PDA

View Full Version : Hard turning vs grinding



gfphoto
07-09-2009, 10:02 AM
After reading Evan's post I looked up "hard turning" and found this article. (http://www.mmsonline.com/articles/hard-turning-might-not-be-as-hard-as-you-think.aspx)

The article says: "For some shops, though, the process of repeatedly turning parts that are harder than 45 Rc to grinding-level accuracies is still a bit cloudy."

What kind of grinding are they talking about that is so accurate?

Is there some way that what Evan showed could be done by grinding?

Any explanations appreciated.

Thanks,

Gary

Evan
07-09-2009, 10:04 AM
It's very difficult to beat proper grinding by any sort of turning operation except in the few circumstances where grinding cannot be easily applied. Hard turning is a very commonly used operation but when you need a really accurate ID or OD on something grinding is the method of choice.

As for what sort of grinding, a good tool post grinder can do both ID and OD grinding on a lathe. It will have accuracy to whatever level of positioning the lathe is capable of as well as the quality of the tool post grinder and the type of stone or wheel used.

The Fixer
07-09-2009, 10:07 AM
The bearing race he was turning would have been factory ground for both precision and for surface finish.

gfphoto
07-09-2009, 11:27 AM
Thanks for the info. I guess I'm just surprised that grinding would give a better finish and be more accurate. Will research more...

Gary

digger_doug
07-09-2009, 11:54 AM
Thanks for the info. I guess I'm just surprised that grinding would give a better finish and be more accurate. Will research more...

Gary

Gary,
I believe you are basing that assumption on what you do on
a bench grinder. In the business this is known as "offhand" grinding
or "snagging" and Norton or other grinding wheel literature
describes wheels for just such usage.

Precision grinding is considered more precise than machining,
as well as producing a far better surface finish.

For even more, google "superfinishing" using abrasives
in a different way, to improve a ground finish even more.
It's used on hydraulic engine valve lifters to give them a
mirror finish, and correct up the roundness just a little more
from what the grinders left.

METALARCHER
07-09-2009, 12:01 PM
To carry this one step further, how would surface grinding as opposed to using a face mill or fly cutter work? using a flaired cup wheel in a mini mill?

pcarpenter
07-09-2009, 12:05 PM
Cylindrical grinding operations are generally a better way to reach high precision and better surface finish than any turning operation would hope to. It's the reason that they make "toolpost grinders" for lathes (for shops that don't have and can't cost justify the high expense of a cylindrical grinder of some sort). They make a mess and require huge amounts of care in shielding the lathe ways etc. to keep the grit off, but they will do for the rest of us....on the rare occasions that we need that sort of precision.

Don't think of some dude standing in front of a bench grinder.....rather, think of the part spinning in precision bearings with a diamond trued wheel counter rotating and taking off a few ten-thousandths of an inch. In effect, then, rather than a single point of contact with the work, you have thousands of tiny points of contact that are constantly rotating past the spinning part, creating an aggregate cut that is very smooth and very precise. Depts of cut may typically be in the tenths range.

For even more precision, some applications call for "centerless" grinders in which the part is turned between two wheels....and idler that effectively rides on the back side of the part, behind the grinding wheel and the grinding wheel itself. This way, the only thing that establishes the part dimension is the part itself. It takes any issues of the precision of spindle bearings (or lack thereof) or heating of the spindle etc. out of the equation.

If you think about a normal turning operation, say you had a precision specified down to "tenths" (ten thousandths of an inch or a "tenth" of a thousandth). If your next to last pass yields a part that is only .0005" over tolerance and your tolerances are (-.0002,+.0000), how are you going to make a turning cut that takes off .0005"? Cutting tools tend not to work very well when not loaded properly. Steel is being sheared in normal turning. You could probably finish by hand with an abraisive strip....but that's not so precise. In the cases where this sort of tolerance is called out, the part might be finished on the lathe to within a few thousandths and then finish ground to size.

If you still tend to think of grinding as an inherently crude operation, remember that eye glass lenses (and lenses far more precise than that) are ground using purpose-built machines. There you are grinding both to the point of optical clarity, as well as to a precise curvature to make your prescription different than another.
Paul

digger_doug
07-09-2009, 12:08 PM
To carry this one step further, how would surface grinding as opposed to using a face mill or fly cutter work? using a flaired cup wheel in a mini mill?


Well not exactly, you need a higher traverse speed for starters.

Look at surface grinders, these are what you would use.
There have been surface grinders built as you descibe
(the wheel orientation, not the actual contruction)
Pratt & witney made one.

Also please notice how dedicated grinding machines do not
have exposed ways. Everything is covered/mounted behind
so as to stop the ingress of the grinding dust/grit.
Some also incoporate flood cooling as well, with the added
gutters on the table, the settling out apparatus, and
a holding tank with pump.

DR
07-09-2009, 12:43 PM
One of the punch and die companies maintains an inventory of heat treated blanks.

When you want a specify size punch it's a simple matter for them to hard turn one of the blanks to the required size.

