Spindle runout issues

[Whetstone]

I just bought and tooled up a Fryer MB-14 as my first CNC. I’ve been chasing a run out issue and it’s driving me mad. We are getting 0.003-0.004” runout at the tool and our surface finishes are horrible. The machine is from 2003 and is in what appears to be decent condition with the exception of this issue. TIR measured at the spindle is .0002, but .003-4” at the er collet nose. I noticed a little bit of play in the spindle bearings and thought that they could be the culprit, so changed them out. Some of my surface finish issues improved and everything sounds much, but still have this TIR issue and am getting bad parts.

I’ve cleaned the spindle taper and don’t see any major dings or marks that could throw off the tool holder. I blued up a tool holder and am getting very good contact all the way around.

I’ll try to figure out how to up load movies of these issues, but I’m beyond frustrated and do t know what else lo look at. A new quill and spindle is $3000 from fryer and that will be the next direction I go unless I can get a recommendation of other things to inspect.

FYI, All the tool holders are brand new from Mari tool.

Indicator at nose of spindle
iCloud

Indicator as high as possible in spindle

iCloud

Run out at er called taper

iCloud

Image if inside spindle taper

Sent from my iPhone using Tapatalk

 

[Whetstone]

Stick a tenths test indicator on a 1/2" - 3/4" toolholder near the bottom, with a rod clamped into the tool socket for leverage. Apply ~25lbs of force on the rod (note the distance to the taper) and see what sort of readings you get for displacement.

Remember that as a bed mill (presumably with both a quill and a sliding headstock) there's a lot of potential "slop" points, so you have to figure out where looseness is coming from. Ideally use multiple indicators so you can check different elements under the same loads, and see if any are particularly loose, or don't return to the original "0" when the load is released.

Let us know the results for more suggestions.

I can definitely do that, but I dont see how that can be affecting my run out issues in a static position. I have 2 tenths at the spindle and 3 thou at the tool holder with no load applied.


Sent from my iPhone using Tapatalk

 

[michiganbuck]

Does clocking the tool holder to 0, 90, 180, 170 show that the .003 follows the tool holder or follows the spindle.

 

[Whetstone]

Does clocking the tool holder show that the .003 follows the tool holder or follows the spindle.

Spindle. Runout is the same, no matter how the tool holder is clocked.


Sent from my iPhone using Tapatalk

 

[cyanidekid]

iI'll say the spindle taper doesn't look all that good, a bit banged up. the blue "print" doesn't look all that good to me either.

if the spindle bearings are properly replaced with quality units, with correct preload and grease, and of course you checked the spindle for bend on a calibrated surface plate with qualified setup and instruments, and ran it in for a while to seat everything, rechecked preload, it STILL should have an in-situ regrind to get everything right with the world. otherwise, you are chasing your tail?

 

[michiganbuck]

Spindle. Runout is the same, no matter how the tool holder is clocked.


Sent from my iPhone using Tapatalk

I not asking about the amount of RO, but if the high place (or low)is to the same position in relation to the place of the spindle as you pull and replace the tool holder in a different position.

Trying to see if the tool holder error may be part of the error.

Guess you could check the tool holder in a V block.


A taper regrind may be due but first you want to know the tool holder is not causing some of the error.

Could indicate 1" deep and 4" deep in the taper with a slow turning and trig out the error/wobble that would occur at the length of the too holder.

 

[Whetstone]

I not asking about the amount of RO, but if the high place (or low)is to the same position in relation to the place of the spindle as you pull and replace the tool holder in a different position.

I’ll have to check tomorrow.


Sent from my iPhone using Tapatalk

 

[Whetstone]

iI'll say the spindle taper doesn't look all that good, a bit banged up. the blue "print" doesn't look all that good to me either.

if the spindle bearings are properly replaced with quality units, with correct preload and grease, and of course you checked the spindle for bend on a calibrated surface plate with qualified setup and instruments, and ran it in for a while to seat everything, rechecked preload, it STILL should have an in-situ regrind to get everything right with the world. otherwise, you are chasing your tail?

Yeah, I think I’m just chasing my tail on this one. I’m one step above a hack trying to figure out what’s wrong.


Sent from my iPhone using Tapatalk

 

[michiganbuck]

QT Op: [eah, I think I’m just chasing my tail on this one. I’m one step above a hack trying to figure out what’s wrong.]

The hack repair would be to set up a tool post or die grinder and grind a used collet holder in place so to run parts, while you wait for the professional spindle grinder hand to come to grind your Taper. Then put the hack holder at the same 180* mark every time. and paint it red so you remember it is a hack holder.

If you prove the taper is the problem.

Yes, use your wife's favorite towels to cover the machine while grinding.
(That so you won't grind with a mill very often.)

Dedicated size tool holders are the most suitable, much better than collets so you do need to make the machine right.

 

[jccaclimber]

Stability of the spindle structure is "foundational", I want to make sure it's good before worrying about the runnout.

Mainly of echoing the above:
1. Stick an indicator on a gauge pin in the tool holder, apply some force by hand, and see how much it moves around. If it moves a bunch under a very light load then firms up you still have something wrong with your bearings. Perhaps incorrectly preloaded or something else. Spindle bearings aren't always the same thing as generic bearings of the same envelope dimensions.
2. Low runout bearings or generic ones? Low/high points aligned. A little bearing runout multiplied down through the nose and tool holder can become a bigger number.
3. A bent spindle or taper that is bad or ground off center will cause similar issues.

