What's new
What's new

New Manual Lathe

Horseshit. When people start talking like this, run. It's an engine lathe. Sure, the lathe maybe can but the operator and the room temp and the tool wear and the material cannot. Maybe on a Hardinge, not anything bigger.



Be still, my heart.
I have no problem holding 0.0005 on my Weiler Praktikant, it is comparable to a Monarch 10EE in accurancy.
 
I have no problem holding 0.0005 on my Weiler Praktikant, it is comparable to a Monarch 10EE in accurancy.

Whoooooosh ....

Okay, to be accurate I should expand. "Holding" in a machine shop has a specific meaning and it isn't supposed to be "have a half pound of heroin in my pocket".

When you come in at 7:00 and there's 200 cutoff pieces of 8620 on your cart, you start the lathe / mill and make parts all day until 3:30 next wednesday, and the parts are within .0002" on all dimensions when you are done, that's "holding". Manual machines simply do not do that. I can imagine a Hardinge on small parts in an airconditioned room in a material that doesn't change, grow, shrink, twist, absorb water, or get hot may be able to do this. But in general it's bullshit, as anyone who has ever run more than three parts at a time on an engine lathe will tell you. Manual machining just doesn't work that way, and worshipping the tool doesn't make it any different. I don't care if your mill was made by the Heavenly Deity Herself, in this level of existence talking about "holding" two tenths on a mill or lathe is silly. Grinder, yes, but that's a different animal and if your measuring is accurate, you've gotta watch your p's and q's even there.
 
Last edited:
Whoooooosh ....

Okay, to be accurate I should expand. "Holding" in a machine shop has a specific meaning and it isn't supposed to be "have a half pound of heroin in my pocket".

When you come in at 7:00 and there's 200 cutoff pieces of 8620 on your cart, you start the lathe / mill and make parts all day until 3:30 next wednesday, and the parts are within .0002" on all dimensions when you are done, that's "holding". Manual machines simply do not do that. I can imagine a Hardinge on small parts in an airconditioned room in a material that doesn't change, grow, shrink, twist, absorb water, or get hot may be able to do this. But in general it's bullshit, as anyone who has ever run more than three parts at a time on an engine lathe will tell you. Manual machining just doesn't work that way, and worshipping the tool doesn't make it any different. I don't care if your mill was made by the Heavenly Deity Herself, in this level of existence talking about "holding" two tenths on a mill or lathe is silly. Grinder, yes, but that's a different animal and if your measuring is accurate, you've gotta watch your p's and q's even there.
I didn't say two tenths, I said five tenths. And I don't do production runs, just prototyping. And yes, one has to be very careful about material expansion due to heating, and be very famliar with material being machined, tooling, etc. Anyway, no big deal, we're talking apples and oranges here with respect to the application.
 
I didn't say two tenths, I said five tenths. And I don't do production runs, just prototyping.

Five tenths is two tenths in the real world, we like bilateral tolerances. And "holding" is production, you can't "hold" a size over one part. It's the same as "repeatability" only in the manual world.

Any effing lathe can take off a half one time by cranking in the dial, even a 1937 flat belt South Bend. That doesn't mean doodly. Coming back and back and back to the same place time after time after time is what "holding" means.

Since the people cannot do that even if the lathe could (and if you are not measuring with good quality tools in a temp-controlled environment, then that is another factor), then the whole concept of saying this is silly. It just doesn't fit the world of machining.

Honestly, if you are 'holding' plus or minus a thousandth on an engine lathe then you're doing good. Hardinge probably better because those are tiny parts but talk 6" pieces of 8620 or something even gummier and things are not like that. And that has more to do with the person running it than the lathe itself. That's why nc is so great -- and talk of silly numbers on an engine lathe is ... open to ridicule :D
 
Five tenths is two tenths in the real world, we like bilateral tolerances. And "holding" is production, you can't "hold" a size over one part. It's the same as "repeatability" only in the manual world.

Any effing lathe can take off a half one time by cranking in the dial, even a 1937 flat belt South Bend. That doesn't mean doodly. Coming back and back and back to the same place time after time after time is what "holding" means.

Since the people cannot do that even if the lathe could (and if you are not measuring with good quality tools in a temp-controlled environment, then that is another factor), then the whole concept of saying this is silly. It just doesn't fit the world of machining.

Honestly, if you are 'holding' plus or minus a thousandth on an engine lathe then you're doing good. Hardinge probably better because those are tiny parts but talk 6" pieces of 8620 or something even gummier and things are not like that. And that has more to do with the person running it than the lathe itself. That's why nc is so great -- and talk of silly numbers on an engine lathe is ... open to ridicule :D

Getting a little carried away here, IMO. I have held tighter than .0005" all day plenty of times on manual machines. The thing is, it falls on the guy running the machine, not the machine itself. If you want to work that tightly on a manual machine, it comes down to a semi-finish cut on every part so you can confirm that the machine is cutting where it's expected to, and frequent checking of the micrometer against a comparative standard or gage block. Not that hard to do if you do it right...
 