Prior to their use of hard turning the process involved rough turning a soft blank, heat treating the rough turned punch, then grinding to size. Relatively long lead time as opposed to the hard turning method.

Glenn Wegman
07-09-2009, 12:45 PM
O.D grinding a hardened steel pin. That's a 14" wheel.
http://img.photobucket.com/albums/v647/Fighter1/OD.jpg

I've ID ground sprag clutch housings (similar to what I believe Evan is making) but can't find the pics or I'd have posted one of them.

Fasttrack
07-09-2009, 01:15 PM
To carry this one step further, how would surface grinding as opposed to using a face mill or fly cutter work? using a flaired cup wheel in a mini mill?


Surface grinder spindles are much more accurate than any mini-mill spindle. In fact, they are more accurate, in terms of TIR, than most any milling machine spindle. Not sure how the newer grinding machines work, but the older models had hydrodynamic bearings. The spindle ran in on a high pressure film of oil. I think someone here (probably Forrest Addy) posted a great link to information about these type of bearings as they applied to grinders. They can't handle the abuse that a good milling machine spindle can, in terms of load and "hammering", but they have less runout.

toastydeath
07-09-2009, 01:35 PM
Couple more differences between the grinding operation and regular machining operation.

Grinders have the magnetic chucks ground ON the machine. So the chucks are as flat as the ways are. Second, the tool itself is "made" on the machine; when you dress the wheel, you are setting the profile of the wheel parallel to the ways of the machine. There is no tramming, et cetera. Third, grinding machines usually have full support of the table. A 6x10" grinder will have a 12x20" bed - the table is supported fully no matter where it is on the machine. Not many milling machines, even today, have this feature.

That the chucking is usually either magnetic or vacuum means there's only infinitesimal distortion of the part. There's also no parallelism error due to stacking vises up, or anything similar. You can grind magnetic parallels just like you grind the chuck itself; all tooling on a grinder is FIT to that grinder.

At the point of contact between the wheel and the part, each grain in the wheel forms a cutting tip. So there's hundreds, thousands of cutting tools all at the same depth of cut. As one dulls, it either fractures (exposing a new, sharp edge), or pulls out of the binder (exposing more grains). Your cutting tools on the milling machine dull, and stay that way. The only time the accuracy of a grinder falls off is when you've used the entire surface of the wheel, and then it's a quick re-dress and you're back in action.

The actual cutting action is one of two primary mechanisms; in a free-cutting wheel, all grains are sharp. Instead of cutting by shear, the highly negative rake on the grains tends to put enough heat in a microscopic portion of the material to plasticize it and then plow it off. There's less deformation that way, since the individual chip is essentially molten. In a dull wheel, the grains have all dulled, and are not either breaking out of the binder, or fracturing. This wheel cuts very slowly, and very hot. However, instead of chipping, it tends to burnish the surface. This leads to very high surface finishes, and good dimensional control. Since the grains are not really coming out or breaking, the wheel is not wearing away and you can go for a very long time without having to re-dress the wheel. That means you can surface a large area without worrying about parallelism from wheel wear. Note that free/dull refers to grain condition only, and has nothing to do with a wheel that is physically worn away from use.

gfphoto
07-09-2009, 03:16 PM
Thank you all for the information and discussion - very interesting. I had been thinking about it more or less in terms of a bench grinder and now can see how the finish would be very good. Also hadn't realized such fine wheels were available.

What still puzzles me is the accuracy. This may be wrong but my understanding of using a regular tool is that you get close, measure the part, and then advance the tool in a few increments to get to the right size - so say moving in by .005" takes that much off the radius. Does it work the same way with a grinder? Doesn't the wheel diameter change as you grind? Or are you starting with a coarse wheel and then for the final cut(s) switching to finer and just not using it up enough to matter?

Also wondering if a vacuum can be used to deal with the dust?

Gary

toastydeath
07-09-2009, 03:35 PM
What still puzzles me is the accuracy. This may be wrong but my understanding of using a regular tool is that you get close, measure the part, and then advance the tool in a few increments to get to the right size - so say moving in by .005" takes that much off the radius. Does it work the same way with a grinder? Doesn't the wheel diameter change as you grind? Or are you starting with a coarse wheel and then for the final cut(s) switching to finer and just not using it up enough to matter?


Yes, .005 takes .005.

The wheel wears ACROSS the surface of the wheel; the whole face of the wheel does not wear down at once. As you grind, the face gets used up. Once the face is completely used, then you re-dress and start again.



Also wondering if a vacuum can be used to deal with the dust?

Vacuums don't get all the dust. Even flood coolant doesn't get all the swarf - some still winds up on the ways if they're exposed.

digger_doug
07-09-2009, 03:35 PM
Gary,
to answer some more of your questions:
1. Finish does not equal precision.

2. Grinders genereally, do both well, in industry
(before the advent of hard turning) grinding was the
final operation, to correct (read accuracy) parts
as they came from turning and milling.

