place......

So to set that up by "feel", what one might do is something like this- Put it together, put the nut that tightens the whole thing on, and while manually rotating the spindle, snug the thing up until you can just feel the bearings start to have some drag. You don't even need to quantify it, just barely perceptable is plenty. Then run the spindle first by hand, five or ten turns, or whatever... feeling for absolute smoothnesss. It should be. Then snug it down right tight, a little at a time, turn, tighten, turn, tighten, until the spindle offers significant resistance. Maybe half a turn if you spin the chuck and let go. It should still be smooth, even though you are (on purpose) making it way too tight. What you've done here is prove the bearings aren't crunchy, seated the bearings properly, just in case the installation was not as perfect as it could have been, and basically just verified everything about the installaton is just right, prior to adjusting.

Now, back off that spindle nut, and bring it down to just snug. Hand tight maybe, but all you can give it. (Yes, that's broad range, but that's OK, it's quite close enough, it's just a starting point.) Then, under power, run the lathe for a few minutes, feeling the headstock as near as possible to each bearing. If it starts getting hot, you're too tight. If it doesn't heat up, you havn't tightened it enough.

Generally, for light applications like this, the IDEAL place to be is right at or just below the point where you can JUST BARELY percieve a temperature change after 20 or 30 minutes of run time, If it warms up tangibly, back it up about a thousandth of an inch. If you havn't felt any temperature change, tighten it about a thousandth until you do just barely percieve a temperature change after that time. Personally, i would be tempted to run these SLIGHTLY on the loose side, which means I would find that "just barely perceptable heat" point after 20 minutes of run time (at a middleish spindle speed), and back the nut up one thousandth of an inch, and call it good.

How do you measure a thousandth of an inch of preload in such an application? With enough metrology equipment that it'll cost more than your lathe... So you don't. That spindle nut and spindle have some thread pitch. I have no idea what it is, but I'm gonna throw out 20 threads per inch. (You're gonna have to convert anyhow, why not.....) So at 20 threads per inch, the nut advances fifty thousandths per revolution. 25 thousandths per half a revolution. I think you're going to have one of those four slot spanner nuts... That makes it easy. 12 thousandths per slot. Call it 10, and envision divisions on there. Call it ten even, for easy estimation. That's close enough. That's about the coarseness of the adjustments you want to make in that case, about one tenth of one quarter of the nut. Draw marks on it if it helps.

This will NOT make high speed spindle bearings for CNC equipment, but if you get it right at that spot just between not making heat and barely warming up at all... you'll be good.

And let me assure you, this is how I personally would set up a bearing. Other people have other methods, and that's fine. So long as you get it in a happy place at the end, it'll all work out just fine. When you're not maximizing the load of a bearing (you will not be) the window for "good" becomes (relatively speaking) a pretty big margin of tolerance. Give it some time and effort (mostly time for those 20 minute run time windows...) and you will get this just fine, even on the first bearings you ever set up.
 
First let me say that hand setting bearings is not that bad. When I type it out it's gonna sound long and drawn out, and it can be. Especially maybe the first time. It is a little bit fiddly, so you can expect it to take some iterations, and some time, but it's very simple steps, very approachable, and a lot less daunting if you know what the end goal is. Don't let this throw you off. In an application like this, they need to be good, plenty good even, but you don't need anywhere near theoretical perfection. The bearings won't ever see any load that's going to challenge them. And I'm going to throw some numbers at you, and while this is a "feel" thing in the end, these (rough) numbers give you some reasonable targets as to what's enough, too much, or too little during process.

If you're able, if the bearing manufacturer has a website, they probably will have some advice on how much preload those bearings can take. That might be useful. In general though, it's based on an installation that has zero flex, and the installer/builder has to add that into the calculation. In practice, if a bearing that size is preloaded to half a thousandth of an inch, maybe to one thousandth of an inch, it'll be in a very happy place. But I don't know at all how much flex is going to be in that headstock casting. Nor do you probably. But since I'm gessing you don't have or want the tools to measure that in order to throw it into any calculations..... Well, that's probably why the manufacturer stuck radial bearings in there in the first place......

So to set that up by "feel", what one might do is something like this- Put it together, put the nut that tightens the whole thing on, and while manually rotating the spindle, snug the thing up until you can just feel the bearings start to have some drag. You don't even need to quantify it, just barely perceptable is plenty. Then run the spindle first by hand, five or ten turns, or whatever... feeling for absolute smoothnesss. It should be. Then snug it down right tight, a little at a time, turn, tighten, turn, tighten, until the spindle offers significant resistance. Maybe half a turn if you spin the chuck and let go. It should still be smooth, even though you are (on purpose) making it way too tight. What you've done here is prove the bearings aren't crunchy, seated the bearings properly, just in case the installation was not as perfect as it could have been, and basically just verified everything about the installaton is just right, prior to adjusting.