Getting a little carried away here, IMO. I have held tighter than .0005" all day plenty of times on manual machines. The thing is, it falls on the guy running the machine, not the machine itself. If you want to work that tightly on a manual machine, it comes down to a semi-finish cut on every part so you can confirm that the machine is cutting where it's expected to, and frequent checking of the micrometer against a comparative standard or gage block. Not that hard to do if you do it right...
Would you normally expect to do that on large batch work ? Most places I worked at would put a grinding allowance on the components and either grind them in house or send them out for grinding if the tolerances were in that sort of ball park.

Regards Tyrone
 
Would you normally expect to do that on large batch work ? Most places I worked at would put a grinding allowance on the components and either grind them in house or send them out for grinding if the tolerances were in that sort of ball park.

Regards Tyrone

Manual machines don't generally get large batch work. If I had huge quantities I'd probably send them out too. Every now and then we'd get an order for 30'ish parts. Did up to those quantities in house all the time. Didn't enjoy it. I didn't quote large quantity commodity type machining, unless just passing it through to a sub. Our work was more often prototyping or stuff that couldn't be sourced in a timely manner elsewhere. And fiddly repair work.
 
Making large batch work on manual machines was quite normal in my lifetime. Back in the early 1980's, I had a sort of temporary early retirement. I bought and rebuilt a 1940's Hardinge manual lathe with the turret and lever cross slide. I found a demand for small fine parts and made thousands of parts on it and my other manual machines and enjoyed every minute, especially the process of planning the turret tooling setups and making special tooling. Even working the levers never got to be a drag. I listened to my favorite radio station and smiled a lot more than when I was at my "real job."

Larry
 
Making large batch work on manual machines was quite normal in my lifetime. Back in the early 1980's, I had a sort of temporary early retirement. I bought and rebuilt a 1940's Hardinge manual lathe with the turret and lever cross slide. I found a demand for small fine parts and made thousands of parts on it and my other manual machines and enjoyed every minute, especially the process of planning the turret tooling setups and making special tooling. Even working the levers never got to be a drag. I listened to my favorite radio station and smiled a lot more than when I was at my "real job."

Larry

And did you have any trouble holding parts within .0005" if you wanted to? Betting not.
 
And did you have any trouble holding parts within .0005" if you wanted to? Betting not.
Well, the parts I made were for items that sold for several thousand dollars each, so it did require care. There were never any quality/fit/appearance issues with my parts, but holding +- .0005" would have been inappropriate for the application. The product was, after all, miniatures of things that were mass produced in the 1940's, so matt black oxide on steel was the main finish.

Larry

Tippmann M1919 1.JPG
 
Getting a little carried away here, IMO. I have held tighter than .0005" all day plenty of times on manual machines. The thing is, it falls on the guy running the machine, not the machine itself.
Seems like we're getting flip-flopped on the meaning of the word "hold".

If you use it the way you describe "I have held .0005 ..." then any machine on the planet will "hold" two tenths. I can take the cheapest thing they have at Harbor Freight, a couple lengths of emery cloth, an etalon indicating mike and "hold" half a tenth.

I would say that a better use of the word in the case of "this machine holds tenths" is that you take twenty pieces of 2" 1018, put them in the chuck, run the cross down to 0 and take a pass. Take the parts out and measure them. If they are all within .0005" of each other then the machine holds a half thou. Not the operator, who can do all the work to make parts the right size with emery paper and twelve measurements. The machine.

If you ever do SPC, this is the first thing you do. Put a statistically relevant sample in the machine and run them. No arfing around, just run them. Then measure all your important dimensions and chart them. Those real numbers will be what the machine "holds". Not speculation, reality.

It's quite possible that drcoelho's weiler will hold a half. But until you've done that, in a controlled environment with certified measuring equipment, you can't (honestly) say that. I can say, "Michael Phelps and Mark Spitz swim the english channel in seven hours" and quite possibly they could but it hasn't happened so it isn't true. It's just a claim.

So many of these bullshit tenths claims flying around these days ... but I guess that's always been true, I remember an older guy laughing about that fifty years ago. We had a Landis grinder that *did* "hold tenths" (yes, ran tests under controlled conditions) but to be honest ? Probably barely, maybe even with a few of the far ends of the curve slipping outside, if one got really critical on temp control and so on. This tenths talks is just bee ess.

can you believe that ? Fifty farking years .... wow :( Poured morgan out of a box underneath the golden gate bridge a long long time ago ...
 
Expanding on EG’s post I can hold tenths all day long on any machine in reasonable condition but as he said if you set the machine to the same position for the batch will all the parts be the same size.

