PDA

View Full Version : Pippin' is easy...lathe diagnostics question



RussZHC
08-04-2011, 09:55 PM
So moved on to experimenting with facing cuts today, that pip in the center slowly but surely disappeared, though I begin to understand the issue with some tool holders and their adjustment
(the knurled nut/wavy washer/hollow hex bolt issues).

Perhaps more importantly, the face is not quite flat...with a straight edge the closer you move to the center, it begins to very slightly rock (and I did notice it required a bit more effort to turn the handle...ahh, the value of a manual lathe to learn on;) ).

This leads to a more general question, is there any sort of diagnostic process (flow chart like?) to sort out issues like this? Or is it just a single cause that a noob has to learn about? e.g. could slop in the cross feed nut account for this slight crown towards the middle...not sure that even makes sense but also as an example of where to look first?

Got and have read both HTRAL and the "Lathe Operation" (Atlas book) and while they have both been extremely helpful, I don't recall anything of the "if this happens, then you do this" nature.

gwilson
08-04-2011, 10:04 PM
Put a piece of metal in your chuck,about 1 to 2" in diameter,and 6" sticking out from the chuck. Take light cuts on it to slightly reduct its diameter. They need to be light,and your tool needs to be sharp. HSS is best for sharpness. Take a few extra passes without advancing your cross slide. Measure the diameter of the cylinder you have turned. If there is a difference in diameters,your headstock needs alignment.

Toolguy
08-04-2011, 10:26 PM
Most likely the carriage moved a little to the right unless you had it locked. Most lathes will make a very slight concave facing cut. A sharp HSS tool is good advice. If using a carbide tool, don't go past center, as the material is coming up on the far side of center and will often chip off the cutting edge of the carbide. If possible, I will put a center drill hole or do a drill operation first in order to stop the cutting tool in the hole and not have to worry about going exactly to center.

beckley23
08-04-2011, 11:04 PM
Pushing away may be the problem, but have a look post #175 in the link for a better test.
http://www.practicalmachinist.com/vb/monarch-lathes/wreck-update-146913/index5.html
The parallel is set-up in the 4 jaw such that when the spindle is rotated 180* the indicator doesn't move, in other words a 0-0 reading. That should be done at the end of the parallel. After that has been achieved, traverse the cross slide across the parallel. Ideally the indicator should read 0 to .0005" concave. Make sure you have a 0-0 reading across the width of the parallel, otherwise you may get false readings.
If you get the desired results, then I would look at the push away situation. If the readings show a convex situation, then you have an alignment problem.
Harry

RussZHC
08-05-2011, 12:23 AM
Thanks for the link...and I even understand the "why/how" of the test.
It will take a few days to get done, mostly to either find or purchase new parallels.

With it in print in front of me, I too suspect movement to the right [though I knew to lock movement down as much as possible, don't know why I didn't :( ]

Bob Ford
08-05-2011, 12:42 AM
Using a magnifying glass look at the cutting edge of the tool and also look for rub marks on the tool below the cutting edge. If it is not sharp it will tend to push away from the work. Likewise if there is not enough relief it will push away from the work.

Bob

darryl
08-05-2011, 01:32 AM
When facing, the closer you get to center, the more the tool will want to push away. You might find that you're turning a slight concave, then it goes high in the center. For a better evaluation of the degree of concave, use very sharp tooling with significant cutting edge clearance, and get it bang on height-wise. Take a few spring passes, with the carriage locked and the crosslide and compound gibs snugged.

I've read that you do want some degree of concave as that is better for parts that have to sit together without rocking. A completely flat facing would be ok, but any convex would likely not be ok. Whether or not a manufacturer takes this into account when machining the carriage, I don't know, but of course it is important to have the headstock axis parallel to the bed- if this leaves you turning a slight concave- be happy.

Black_Moons
08-05-2011, 01:46 PM
Put a piece of metal in your chuck,about 1 to 2" in diameter,and 6" sticking out from the chuck. Take light cuts on it to slightly reduct its diameter. They need to be light,and your tool needs to be sharp. HSS is best for sharpness. Take a few extra passes without advancing your cross slide. Measure the diameter of the cylinder you have turned. If there is a difference in diameters,your headstock needs alignment.

