Acceptable lathe chuck runout?

Hi,
Out of curiosity - what's considered to be an acceptable runout when placing the magnet foot on the firewall and then measuring against the exterior OD of the chuck "body" while rotating around? (by hand I might add..)

Measured 0.01 mm max difference when checking now. It's a practically new lathe with a clean 10 inch chuck.

Are lathes outfitted with adjustment knobs for the spindle/bearings to get it 0.00 mm? Guess that would come in handy if a crash occurred.

Schjell said:

Hi,
Out of curiosity - what's considered to be an acceptable runout when placing the magnet foot on the firewall and then measuring against the exterior OD of the chuck "body" while rotating around? (by hand I might add..)

Measured 0.01 mm max difference when checking now. It's a practically new lathe with a clean 10 inch chuck.

Are lathes outfitted with adjustment knobs for the spindle/bearings to get it 0.00 mm? Guess that would come in handy if a crash occurred.

Click to expand...

it's 0.0003", so 3 tenths, in reality id be happy if mine repeated 3 times that. currently mine is 6X that amount. so for a 3 jaw its fine unless you are splitting microns. only need it as accurate as your parts need to be.
I think most new chucks are only that or a hair better, like 2 tenths of an inch

Schjell said:

Are lathes outfitted with adjustment knobs for the spindle/bearings to get it 0.00 mm?

Click to expand...

No, you can only get them dialed in to 0.0000000000000001 mm.......................

That's like calling a part "perfect".

Schjell said:

Hi,
Out of curiosity - what's considered to be an acceptable runout when placing the magnet foot on the firewall and then measuring against the exterior OD of the chuck "body" while rotating around? (by hand I might add..)

Measured 0.01 mm max difference when checking now. It's a practically new lathe with a clean 10 inch chuck.

Are lathes outfitted with adjustment knobs for the spindle/bearings to get it 0.00 mm? Guess that would come in handy if a crash occurred.

Click to expand...

Shall we guess your home was one of those just mudslid into the sea? Or that Norway's "daytime TV" is as boring as everywhere else?



Just WTF part to be turned is it that you mount to "the exterior OD of the chuck 'body'" anyway? That's a BALANCE issue, not workholding.

Top-grade CNC "non-scroll" operated power chuck specs are published by their makers. Who BEAT their specs, "Day Zero". They build so as to HOLD to them for hard running long enough to earn the price of new or a factory rebuild. ELSE a competitor who can do will eat their lunch.

You have some OTHER grade of chuck? You got whatever you GOT!

Spindle & bearings are a whole 'nuther matter, and best vetted directly on the mount or bore taper. AND NOT on a(ny) workholding system chuck, faceplate, nor collet.

That is a sore TEDIOUS job to do reliably, with "millionths" (of an INCH) metrology, too.

DAMHIKT!

BT Fabrication said:

it's 0.0003", so 3 tenths, in reality id be happy if mine repeated 3 times that. currently mine is 6X that amount. so for a 3 jaw its fine unless you are splitting microns. only need it as accurate as your parts need to be.
I think most new chucks are only that or a hair better, like 2 tenths of an inch

Click to expand...

Should have measured when the lathe was brand new out of the box a month ago.
Reason for asking is that I had an incident today. Managed to add a 12.5mm offset to a static 25 mm drill. Luckily I was working on a fairly long POM part at low RPM so when the drill fed in with an offset it only resulted with a bent 13mm Shank inside the holder and damaged self-esteem. Checked the turret for misalignment, I was lucky.

I converted your figure of 0,0003 inches to be 0,001 mm (one thousandth of a mm).
Are they really that true when set up properly/new?

Really pisses me off. New rule for myself now. No machining after lunch on a Friday, my head was clearly somewhere else. Should have run single block, but thought I could stop it visually. Clearly not.

We are a subsea tooling company where I try to keep parts to within 0.1 mm tolerance when I design them. Quite forgiving stuff really. Tight tolerances subsea only lead to trouble.
My first unplanned "contact" today on the lathe so just became curious to how sturdy these machines are and what to expect when checking them. Made a few cuts, seems it still held accuracy after the incident. Happy I wasn't turning 316 like I did yesterday..

Schjell said:

I converted your figure of 0,0003 inches to <funny numbers>
Are they really that true when set up properly/new?

Click to expand...

Damned well HAVE to be, don't they?

Look at the standard specs on collet systems they have to mount and compete with.

TWO tenths is not the least bit uncommon.

All this stuff is "published information". Just look it up - especially on any machine as NEW as your one.

"Very ordinary" CNC spindles and workholding of the current era hold tolerances that only the "super precision" toolroom/tool & gauge spindles once held.. and those were published "numbers" as well - even 60 and 80 years ago.

RTFM.

Not a damned bit of "mystery" to it ..

Possible exception where to find the MONEY to pay for "the good stuff"?

And the situational awareness to not f**k it up?

Danish firm I once consulted to, the company funded the fresh bottle of "Gammel Dansk" on a Friday luncheon table. You'd have to know Scandihooligans?

Afraid to even ASK what the granite-headed Norsk ... who EXPORT it, rather than DRINK it ... consider a better deal for "home use"!

You'd have to know "Fridays"? Or at least gra_NEET!

