Chuck & spindle nose flex?

[HammondR]

Hi folks,

Been having some trouble with chatter on a Viceroy which I've recently bought and refurbed. I'm trying to ascertain whether the bearings are shot, or whether there's just a problem with the chuck mount.

I posted about this over on the MadModder forum but someone suggested I ask here too.

The crux of the problem is this: If I chuck up a 6 inch steel bar, I can press on the end of the bar with my fingers and cause the chuck body to move up or down by about 0.03mm (ie about 1.25 thou).

If I put a 200mm long test bar in the spindle taper with no chuck attached, and a sensitive indicator on the spindle nose, I can deflect the spindle nose up and down by just under 0.01mm.

I'm trying to find out if this is the normal amount of deflection or not, and therefore whether it's a factor in my chatter problem. The chuck is definitely fitted tight up against the shoulder on the spindle, and there are no burrs or other irregularities on either the chuck or the spindle nose.

I don't suppose anybody knows if this is normal? Or could do a similar check on their lathe to see?

Thanks,
Richard

[dazz]

Hi
The Denford is a relatively small and light lathe.
Flex is normal.
Too much is abnormal.

I would run a few cuts on a test piece to check that you can turn a parallel bar.

You have to allow for deflection in order to achieve accurate dimensions.
Specifically you should aim to reach a target diameter with a series of equal steps rather than creeping up on the diameter.
Using this method I can hit a dimension 1/20th depth of cut.

[HammondR]

Hi Dazz,

I'm concious that the lathe does have it's limits, but I do wonder whether something's up, it starts to struggle at about 1mm depth of cut on a 20mm bar (only 50mm extended from teh chuck), which I think should be achievable. If you introduce the tailstock though it doesn't struggle at all and will take the same cut on a 50mm diameter bar easily.

It is turning parallel, I confirmed this by turning an 8 inch long, 60mm dia aluminium bar - had to take very light cuts to avoid chatter but I got it done and it only tapers about 0.01mm in 8 inches. I had to rescrape the head to achieve that though (to begin with it was pointing down and towards the operator).

For some reason on this particular Viceroy the rear bearing is a ridiculously tight fit on the spindle, which is making setting it a bit tricky, I'm going to have another fiddle with that but then I think I might go for some new bearings to see....

I would like to know what the normal flex amount on the chuck is though.

[rpwilson]

check the bearing preload before you do much else. Its explained in earlier posts. As a rough guide, when the preload is correct, the headstock near the bearings should be comfortably hot to touch after about an hour running at top speed.

[dazz]

Hello

My lathe and I are currently on opposite sides of the planets so I can't take any measurements off my lathe.
The torque drag method is most commonly quoted for checking preload.
A better method is to measure axial deflection to a set force.

In practice this would mean applying a set force (say 25kg) against the face of the spindle and measuring how far is moves into the headstock. This needs to be calibrated by starting with a lathe with correct preload. Chicken and egg.

The bearings are readily available. They are used on the wheels of Cessna aircraft.
The inner/outer cones are purchased separately. Be sure to get the short version of the inner cones.
True precision grades of these bearings are not available.

If you are thinking of replacing the bearings, there won't be any harm in cranking up the preload to see if that cures the problem.

[HammondR]

Cheers guys. I'm going to double check the spindle preload again though I don't think there's anything else I can do on that front, it's already really tight. The bearings do get pretty hot.

I can't find the bearings with a 'short' inner cone but I've looked carefully and I can't see that it will make a difference, I have removed all the gears from inside the headstock and just have a pulley directly on the spindle (for a VFD drive), connected straight to the layshaft in the cabinet, so I think any difference in bearing depth will in a worst case scenario mean I need to machine up a spacer of some kind between the rear bearing and the lock nuts. I think if the gears were still present this could cause alignment issues though.

I also can't find the 'flanged' version of the bearings in stock anywhere, on my lathe the back of the bearing pocket is 13mm deep but the bearing is only sitting 11mm deep and is resting on a flange on the bearing cup, on the face of the headstock. The bottom of the pocket is machined though so I think I will just press the unflanged bearings in until they bottom out. This will block off the grease port but I never use those anyway.

I can't imagine any of these differences will cause a problem but if anybody knows different please let me know

Will post some updates once I'm done.

Cheers,
Rich

[dazz]

Hello
If you look on the edge of the bearings, you will find the part numbers etched.

The bearing will run hot with too much preload, too much or too little grease.

I got a 10 deg C temp rise when I set preload on my lathe.
See here: Setting Spindle Preload

[HammondR]

Hi Dazz, cheers for your help, I think I have this sorted now.

The rear bearing is, presumably erroneous, a tight press fit on the spindle. As I've ordered new ones I didn't fear too much overtightening the rear bearing nut, so I got a 2ft long spanner and went at it... At first it wouldn't move, then it cracked and I moved it maybe a quarter turn.

End play disappeared and the spindle nose deflection almost disappeared, now, I can take a 0.4mm deep cut on a 40mm steel bar 6 inches stickout, without any chatter -> https://www.youtube.com/watch?v=Iru5wiplJkU

Really chuffed with how smooth it's cutting - that's at about 1800rpm, mild steel. I just finished doing a full rescrape of the whole lathe (except the bed itself), and it's cutting absolutely zero taper.

Previously the exact same steel bar would chatter at anything more than 0.06mm

The bearings are getting really quite hot, I can keep my hand on the spindle nose or my finger inside the spindle, but not by much.

At some point I might swap the bearings for the new ones and the spindle for my spare spindle, which is 0.01mm smaller where the rear bearing fits.

Cheers,
Rich

[dazz]

Hi

That's good news.

For that length of work piece, I would normally use the centre or a steady. It makes a big difference to work stability and surface finish.

I retained the back gears for my VFD conversion. I find it is useful to have the extra torque when required. I probably engage back gears around 25% of turning.

Looks like you have a few mini-projects to do, like replacing the broken hand wheel. I also suggest a rear chip shield.

I had a similar experience to you with instability and poor performance. It was mostly me. I can now turn out shiny chunks of metal to tight tolerances.

[HammondR]

Yeah I normally would use the tailstock but it's a test bar (for testing for parallel turning) so the TS would defeat the object, it looks long but it's actually only roughly 3x the diameter stickout which is fine.

The broken handwheel is a pain, I have a metal one I can retrofit though. I really want to put a handle on the tailstock handwheel too, it baffles me that they didn't equip it with one, it's a real pain drilling with it.

I know what you mean about the backgears, I was very tempted to leave them but I rarely turn slowly and I didn't want the extra noise or complication. I overspecced the motor (3hp) so I can turn it on a really low frequency and still get pretty good torque I just put a taper lock pulley straight on the spindle, runs nice and quietly. I can get fairly good torque down to about 100rpm. Need to equip the motor with an extra fan like yours though.

You're probably right about the rear chip guard too, I could probably do with a decent front one as well.