Industry accuracy for manual machining.

[Peri]

Hi,

For the last 8 years I've worked in education, and I have access to the colleges lathes, mills and cnc shops (Bridgeports, Colchester/Harrisons and XYZ SMX and SLX).

I spend every free moment using and learning these machines, and I'm.......not as bad as I once was I've made a few projects like the engine that's pictured.

I'd like a career change, and I've been browsing around for a machinists job. Many of the jobs I see state "Must be able to work to 0.02mm ( 0.0007") tolerance"

Is that a standard thing? Some people I talk to say yes, others say "Anything needing that level of precision would be machined over size and ground to final spec".

I've tried working to 0.02 - I know I can't do it consistently !

Thanks,

 

[William Payne]

In my world that is a standard everyday tolerance and done manually very often.

 

[Ralph_P]

Stay in education.

 

[Peter from Holland]

The machine you do it on makes a big difference in how easy or difficult that might be

Peter

 

[Overland]

I would suggest a company expecting that level of accuracy from their employees would have equipment capable. So don't let that put you off, go talk to them. You'll gain a lot from an interview where you can discuss your skill level and experience with their expectations. If they like you they may offer some training.
There's a lot of openings out there, very often the most important things to a prospective employer are attitude, reliability and willingness to learn.
Good luck,
Bob

 

[jim rozen]

The simple fact that you understand what that tolerance spec means, indicates you are a desireable employee for that job. Go for it!

 

[jscpm]

Machining is a skill. One of the symptoms of unskilled workers is that their parts are the wrong size. So, the "requirement" you are reading can be interpreted as meaning: this is a skilled job.

Of course, anybody can produce a precise part given enough time. What matters is not only being able to make precise parts, but make them quickly.

 

[Kingbob]

I would suggest a company expecting that level of accuracy from their employees would have equipment capable. So don't let that put you off, go talk to them. You'll gain a lot from an interview where you can discuss your skill level and experience with their expectations. If they like you they may offer some training.
There's a lot of openings out there, very often the most important things to a prospective employer are attitude, reliability and willingness to learn.
Good luck,
Bob

This 1000% - Doctors have the good sense to call what they do "practicing". If someone smart and tough enough to get through 8 years of medical school and then 5 years of residency calls what they do "practicing"; then I think we humble machinist can and should adopt the same language and mentality about what we do. Every job is practice for the next one.

 

[Gordon Heaton]

.0007" when required isn't bad, and CNC doesn't necessarily mean tight tolerances. No matter who you are, or what your experience/talent is, working to tight tolerances requires more time.

 

[eKretz]

Easily possible. Mostly a matter of good technique over machine quality. Machine quality comes into play more when needing to hold that tolerance over a longer length. Even then (poor quality or condition of machine) a lot can be done with technique, but you may need a more skilled person running the machine. That plays into Peter's comment about how easy or hard it will be to do such work.

The varied comments you've received are probably more an indication of quantity of parts playing a role. For a few parts, not a problem at all. If I had hundreds or the parts were hardened, it would probably be a faster and better quality job if the parts were OD ground. Remember that with any job, the higher the quantity the less an extra setup counts against the job in terms of cost. Setup time is often the most costly part of a job on small parts when quantity is low.

 

[Peri]

Thanks all, very enlightening.

I'm not too bad on the lathe, pretty consistent to less than .04mm.
The mill is another matter - it might not help that all our mills are old Bridgeports (1970's-80's), with probably not the best maintenance - simple blocking up tends to be .02-.03 but adding features means errors add up quickly (I do realize thats a lot to do with my lack of skill).
Ironically, the new (less than 8 years old) XYZ conversational CNC mill we have is probably the worst for accuracy - you can't alter the nod, and the bed runs out 0.02mm across the table.
These are the only machines I've ever used, I kind of assumed they would all be like that

Thank you again,

Steve.

 

[PackardV8]

Much good advice above. Long term, no company can ask for/demand tolerances their machines and machinists can't reliably produce.

Only slightly OT, but a younger estimater underbid/overspecced a job and it wasn't happening. He and the also young supervisor came down hard on the experienced machinist, who said, "I'll stand here off the clock and watch you two meet the units and tolerances you spec. Once I learn how from you, then I'll guarantee to be able to do what you do." Of course, neither the supervisor nor the estimator could even set up the stock and tooling in the machine and the job had to be turned back for rebidding.

FWIW, in production engine rebuilding shops, those with younger, less skilled machinists have to have the newest, tightest tolerance machines.

The shop which does our crankshaft grinding has two old guys who've worn out the machines they run every day. They can still hold whatever tenths they're asked to provide. A new guy, who'd be OK on current technology machines, coming in on those old, worn machines would produce junk for weeks and might never learn how to grind cranks to spec in a production time frame.

jack vines

 

[Richard King]

I am a professional Machine Builder / Rebuilder and when we build machines we hold tolerance per 12". On what I call a Conventional machine category; Knee mill like a Bridgeport, Engine lathe Like a Leblond or South Bend,, low end surface grinders, etc. I hold a tolerance of .0002" / 12". A Precision machine category; CNC VMC & HMC, Super precision lathes like a Monarch EE or Hardinge, Drake thread grinders, Crank Grinders, Studer Cylindrical grinders, Okomota surface grinders, surface plates, straight-edges, etc. I hold .00005" in 12". so if the ways are 48" long I shoot for .0002".