For example I was turning 8 brass pieces to an internal bore of 23mm +-.01mm 80mm deep. I would rough to within 0.4mm then set to within 0.2mm and bore in and out. I would then measure the bore with a bore gauge adjust to the amount which needed to be removed and finish bore in and out. Every time the finished bore was within the .01mm and generally less. But every single time the amount to be removed varied by about .02mm in either direction.

I would understand this to mean that I am capable of holding tenths on that machine but the machine is only capable of repeatably holding a thou.
 
People who talk like that have never actually run production on an engine lathe. Hell, they've probaly never made three parts the same on an engine lathe. It's like people who talk tenths in a non-airconditioned room. It's stupid. It'd be like me talking about the time I won at LeMans. Well, I read a book about it ....
That was the year after I did , wasn't it?
 
Seems like we're getting flip-flopped on the meaning of the word "hold".

If you use it the way you describe "I have held .0005 ..." then any machine on the planet will "hold" two tenths. I can take the cheapest thing they have at Harbor Freight, a couple lengths of emery cloth, an etalon indicating mike and "hold" half a tenth.

I would say that a better use of the word in the case of "this machine holds tenths" is that you take twenty pieces of 2" 1018, put them in the chuck, run the cross down to 0 and take a pass. Take the parts out and measure them. If they are all within .0005" of each other then the machine holds a half thou. Not the operator, who can do all the work to make parts the right size with emery paper and twelve measurements. The machine.

If you ever do SPC, this is the first thing you do. Put a statistically relevant sample in the machine and run them. No arfing around, just run them. Then measure all your important dimensions and chart them. Those real numbers will be what the machine "holds". Not speculation, reality.

It's quite possible that drcoelho's weiler will hold a half. But until you've done that, in a controlled environment with certified measuring equipment, you can't (honestly) say that. I can say, "Michael Phelps and Mark Spitz swim the english channel in seven hours" and quite possibly they could but it hasn't happened so it isn't true. It's just a claim.

So many of these bullshit tenths claims flying around these days ... but I guess that's always been true, I remember an older guy laughing about that fifty years ago. We had a Landis grinder that *did* "hold tenths" (yes, ran tests under controlled conditions) but to be honest ? Probably barely, maybe even with a few of the far ends of the curve slipping outside, if one got really critical on temp control and so on. This tenths talks is just bee ess.

can you believe that ? Fifty farking years .... wow :( Poured morgan out of a box underneath the golden gate bridge a long long time ago ...

Hardly any machine will do that, including CNC. *All* of them have thermal drift. *All* of them need adjustments for tool wear. The newer ones are better, but even they aren't perfect. This has become an argument of semantics, I think. The guy running the machine still is the one *holding* the sizes in most every instance. I pretty rarely touched a piece of emery when holding tight sizes. It's all in keeping the tools sharp and using that semi-finish cut to one's advantage. And definitely also good readouts. Much easier with those.

But yes, there are lots of claims out there that don't always hold water. Nothing new about that.
 
Hardly any machine will do that, including CNC. *All* of them have thermal drift. *All* of them need adjustments for tool wear. This has become an argument on semantics.
Nope. This is exactly how SPC works. First thing is to determine what the real tolerances of the machine are. You do that by just running it and measuring a statistically significant number of parts. NC is obvious, but you could also do this for manual machines by putting the blank in the chuck, put the face tool at 0 on the indicator, put the tool at 0 on the cross, take a cut, then measure measure measure. Now you know what the machine will do.

Now, with spc, you don't ask it to do more than it's capable of. If you do, your processes will not be reliable.

This has become an argument of semantics, I think. The guy running the machine still is the one *holding* the sizes in most every instance.
Not correct. This is THE basis of reliable production machining.

Even if you have a guy making tool offsets for wear, if the tolerance on the part .0002" and the machine wanders around by .001", then your process is out of control and the guy making offset changes is not going to be able to handle it. The machine tolerance is quantifiable. If the person is chasing it all over and catches the drift most of the time or even all of the time, the machine is still not "holding half a thousandth." To say a machine is capable of holding a half thou, it's got to be able to do it on its own.

We need carbide bob over here ... carblob ! debate on aisle three ! :)
 
Last edited:
Nope. This is exactly how SPC works. First thing is to determine what the real tolerances of the machine are. You do that by just running it and measuring a statistically significant number of parts. NC is obvious, but you could also do this for manual machines by putting the blank in the chuck, put the face tool at 0 on the indicator, put the tool at 0 on the cross, take a cut, then measure measure measure. Now you know what the machine will do.

Now, with spc, you don't ask it to do more than it's capable of. If you do, your processes will not be reliable.


Not correct. This is THE basis of reliable production machining.

We need carbide bob over here ... carblob ! debate on aisle three ! :)

I think they would be reliable still, just maybe not as fast as they could be. I have a hard time adapting to that approach, due to my not wanting to make *any* scrap parts mentality! :D
 








 
Back
Top