Please do NOT RECOMMEND headstock adjustments after the most BASIC of tests without massive instruction on how to do it just right, And the fact that its much more likey to be bed twist then headstock alignment if the lathe has not been set up right.

the last thing a noob should do is try and adjust the headstock alignment. And by last I mean 'never untill hes no longer a noob'

Al Messer
08-05-2011, 08:55 PM
If your lathe is "normal". it was set up at the factory to always cut a slight concave surface when facing.

Al

RussZHC
08-06-2011, 01:57 AM
Locking the carriage down appears to have solved the "crowning" of the facing...I still think something is not quite right however...I took more note of the "how" of the cross slide movement and it feels like it takes more feed pressure from the 4 o'clock to 10 o'clock positions (no power cross, manual;) feed pressure as in effort to turn by hand).

The whole thing appears to be turning a taper...been brushing up on "solutions" so I guess at the top of the list soon is seeing how much run out the chuck imparts, double checking the crowning of the face via method suggested previously, checking level of lathe (not done since complete final assembly so...) then fiddle with tail stock end adjustments.

Edit Sat morning: you know how you have those moment of lucid thought just when you wake up?
While it dawned on me what the overall "goal" here is...to make one of those cylinder squares that is within your needed tolerances each and every time and then, if needed, to bore it out so that the removed "space" is now within a similar set of tolerances...with clearly any desired taper being the exception

RussZHC
08-06-2011, 11:00 PM
Spent most of the afternoon and early evening measuring and tweaking. Not saying its entirely "beat" but IMO much better than it was...
vertical lift of spindle = .0014 (so about double the small suggested by SB for similar lathe w bronze bushings/bearings)
run out chuck mounting plate edge = .0003 (doesn't get to .0002 either side of the zero)
edge of chuck body proper = .004 (.002 either side of the zero) [the more I look at it, now that I have a few to compare it to, this chuck is the more "beat up" of the older used ones; I think the 4 jaw will be much better esp once I learn to use/adjust it properly ;) , next on the list]
chuck face = .003

used test suggested by beckley23 (above, and thanks again by the way, have enjoyed reading all your threads on PM...lots of info for a noob even if I can't use/fathom a fair percentage :) and the movement along the parallel is concave and falls somewhere around .00075 (could not find really fine dial...more than half an increment and could still see space between edge of indicator arm/hand and next increment)

read up on adjusting taper "out" and its now at the point where a single sheet of notebook paper gets stuck about 2.5" along the 4" blade of engineers square...may try and get it better but will let it "settle" first and decide how much "fussing" I want to do...

Thanks all for the suggestions and help:cool:

Forestgnome
08-07-2011, 11:18 AM
If your lathe is "normal". it was set up at the factory to always cut a slight concave surface when facing.

Al
How could that be normal? A lathe is designed to make square and parallel cuts. I can't even imagine how that would be "designed in".

dalee100
08-07-2011, 02:48 PM
Hi,

Because there is no such thing as perfect, many lathes were and still are made to face just a little bit concave. This ensures that if you set it up on end, it will stand upright without a wobble. Normally this little bit is around .001" or less depending on the diameter of the piece in question. This very seldom bothers the final part in any way.

dalee

Forestgnome
08-08-2011, 10:34 AM
Hi,

Because there is no such thing as perfect, many lathes were and still are made to face just a little bit concave. This ensures that if you set it up on end, it will stand upright without a wobble. Normally this little bit is around .001" or less depending on the diameter of the piece in question. This very seldom bothers the final part in any way.

dalee
Could you please explain how this is achieved?

hareng
08-08-2011, 10:51 AM
I would love to know as well.

brozier
08-08-2011, 12:13 PM
When looking from above, the head stock is not quite parallel with the ways that the carriage run on. It very slightly points towards the operator.

This effectively deepens the cut the closer it gets to the centre.

I think the Myford inspection defines this as angle as so many thou when measured over the length of an inspection bar held in the spindle taper.... i.e. it is very slight.

Cheers
Bryan

Forestgnome
08-08-2011, 04:55 PM
It's hard for me to believe lathes are made with built-in taper. The test sheet for mine doesn't indicate anything like that. It specifies "less than" for the taper, which implies zero would be perfect. Otherwise there would be a minimum limit.

