Weird runout issue

[aaronpadilla]

Hi there, I am trying to build an adaptor for my Van Norman 12.

I had never tried some thing of this level of precision, and right now I am getting bad results, I am running over an issue that I just don't get.

My idea is to turn a 3/4 rod and then mount over this a er32 to mt2 adapter, and then mount my prevoiusly made adapter over the mt2 in order to correct the taper with the toolpost grinder.

Well I did a try and the result was not that good, I got a runout of 1 mil near to the mill spindle, wich worsens on the tool.

At this point I had always asumed that a fresh turned rod over the chuck turns truth, but I was unable to find the source of runout so I did a check of my lathes.

I have a 10 inch taiwan made lathe and a standard modern 13 inch series 2000 utilathe

Well, that's what I don't undestand: a fresh turned rod that still mounted in the chuck runs out about 0.02mm or 0.8 mils, it seems to be a little better on the taiwan lathe, but when I check my cut with micrometer, it seems to be perfectly round, under a tenth of mil.

I have tried different combinations of materials, cutting tools, chucks, preload of spindle bearings and checked the spindle runout, everything looks normal but I get allways the same result

Some hint?

 

[Asm109]

Well, that's what I don't undestand: a fresh turned rod that still mounted in the chuck runs out about 0.02mm or 0.8 mils, it seems to be a little better on the taiwan lathe, but when I check my cut with micrometer, it seems to be perfectly round, under a tenth of mil.


Some hint?

Your lathe is cutting a part that is not circular, it is lobed, probably 3 lobes. You measure the rise and fall of the lobes with the indicator.
With a 2 point measurement of a micrometer you can't pick up the lobing. There are V block type mikes that can pick up this error.
Don't ask me how or why your lathe is cutting a lobed part,

 

[markba633csi]

Have you checked the inside taper of your lathe spindle(s)? It's hard to imagine the lathe cutting the type of runout you describe, unless it is a lobing issue as mentioned above. Suspect your measuring technique when things don't make sense
-M

 

[aaronpadilla]

Your lathe is cutting a part that is not circular, it is lobed, probably 3 lobes. You measure the rise and fall of the lobes with the indicator.
With a 2 point measurement of a micrometer you can't pick up the lobing. There are V block type mikes that can pick up this error.
Don't ask me how or why your lathe is cutting a lobed part,

Wow!
Thank you

 

[Ken226]

Was your lathe bought new, or used? Spindle bearings ever replaced?

In lower precision grade bearings, one or more rolling elements (balls or tapered rollers), that aren't sufficiently round will create a lobed runout.

Since the roller is smaller in diameter than the inner race, the out-of-round rollers will make several revolutions per revolution of the inner race, resulting in a multi-lobed "potato" shaped runout pattern.

Is it possible that someone it the past put some wheel-hub bearings in your lathe, or even that it came from the factory with cheap bearings?

Your talking about runout in the .02mm range. P0 bearings allow up to twice that amount and are still considered in-spec.

I would never accept bearings of less than P6 precision in a lathe spindle. While some brands, such as Timken, NSK and FAG typically have better runout than their precision class allows, Chinese bearings do not.

A P6 class allows up to .0004" of radial runout. A Chinese P6 tapered roller bearing, will have every bit of that.

 

[aaronpadilla]

Was your lathe bought new, or used? Spindle bearings ever replaced?

In lower precision grade bearings, one or more rolling elements (balls or tapered rollers), that aren't sufficiently round will create a lobed runout.

Since the roller is smaller in diameter than the inner race, the out-of-round rollers will make several revolutions per revolution of the inner race, resulting in a multi-lobed "potato" shaped runout pattern.

Is it possible that someone it the past put some wheel-hub bearings in your lathe, or even that it came from the factory with cheap bearings?

Your talking about runout in the .02mm range. P0 bearings allow up to twice that amount and are still considered in-spec.

 

[aaronpadilla]

Hi Ken226,

You may be hit the nail on the head, the taiwan one, well, it's born cheap.
The SM I have not changed any bearing, it is old and some gears in the headstock are worn, there was many metallic debris when I changed oil so, many of then have certainly passed multiple times inside the spindle bearings.

Thank you,

 

[Ken226]

Hi Ken226,

You may be hit the nail on the head, the taiwan one, well, it's born cheap.
The SM I have not changed any bearing, it is old and some gears in the headstock are worn, there was many metallic debris when I changed oil so, many of then have certainly passed multiple times inside the spindle bearings.

Thank you,

What makes those types of issues even more confusing to measure, is that the lobe pattern precesses along the length of the turned shaft. Like, the runout pattern turns helically, like a coarse thread.

The ratio of roller circumference to race circumference is likely never a whole number, integer ratio.

So, every revolution of the spindle will precess the lobes and troughs along, a few degrees further than the last revolution.

 

[epanzella]

Sweep the part under power with an indicator mounted in your toolpost. Lobes will be obvious.

 

[Tim9]

Hi Ken226,

You may be hit the nail on the head, the taiwan one, well, it's born cheap.
The SM I have not changed any bearing, it is old and some gears in the headstock are worn, there was many metallic debris when I changed oil so, many of then have certainly passed multiple times inside the spindle bearings.

Thank you,

I’m not an expert but in my opinion, if there’s metal in the crankcase from bearings... the metal from bearings tends to be little tiny flakes. Metal from gears is more of a dark black dust consistency. But bearing debris is made of chrome flakes. It’s a very distinctive sort of debris.
Now granted, all the above is a rough rule of thumb. There will always be the odd case which none of the above applies such as flakes coming off of bearings which is the sitting in a crankcase for years which has water in the bottom of the crankcase. That might get rusty. But, I would think most bearings shed bright flaky metal.