Schjell said:

.......My first unplanned "contact" today on the lathe so just became curious to how sturdy these machines are and what to expect when checking them. .......

Click to expand...

Since you have not said anywhere in this thread what the machine is, there is no way for anyone to begin to say how close it should run nor how sturdy it might be.

thermite said:

Shall we guess your home was one of those just mudslid into the sea? Or that Norway's "daytime TV" is as boring as everywhere else?<snip>

Spindle & bearings are a whole 'nuther matter, and best vetted directly on the mount or bore taper. AND NOT on a(ny) workholding system chuck, faceplate, nor collet.

That is a sore TEDIOUS job to do reliably, with "millionths" (of an INCH) metrology, too.

DAMHIKT!

Click to expand...

exactly ,

Not sure I understand the question as most lathe chucks OD are not ground in situ to be concentric -ish and parallel to the spindle bearings.

Like what thermite says it would only be a balance issue.

I.e. the part cut for example in soft jaws or collet reflect more directly the spindle runout / capability + skill in loading the bearing in cut.

@op

1 micron (0.001 mm) is roughly 40 millionths of an inch. 0.00004"

roughly 2.5 micron per "tenth"

0.0003" three "tenths" is roughly 7.5 ish microns - inside / shy of a thousandths. 0.001"

0.0004" ≈ 10 micron or 0.01 mm

___________________________________________

Like what Thermite mentions some internal surfaces are ground in situ to the spindle bearings but even there that's not a representation of what spindle runout makes to part "roundness' etc.

I'll double check my math at some point.

cameraman said:

I'll double check my math at some point.

Click to expand...

It's "the (other) math" as gets complicated.

Example. "Precession" in the bearings. A full day, each of two setups more than a year apart. First one, all I had was a ten-millionths per division Hamilton 4 AGD mechanical DI.

Second go, I also had a Mahr "millimess" electronic as was 20 millionths per-division capable. Also by pushbutton some metrifuckated nonsense, but who gives a shit? Metric sytem is mostly used for boolit sizes.

So what transpires after a whole day of polar-cooordinate charting and averaging many, many, readings ... is that that particular 10EE generates a "D" shaped graph.

One rolling element in its precision spindle bearings is flawed, and the repeat has a low-spot that nets to 130 millionths TIR, not the 50 millionths it had shipped with nearly 80 years earlier. Maybe.

ISTR the repeat was every 13 revolutions?

MEANING.. that each SINGLE revolution, the "low spot" was one thirteenth of 360 degrees displaced, "polar coordinate" wise from the previous revolution.

Now.. does that do a form of "averaging" down the error?

What that does to the departure from a perfect circle of work being single-point turned at a rate of "n" advance per revolution and with a tool-tip that has finite WIDTH to it? Depends. Comes damned good with abrasive final-final, anway.

So not enough .... to cause me to pop for upwards of $2,000 worth of new spindle bearings is what.

"Run what you got". To the best it can deliver.

Not to what you WISH it could deliver!

Vancbiker said:

Since you have not said anywhere in this thread what the machine is, there is no way for anyone to begin to say how close it should run nor how sturdy it might be.

Click to expand...

Apologies, ST-30Y.

thermite said:

Damned well HAVE to be, don't they?

Look at the standard specs on collet systems they have to mount and compete with.

TWO tenths is not the least bit uncommon.

All this stuff is "published information". Just look it up - especially on any machine as NEW as your one.

"Very ordinary" CNC spindles and workholding of the current era hold tolerances that only the "super precision" toolroom/tool & gauge spindles once held.. and those were published "numbers" as well - even 60 and 80 years ago.

RTFM.

Not a damned bit of "mystery" to it ..

Possible exception where to find the MONEY to pay for "the good stuff"?

And the situational awareness to not f**k it up?

Danish firm I once consulted to, the company funded the fresh bottle of "Gammel Dansk" on a Friday luncheon table. You'd have to know Scandihooligans?

Afraid to even ASK what the granite-headed Norsk ... who EXPORT it, rather than DRINK it ... consider a better deal for "home use"!

You'd have to know "Fridays"? Or at least gra_NEET!

Click to expand...

Thanks, no I read the specs in the glossy manual, but I was just wondering how these figures compared to real world values.If they actually are as precise as advertised then I'm well impressed. I agree with other comments that the outside of the chuck is prob not the best place to check runout.
Just consider myself lucky today to learn something without ruining a new machine. Close call.

AS for the booze, we enjoy a drink or ten in the weekends, have to admit that. The Gammel Dansk is Danish though, we Norwegians are big on Akevitt (potato spirit) :-)

Have a good weekend guys, thanks for the comments.

Schjell said:

The Gammel Dansk is Danish though,

Click to expand...

WAS .. for forty years.. Made in Oslo since 2014 though.

Gots to do SOMETHING profitable with the rejects from distilling blighted potatos wild animals have rooted-up and shat upon, yah?

As with Jägermeister, there's enough in the way of weird "herbals" in there to mask the flavour of tincture of cigar butt in JP4!

Takes an "outsider" to spot some things. We have a one here called "Southern Comfort", the flavour a mystery. Until a European tastes of it. Figures it is basically grain ethanol, Peach Schnapps... and Prune juice!