Black_Moons
08-08-2011, 05:02 PM
Im pertty sure its the cross slide thats 'off' to add in facing taper, Not the spindle (Or you would turn a taper along the length too)

And it would be something like +0 degrees -0.01 degrees tollerance

Ie: Yes, Flat is perfect and what you idealy aim for. However you do NOT want convex at all, And as nothing is 100% perfect, you are gonna aim for ever so slightly concave, Maybe the odd lathe comes out with an unmeasureable* convex/concave (NOTHING IS FLAT!!!), But the majority would of been aligned ever so slightly concave.

*Unmeasurable with whatever measuring equipment was in use at the time.

Its the exact same thing as 29 vs 30 degree cutting angle for threading.
Its CRITICAL you not be above 30 degrees, 30 degrees or *less* is however OK, So you aim for 29.5~ or 29.

Forestgnome
08-08-2011, 05:09 PM
Im pertty sure its the cross slide thats 'off' to add in facing taper, Not the spindle (Or you would turn a taper along the length too)

And it would be something like +0 degrees -0.01 degrees tollerance

Ie: Yes, Flat is perfect and what you idealy aim for. However you do NOT want convex at all, And as nothing is 100% perfect, you are gonna aim for ever so slightly concave, Maybe the odd lathe comes out with an unmeasureable* convex/concave (NOTHING IS FLAT!!!), But the majority would of been aligned ever so slightly concave.

*Unmeasurable with whatever measuring equipment was in use at the time.

Its the exact same thing as 29 vs 30 degree cutting angle for threading.
Its CRITICAL you not be above 30 degrees, 30 degrees or *less* is however OK, So you aim for 29.5~ or 29.
That would make more sense, however if it cut concave going in it would cut convex if you continued through beyond center or came from the back side. it would be good to know if it was set up that way.

beckley23
08-08-2011, 07:10 PM
"When looking from above, the head stock is not quite parallel with the ways that the carriage run on. It very slightly points towards the operator."


That's utter nonsense. The headstock is aligned parallel to the bed ways. It is the cross slide alignment that is critical. The tolerance is 0- .0005"/12" concave perpendicular to the lathes' centerline. Check the link in my earlier post for how to set-up the alignment check. It's very easy to do, but getting to that point is whole 'nother story.
Harry

Black_Moons
08-08-2011, 07:51 PM
That would make more sense, however if it cut concave going in it would cut convex if you continued through beyond center or came from the back side. it would be good to know if it was set up that way.

Come to think of it, That has an intresting implimencation for rear mounted cutoff blades:

They leave a convex surface on the work left in the lathe (Assuming ideal alignment and all things don't screw up) But a concave on the part that falls off.

Front mounted cut off blades would do the opposite.

(Not that any of my parting cuts have been nearly accurate enough to notice..)

RussZHC
08-08-2011, 09:57 PM
Black Moons: that brings up a less important question (for me, at this moment)...is there any reason a parting tool could not be used for facing?

I mean, essentially, when you part off stock, aren't you creating two faces?
Of course one may need/want to go back and improve it but...

Asked partially as I likely bought several parting bits that are way big for my lathe...hey, call it a noob mistake but they are very good quality and could come in handy.

Black_Moons
08-08-2011, 10:23 PM
Black Moons: that brings up a less important question (for me, at this moment)...is there any reason a parting tool could not be used for facing?

I mean, essentially, when you part off stock, aren't you creating two faces?
Of course one may need/want to go back and improve it but...

Asked partially as I likely bought several parting bits that are way big for my lathe...hey, call it a noob mistake but they are very good quality and could come in handy.

No reason other then they won't resist side to side bending forces worth a damn, So basicly, they would only be able to do very fine cuts with minimal tool overhang before problems occure.

They do make great grooving tools however, Or symmetrical form tools. (Any task where they plunge right in and are not exposed to much side force)

Consider how useful cerclips are. Would'nt you like a few cerclip grooving tools?

BadDog
08-08-2011, 11:02 PM
I agree with headstock axis being parallel to bed ways to the greatest extent possible. The only "intentional errors" I've ever heard of are the making the tailstock a thou or two high to allow for wear/droop, and making the cross slide a tiny bit out of square so that a typical near side facing cut produces a tiny concavity (like a few tenths in a few inches).