That's on new machines and rebuilt machines. The tight tolerances can only be done in a temperature controlled room. You should buy a book called "The Foundations of Mechanical Accuracy" By Wayne Moore. I have seen experienced machinists hold incredible tight tolerances on worn out machines because they know where the bad places are and work around the worn spots. I was in MN Grinding rebuilding a Myford Grinder (00005/12") and a journeyman grinder machinist was holding millionths on a worn machine . The machine ground a part big in the middle of a 16" shaft and he would put his finger on the shaft and he pushed it in the area that was worn.
I'm not recommending you do that as it's dangerous, but he knew how to make good parts. Also I am friends with the owners of Professional Instruments in MPLS. They make air bearings. They said when they hire a machinist they give him a test. They lay on a table a Starrett Micrometer .001" per graduation, a Micrometer with tenth graduations, a pressure mic, a 3 anvil pressure mic and when they interview new people they hand him a ground bar and ask them to measure it . If he picks up the Starrett mic they don't hire him, if he picks up the 3 anvil pressure mic they consider him.
Then they ask him how he would measure the runout on a spindle bearing outer race? A precision machinist would say on a air bearing rotary tester.

My point being is what are you measuring, where are you measuring it and what are you measuring it with?
Machinists have spec's on their blueprints. There is more I could say, but the others can help now, my fingers are getting cramps..lol

 

[jim rozen]

Again, the OP is asking about the possiblity of a career CHANGE and while he's had experience, he does not have the equipment for, nor has he done, machining to that tolerance level.

1) employers will have whatever measuring and machining equipment to work to the tolerance they require.
2) you'll be able to do this with a minimum of hand-holding if you take a job like this.

 

[GregSY]

It's reasonable to assume they will provide the tools...so it is as much their challenge as yours.

I remember the machinists in our shop complaining that they were given prints with tolerances of .001" when the lathe they had to use had .0015" slop in the spindle. These were old, big lathes and they were turning new, big shafts, usually forgings that were 10' long and 2' diameter, so it wasn't considered OK to scrap one every so often.

 

[Richard King]

I am a professional Machine Builder / Rebuilder and when we build machines we hold tolerance per 12". On what I call a Conventional machine category; Knee mill like a Bridgeport, Engine lathe Like a Leblond or South Bend,, low end surface grinders, etc. I hold a tolerance of .0002" / 12". A Precision machine category; CNC VMC & HMC, Super precision lathes like a Monarch EE or Hardinge, Drake thread grinders, Crank Grinders, Studer Cylindrical grinders, Okomota surface grinders, surface plates, straight-edges, etc. I hold .00005" in 12". so if the ways are 48" long I shoot for .0002".

That's on new machines and rebuilt machines. The tight tolerances can only be done in a temperature controlled room. You should buy a book called "The Foundations of Mechanical Accuracy" By Wayne Moore. I have seen experienced machinists hold incredible tight tolerances on worn out machines because they know where the bad places are and work around the worn spots. I was in MN Grinding rebuilding a Myford Grinder (00005/12") and a journeyman grinder machinist was holding millionths on a worn machine . The machine ground a part big in the middle of a 16" shaft and he would put his finger on the shaft and he pushed it in the area that was worn.
I'm not recommending you do that as it's dangerous, but he knew how to make good parts. Also I am friends with the owners of Professional Instruments in MPLS. They make air bearings. They said when they hire a machinist they give him a test. They lay on a table a Starrett Micrometer .001" per graduation, a Micrometer with tenth graduations, a pressure mic, a 3 anvil pressure mic and when they interview new people they hand him a ground bar and ask them to measure it . If he picks up the Starrett mic they don't hire him, if he picks up the 3 anvil pressure mic they consider him.
Then they ask him how he would measure the runout on a spindle bearing outer race? A precision machinist would say on a air bearing rotary tester.

My point being is what are you measuring, where are you measuring it and what are you measuring it with?
Machinists have spec's on their blueprints. There is more I could say, but the others can help now, my fingers are getting cramps..lol

Thank You Peri for the "LIKE" . I hope I helped you understand. I must have read it differently then Jim. I thought you asked about tolerances and expected spec's.

 

[William Payne]

I should probably add to my earlier post. Just remember .02 won’t be a tolerance for everything on the part. Just important details. I have never had the pleasure of working in a climate controlled shop so I’m only talking from my experiences in my world.

0.02mm is not that bad. I’ve done .01mm and even 0.005mm as well. What I try to do is aim for the top of the tolerance and if need be if I don’t hit size and it’s a bit up it’s nothing some minor polishing can’t get to fall into size.0.02mm would be common in shafting or any sliding fit part. Just have to be careful and don’t trust your readout without verifying first.

Check when you do your finish cuts. If you are worried about wear in the machine using dial gauges to measure you movements is not dumb, I do it often even with a readout.

Just have to be careful.

If you are doing multiple parts in a production environment you set yourself up using your first one so that you can come up with a plan to make repeated in tolerance parts.

 

[plastikdreams]

It doesn't just take the machine being good but you need to know the machine, it's quirks and limitations. My measuring equipment is good to .0005, I can hit 0 with little effort. I check my head tram often and square my vise to 0/0.

 

[Richard King]

When I send a part out to be ground I either set it on 3 points and block around it and mag the blocks down and not the part as if you mag down a warped part and grind it, then release the magnet the part bends back and it is not flat. Or I hand-scrape flat the side I'm going put on the magnetic chuck. Same goes for chucking something in a vise. Those are just some of the "tricks of the trade" old fart machinists know,

 

[William Payne]

When I send a part out to be ground I either set it on 3 points and block around it and mag the blocks down and not the part as if you mag down a warped part and grind it, then release the magnet the part bends back and it is not flat. Or I hand-scrape flat the side I'm going put on the magnetic chuck. Same goes for chucking something in a vise. Those are just some of the "tricks of the trade" old fart machinists know,

One day I would love to attend one of your classes. I sadly feel you will probably have retired by the time there was enough demand to do a class here in New Zealand.