Standard Lathe, Size and Coaxial Accuracy vs. Standard CNC Milling, Size and Coaxial Accuracy 1

[SpaciouS]

Hi,

I'm investigating the difference in precision between standard lathes and standard CNC milling machines, with respect to diametrical size tolerances and coaxial tolerances.

What are, typically, the tightest tolerances for each?

I'm an opto-mechanical designer and I'm currently designing lens barrels, and this information would be helpful.

Thanks,

Thanks,
Sean

 

[ornerynorsk]

Lathes will generally be more accurate and more cost effective for producing any kind of "turned" cylindrical part.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[SpaciouS]

ornerynorsk,

Would you have any, general, tolerance numbers for diameters that range between 10 mm and 50 mm?

How much more precision can I get with a lathe over CNC milling?

Thanks,
Sean

 

[ornerynorsk]

I don't have any reference material handy at the moment. What you can do is to look up various mill and lathe manufacturers spec sheets and refer to positioning and repeatability. The thing to consider is that with a lathe, the roundness is being produced by rotation of its spindle and the inherant accuracy of the bearings, while with a mill, the roundness is produced by 2 axis' positioning themselves in a "step-by-step" manner. I will believe that the lathe has the advantage.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[SpaciouS]

Thanks

Thanks,
Sean

 

[SpaciouS]

ornerynorsk,

Last question; I will also do what you recommended earlier.

Can I achieve +/-1um [.00004"] bore size, and +/-5um [.0002"] coaxially from one bore to the next inline, using a milling machine?



Thanks,
Sean

 

[MikeHalloran]

High-end CNC mills are capable of jig-borer accuracy for location of bored holes. I don't think they're quite that good on whirled/ milled holes.

Manual lathes and mills are not capable of micron accuracy in the hands of an idiot like me. ... but they are, in the hands of an experienced machinist, especially one who grew old along with the particular machine in question.

If your operation has its own machining shop, you could learn a lot by spending an occasional hour or day looking over the machinists' shoulders.
Do not touch _anything_ without permission.
Be polite and respectful.
Bring donuts and coffee. ... but don't bring them near the machinery.



Mike Halloran
Pembroke Pines, FL, USA

 

[berkshire]

SpaciouS
At the accuracy you are asking for, the most common method is to turn to close size, then grind to finish size. or hone the bore.
Or use a jig bore.
B.E.

You are judged not by what you know, but by what you can do.

 

[SpaciouS]

Great.

Thanks everyone!

Thanks,
Sean

 

[racookpe1978]

First question: Can you get them made that accurately? Yes.

Next question: For a LOT more money and much more time and by many fewer machine shops.

Next question: Do you really, really, really need them made that accurately?

 

[ornerynorsk]

0.00004 is going to be extremely difficult by any method other than abrasive finishing (honing, lapping, jig grinding) in my opinion. I toured a factory years ago that makes power steering pumps and the like, they are doing precisely what your requirment is, and they ended up building their own in-house "diamond sizing" machines, if I remember their moniker correctly. They had invested the better chunk of a million US dollars to achieve production volume capability down to that level of accuracy and repeatability.

Temperature controls of all aspects (shop, machinery, coolant, material, inspection) are going to be critical at those tolerances, as well.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[Viktor]

Can I achieve +/-1um [.00004"] bore size, and +/-5um [.0002"] coaxially from one bore to the next inline, using a milling machine?

No - you can't. Typical spindle runout+tool holder runout+tool runout+...are much greater.

 

[SpaciouS]

Viktor,

Q: "Can I achieve +/-1um [.00004"] bore size, and +/-5um [.0002"] coaxially from one bore to the next inline, using a milling machine?"

A&Q: My guess is that you cannot, but can you get there, or close, using some type of lathe. Radially symmetric features can generally be made more precisely on a lathe than from a milling machine (I believe), but by how much? What's the factor between the two? And what are typically the tights lathe tolerances when operated by a qualified machinist?

If I could get some kind of "best" baseline numbers, I can contiue to investigate how to get to those final numbers, such as with honing etc.

Thanks,

Thanks,
Sean

 

[Viktor]

SpaciouS

"Radially symmetric features can generally be made more precisely on a lathe than from a milling machine (I believe)" - believe sounds as church activities whereas guess relates to Las Vegas. It does not matter what type of machine we are talking about. Rather, it is a mater of spindle accuracy and rigidity, tool holding and quality of tool itself.

As I mentioned, you cannot get better results as the stack of you runouts from the spindle, fixture, tool holder and tool - it's just physically impossible. The best spindle you can have today has 0.5 microns ronout, the best tool holder - 3 microns (say the latest and greatest Tendo by SCHUNK), tool (the best reamers ground to 4 microns runout...now you do simple math. Note that I did not add anything related to workholding - it is normally 5 microns.

However, for a given setup you can probably achieve +/1 3 microns tolerance if you willing to spend tons of money on high-precision machining components. Less than that is a nice dream.

Viktor

http://viktorastakhov.tripod.com

 

[SpaciouS]

Viktor,

That's what I was looking for. I just need some baseline numbers to put into my trade table and as I delve deeper into my research I will adjust those numbers accordingly. Cost is usually a big factor, but not for what I'm working right now. I'm working on an extremely complex and precise optical system and some of the expectations for packaging, accuracy and stability are near impossible. So I'm working on a concept that has not been done, but if it can be made, it could solve the impossible. That being said, cost concerns are taking a back seat.

Thanks again!

Thanks,
Sean

 

[Viktor]

SpaciouS (Mechanical)

If cost is not a big factor than you have another game. First of all, select a good tool maker who has high-precision grinding machine and Zoller genius 3 (or Walter Helichek Pro); make sure that they grind the tool in the holder as a monoblock and then check the resultant runout on Zoller with a detailed inspection report: make sure that they use the application-specific tool material not chip on-shelf junk and so on. Anything can be done - it is just a matter of cost, time and understanding what you are doing..Good luck anyway.