Now, back off that spindle nut, and bring it down to just snug. Hand tight maybe, but all you can give it. (Yes, that's broad range, but that's OK, it's quite close enough, it's just a starting point.) Then, under power, run the lathe for a few minutes, feeling the headstock as near as possible to each bearing. If it starts getting hot, you're too tight. If it doesn't heat up, you havn't tightened it enough.

Generally, for light applications like this, the IDEAL place to be is right at or just below the point where you can JUST BARELY percieve a temperature change after 20 or 30 minutes of run time, If it warms up tangibly, back it up about a thousandth of an inch. If you havn't felt any temperature change, tighten it about a thousandth until you do just barely percieve a temperature change after that time. Personally, i would be tempted to run these SLIGHTLY on the loose side, which means I would find that "just barely perceptable heat" point after 20 minutes of run time (at a middleish spindle speed), and back the nut up one thousandth of an inch, and call it good.

How do you measure a thousandth of an inch of preload in such an application? With enough metrology equipment that it'll cost more than your lathe... So you don't. That spindle nut and spindle have some thread pitch. I have no idea what it is, but I'm gonna throw out 20 threads per inch. (You're gonna have to convert anyhow, why not.....) So at 20 threads per inch, the nut advances fifty thousandths per revolution. 25 thousandths per half a revolution. I think you're going to have one of those four slot spanner nuts... That makes it easy. 12 thousandths per slot. Call it 10, and envision divisions on there. Call it ten even, for easy estimation. That's close enough. That's about the coarseness of the adjustments you want to make in that case, about one tenth of one quarter of the nut. Draw marks on it if it helps.

This will NOT make high speed spindle bearings for CNC equipment, but if you get it right at that spot just between not making heat and barely warming up at all... you'll be good.

And let me assure you, this is how I personally would set up a bearing. Other people have other methods, and that's fine. So long as you get it in a happy place at the end, it'll all work out just fine. When you're not maximizing the load of a bearing (you will not be) the window for "good" becomes (relatively speaking) a pretty big margin of tolerance. Give it some time and effort (mostly time for those 20 minute run time windows...) and you will get this just fine, even on the first bearings you ever set up.
Now, that's just what I needed, thank you. :)
 
If you have a 2 speed headstock, consider installing metal gears while you have the spindle apart .


You will break these at some point in time and replacing them requires spindle removal. I still use a plastic gear on the intermediate shaft for a few reasons. It keeps the noise down, replacement is quick and easy and the gear acts like a fusible link. It's easy to dig a tool in on these machines, something is going to break and it may as well be a cheap part.
 
If you have a 2 speed headstock
I have the version with a brushless '750W' motor (how close 1 HP the output is, remains to be seen ;) ), so no hi-low selector is present on mine.

I may well get a set of metal change gears, if I see any significant wear or cracking. All the plastic (acetal?) gears seems to currently be in good order though.
 
Yeah, I started looking at bearing pullers on Amazon (UK) but I got confused and gave up. Do you have any recommendations?

But yes, I think you're right; given the apparent lack of wiggle, I guess I might leave it for now.
I bought a hydraulic puller from Harbor Freight....
 
I have the version with a brushless '750W' motor (how close 1 HP the output is, remains to be seen ;) ), so no hi-low selector is present on mine.

I may well get a set of metal change gears, if I see any significant wear or cracking. All the plastic (acetal?) gears seems to currently be in good order though.
Little machine shop carries all kinds of parts for those including change gears....
 
I have the version with a brushless '750W' motor (how close 1 HP the output is, remains to be seen ;) ), so no hi-low selector is present on mine.

I may well get a set of metal change gears, if I see any significant wear or cracking. All the plastic (acetal?) gears seems to currently be in good order though.
None of my plastic change gears are showing any signs of wearing out. They are so old that the lathe has a 3 digit serial number (184). I'm the 3rd owner, my buddy got it in 98 from the guy that bought it new.
 
At least the forum is more civil than the spectating the Houses of Parliament!

Here's one, but I didn't check the .co.uk site. It installs and removes. Check to see if these tools come with the sizes you need first.


Preload is a big deal, because things warm up in operation, and the tighter they are, the more it adds to heat. The bearing manufacturer may specify a preload torque. In those cases, you can a "fish scale" or hanging spring scale (not digital) to pull from a known radius to set preload. You can also do it with a temperature gun in some cases.. Bearing manufacturers provide info too. And then there's always the wheel bearing technique... There are people here who definitely know more than I do about bearings, so I suggest posting a query thread and seeing what rolls in (pun intended).
 
:D

I think most countries legislatures have got a little less 'civil' these days. ;) Not had any punch-ups in the commons yet though (well, not since the days of Charles I ;) ).

Ahhh...one of those. Excellent, I now know what to buy, lots of those for sale on amazon UK. Big thanks for that, appreciate it. :)

As for preload I'll go with the instructions Jake M has provided and see how I go. Might give ArcEuroTrade a bell for some advice if I get stuck.

Feeling a lot less nervous about swapping out the bearings now. Cheers.
 
Back
Top