3. Grinding is used after heat treatment to correct
warpage, it is common to machine a part and leave
say .015 excess stock on the part as grinding stock.
The part would be heat treated, then sent to a grinder
the .015 is needed to correct any warpaged, and
remove scale (finish) from the heat treat process.

4. Before hard turning came about, grinding was about the only way
to machine something AFTER hardening.
E.D.M. is available, but not too good with the finish.

5. Wheel wear, yes it's present, you are not (generally)
taking off the material like you would with a lathe or mill.
But you do like to take a last pass, with no infeed,
called a "spark out pass", to correct any difference
in wheel wear.

6. Accuracy, yes sneak up to the required number, mike it
again, adjust accordingly. Holding .001 with a grinder ?
No problem, the B&S surface grinder I ran had a small
wheel (fine down feed) that 1 revolution was .001, so the machines
are built differently from a lathe/mill. I see a gallymer & livingston
(very large surface grinder 18" x 72") that the downfeed
could be selected to drop .0001 with each pass over the table.
BTW none of these machines had any computer controls
of any kind.


Are there variations to these words I wrote ? yes of course.

An example:

A certain diesel engine plant has eliminated milling camshafts
(and the corresponding heat treat, straightening, and grinding
operations) by purchasing PRE-hardended bar stock, and with
a very sophisticated grinder, hogging the camshaft out of the
solid (lobes and all bearing journals) in one shot with a
very stout machine.

gfphoto
07-09-2009, 04:44 PM
Thanks again. It sounds like something I'd like to try and do.

Would some trash-bag plastic and duct tape work to contain dust?

Are the wheels [url=http://www.mcmaster.com/#grinding-wheels/=2o9gv2]here[/url the correct type?

For my 9" lathe is the 3" size right?

Gary

pcarpenter
07-09-2009, 05:47 PM
Yeah...if you have a toolpost grinder and want to try it on your lathe, be really anal about covering everything. Even then, you should figure on wiping all way surfaces with kerosene after "untarping" it and then drying them and re-lubricating. Wheel grit mixed with oil=lapping compound.

Just to correct something you said.....wheels that produce a fine surface finish are not necessarily "fine". Certainly finer wheels often result in a finer surface finish, but overdo it and you end up with a wheel loaded with metal particles which then result in a "galled" surface and a real mess. This is likely more esoteric than you want to get right now, but as a previous poster mentioned, often an open, friable wheel is the right answer so that the wheel does not load up. There are all sorts of "grades" of hardness and openness of the wheel material in addition to grit size to consider. Think of grinding as it's own class of machining....with its own set of stuff to learn. In any case, a "fine" wheel may still be 80 grit with 60 much more common and 46 often used and still producing surfaces with a better surface finish than milling will typically do.

Paul

Evan
07-09-2009, 08:06 PM
Grinding IS it's own class of machining called abrasive machining. There are just as many improvements in abrasive machining in recent years as there have been in cutting tools that make hard turning practical. Modern super-abrasives make possible grinding operations that were impossible just a few years ago. Using CBN wheels in profile configuration very little wear occurs to the wheel and it maintains it's shape for a long time. Almost no grit from the wheel is released to contaminate the work area.

Anothe recent technique is "creep feed grinding" where very deep cuts are taken at low advance rates. This is also only possible with wheels that don't wear quickly.

Here is an example of creep feed grinding where I made a form tool for milling plastic parts. The tool is ground to profile with a copy attachment and the wheel is fed to full depth of cut in a single pass. It is an 8 inch CBN (Cubic Boron Nitride) wheel and doesn't require dressing.

First the edge profile is ground and then the clearance and rake angles are ground, each in just a single pass.

[edit] The material being ground is High Speed Steel.

http://metalshopborealis.ca/pics3/bushing2.jpg

http://metalshopborealis.ca/pics3/bushing3.jpg

gfphoto
07-10-2009, 01:13 PM
Here is an example of creep feed grinding where I made a form tool for milling plastic parts. The tool is ground to profile with a copy attachment and the wheel is fed to full depth of cut in a single pass. It is an 8 inch CBN (Cubic Boron Nitride) wheel and doesn't require dressing.

Evan,

How is the blade attached?

Gary

Evan
07-10-2009, 02:40 PM
The last time I posted this I caught a bunch of flack because it was supposed by some that every newbie to machining was going to make something like this and try to machine a log of 4140 with a three inch wide form tool.

I don't make the assumption in my posts the the people reading this are stupid. I have yet to meet a stupid machinist although some can be ignorant at times. ;)

Anyway, don't try this at home folks unless your equipment can handle the load with all variables considered.

The part design was slightly modified to accommodate the change to Delrin.

http://metalshopborealis.ca/pics3/bushing4.jpg

http://metalshopborealis.ca/pics3/bushings5.jpg

gfphoto
07-10-2009, 03:20 PM
Very neat!

Is that a horizontal mill?

Gary

Evan
07-10-2009, 05:59 PM
Yes it is. You can read all about here:

http://bbs.homeshopmachinist.net/showthread.php?t=25928