 

[michiganbuck]

And proper bearings need to be eccentric marks lined up to run true.

 

[guythatbrews]

Guys will slam a toolholder in the spindle at an angle a lot. Pull stud dings up the taper. You don't have to manually load the thing straight. It's like a funnel, right? Used to drive me nuts.

It's really hard to see all the way to the small end of the socket. May be nicked or have a small chip mashed in there. Not likely tho since you surely looked very close when new bearings installed.

 

[eKretz]

That's not Hi-Spot on the toolholder, is it? Doesn't look right. Don't use layout blue for that. I think at this point I would apply Hi-Spot to the adapter and use that to print the spindle. A new Mari Tool adapter should be almost reference quality. Use that to check your spindle taper. Sounds like it may not be seating correctly. You could also try setting a pin in a collet chuck on the lathe and then gripping the same pin in the adapter collet. Then indicate the adapter's tapered OD while rotating the lathe's spindle - if there's an anomaly with the adapter, that should show it. Considering you sound like you have had this problem with multiple adapters, I don't think that is the issue.

 

[Turbowerks]

Spindle test bar, not cheap but makes trouble shooting a problem like this much easier. If your carful and good at turning make one between center and it should be ok to get a grip on whats going on without spending the big bucks.


When I find it I don’t need it
When I need it I can’t find it!

 

[cyanidekid]

Yeah, I think I’m just chasing my tail on this one. I’m one step above a hack trying to figure out what’s wrong.


Sent from my iPhone using Tapatalk

just be sure to work in a linear, stepwise fashion. if you do, its impossible to chase your tail!

start with the most basic thing, the trueness of the spindle it's self, if you can. there is a ton of good advice in this thread already, particularly MIbuck.

one note of caution, NOBODY'S off the shelf anything, even from someone with a good rep like Maritool, without certification or better yet proper in house qualification on certified stuff, can be considered a "reference". you may be lucky and its really good, but unless you know with some reasonable certainty, that is the surest way to start going in circles!

 

[eKretz]

just be sure to work in a linear, stepwise fashion. if you do, its impossible to chase your tail!

start with the most basic thing, the trueness of the spindle it's self, if you can. there is a ton of good advice in this thread already, particularly MIbuck.

one note of caution, NOBODY'S off the shelf anything, even from someone with a good rep like Maritool, without certification or better yet proper in house qualification on certified stuff, can be considered a "reference". you may be lucky and its really good, but unless you know with some reasonable certainty, that is the surest way to start going in circles!

That's why you check it... Pretty easy task with an indicator and stand, sine bar and gage blocks. As far as I'm aware, Mari checks their stuff pretty closely and uses some very high end equipment. I would trust them to be dead nuts. But if I found a problem I would of course check the adapter myself; quick and easy to do.

 

[cyanidekid]

That's why you check it... Pretty easy task with an indicator and stand, sine bar and gage blocks. As far as I'm aware, Mari checks their stuff pretty closely and uses some very high end equipment. I would trust them to be dead nuts. But if I found a problem I would of course check the adapter myself; quick and easy to do.

Maritool would be the first to say they don't ship every unit "dead nuts". (what is that anyway, everything has a tolerance?)

hey, im not sayin I wouldn't do it EXACTLY the same way you would, lol, but if you are laying out a strategy to avoid chasing your tail, im just sayin check/verify stuff FIRST, then test the next thing with some confidence you have an idea what's going on without having to go back a few steps.

bottom line, no off the shelf holder can be truly used as a spindle test bar. how do you know the .0002 you are looking at is from the holder or the spindle, and is that additive subtractive or neutral? sure, flip it around, but to use that as truly diagnostic, your making another set of assumptions.

 

[michiganbuck]

If you indicate the spindle taper ID at 1" deep and at 4" deep with slowly turning, with clock-in to the place, so to be that the radial position place is noted. Then think about, is it pretty much in the same clock place for high and low for a lop-de-lop... not a wobble.

likely you would need to draw/sketch the taper with 0-90-180-170* and mark the drawing as you indicate it.

A loop-de-loop all the same way usually gives the same error when at tool holder's length...a wobble can magnify the error the farther out you go.

Someone said between centers make your oun taper gauge..That is a very good idea.

 

[cyanidekid]

Someone said between centers make your oun taper gauge..That is a very good idea.

.. especially if you have your Okamoto cylindrical grinder and taper offset setup dialed in. .

 

[eKretz]

Maritool would be the first to say they don't ship every unit "dead nuts". (what is that anyway, everything has a tolerance?)

hey, im not sayin I wouldn't do it EXACTLY the same way you would, lol, but if you are laying out a strategy to avoid chasing your tail, im just sayin check/verify stuff FIRST, then test the next thing with some confidence you have an idea what's going on without having to go back a few steps.

bottom line, no off the shelf holder can be truly used as a spindle test bar. how do you know the .0002 you are looking at is from the holder or the spindle, and is that additive subtractive or neutral? sure, flip it around, but to use that as truly diagnostic, your making another set of assumptions.

That could be said about the test bar too... For that matter about anything you have not verified yourself.

And BTW I was referring to using the tapered surface of the adapter to check the tapered surface of the spindle with Hi-Spot. Not using the adapter as an impromptu test bar. The taper needs must be about dead nuts (yes, highly sophisticated technical term... ) or nobody would buy the adapters...