How perfect are tools?

[mmcmdl]

I had just read someone had bought a 5c collet and called it junk because it ran out .001 thousands.

I think it is silly to expect everything to be perfect.

I think it is silly to think a cutting tool has to be a certain degree.

I would hate to be a new guy trying to match these standards.

I think it is silly to tell a new guy trying to learn machining to read a book with out dated info about how to grind a tool. Info about a 100 years old.

To me it is like taking a teen out and saying I am going to teach you how to drive
then taking him out to a model A Ford.

I have to agree on all the above . Machinist's make parts to prints , sketches or verbal instructions , the tools do not . Can't remember EVER checking out runout with a collet . And no , nothing is perfect in a real machinist's world and never will be . Perfection is a whole nother rabbit hole to wonder into . I thought there was a similar post on here lately regarding your topic . Myself as a long time tool maker just laughs it off knowing how things really are .

 

[mmcmdl]

Going overboard with precision when it isn't justified is not productive.

This sums it up nicely ! You're taught form , fit , and function . Keep the customer happy is the bottom line but going overboard costs you and your employer $$$$$ .

 

[Ulma Doctor]

It’s amazing how superior one feels when talking about working to impossible accuracy.

 

[RJSakowski]

It's the difference between machinists who have worked in the trade and hobbyists who are learning from books or Youtube but have yet to be told "Quit f***ing around and get that job done!" by an impatient boss. I see it here frequently, dithering about tenths when ten thousandths is plenty close. And then there's speeds and feeds. I never met a manual machinist who calculated IPM. I believe those online calculators actually prevent people from developing an intuitive feel for what they are doing.

It is only with the recent advent of VFD's for lathes and mills that IPM calculations were even feasible. My Grizzly G0602 had only six choices for spindle speed with incremental increase ranging from 29% to 100% with an average of 76%. The Atlas/Craftsman 6 x 18 fared better with the stock pulleys with the incremental increase ranging from 16% to 56%, averaging 32%. The incremental increase for my RF30 clone ranges from 8% to 103%, averaging 37%.

Power feed rates on the lathes can be adjusted to a finer degree providing you are willing to change gears. This is usually more bother than it is worth. The mill is manual feed only so IPM is meaningless. I generally adjust feed rate by sound and feel. I have done this from the very beginning and continue to do so even though I have the ability to fine tune my feeds on the Tormach and the 602 with variable spindle speed and push button feed rates.. Old habits die hard.

 

[great white]

What I would like to hit for accuracy (at least once) are those air bearing pieces.

Can’t say why I think so, but it just seems cool as heck to me to machine to such a tolerance that the air molecules themselves are the bearings.

Of course, for me, it would be a completely academic exercise I have no need of that level of precision in the things I machine. I just think it would be cool to make one sometime…but I also don’t want to invest that amount of money in machines capabaly of hitting that tolerance. Would be wasted capability %99.999999999999999999999999999999 of the time in my hands...

 

[Shotgun]

When turning anything over a couple inches long, I always hit my measurement within a tenth. . . somewhere along the length.
So many of the machines we have are so worn near the chuck that we're always turn a curve. Most people talk about cutting a taper, but if a measurement was taken every half inch or so, I'd bet most would have curves.

 

[WobblyHand]

I wouldn't bother checking the TIR on a collet made by a reputable manufacturer, but us hobby machinists don't always buy from reputable manufacturers. You'd better believe that I check equipment that comes from distant sources, because sometimes it has to be returned. I've had Morse tapers that were NFG, their tapers were way off and had sub 1% contact.

If you machine for a living, you don't have to deal with this as you don't buy bottom shelf tooling often. But as hobbyists we often have to economize. When you buy lower end stuff you have to do your own QC and incoming inspection. So yeah I check TIR of the collets I bought, just to make sure I didn't get a lemon. 95% of the time it works out ok. It's the 5% stuff that can be troublesome. I'm just looking that the runout isn't awful, or the bearings aren't full of grit, or bearing housings aren't broken, things like that.

 

[Shotgun]

This Old Tony has a video. . . "How much does a cheap collet cost". . . at least, I think that is the name of it. His cheap collet turned out to be very expensive.