Is manual milling really a prerequisite for CNC work? (Warning: long and babbly)

[beanbag]

I used to think "yes", based on what I've seen a lot of "CNC only" people do. Where I do my machining, there is a Tormach machine that people are able to use if they take a brief class on it. Since there is a manual mill next to it, sometimes they will do a quick cut on that instead. I once had a person lean over and ask me "Don't you think this end mill is cutting poorly?" and I said "yes, if you spin the bit backwards." Other times, I see people just using totally the wrong speeds and feeds for what they are cutting. Lots of folks these days jump straight to buying or making or using CNC machines as a project or coz it's cool, but don't seem to really know what is going on.

As a brief personal history, I've done a few years worth of manual milling, but it was never with an eye towards doing it properly. It was always some ancillary task for some larger project, and the only reason I would mill was because I needed bracket A to fit slot B, or because the professional machine shop had too long of a lead time. I never thought about rpm=cs x 4 / d, or anything else and went thru end mills pretty quickly. (I wasn't paying for them.)

When I left that environment and now started having to pay for my own end mills, suddenly I started caring a lot more.
Now I haven't done CNC yet (plan to soon), but all of the "proper" machining practices I've learned recently seem to be CNC related, like climb vs falling cut, feed rate, importance of no backlash, etc.

I would have thought that somebody who had proper CNC training were taught these things from the start, as they are fundamental to expeditious metal removal. And they don't worry about stuff like backlash, so of course climb cutting makes sense.

Sometimes I feel like manual machining is like driving an old stick shift car, where you manually have to adjust the timing, or playing around with the stupid XF86config files (for your linux dorks). It's kind of like, sure you get a better "feel" for what is going on, but who cares, it doesn't work as well as the modern equivalent. And some things are better left to the dustbin of history.

I'm just a young punk (in terms of machining), so maybe some older and wiser folks have a better perspective.

[Doc Nickel]

This is a loaded subject, but in my opinion, yes. The CNC machinist (as opposed to just a part-loader/button-pusher) should have at least some hands-on experience with manual machining. He or she doesn't necessarily need to be an expert or a Master Machinist, but I think should at least have enough experience to be able to both see and feel the difference between climb and conventional milling, to see how dull tooling affects the cut, and to have a few hopefully-educational mistakes.

There's a big difference- at least for most people- between the theoretical concept and the actual implementation. The old example is reading a book on how to ride a bike. You may know how, conceptually, but 'til you actually do it, it's just an abstract.

Yes, I'm aware there's lots of highly-skilled CNC machinists who have rarely, if ever, cranked a manual handle. I'm just saying that, in my opinion, the manual experience would likely be very helpful.

Doc.

[John Stevenson]

At first I would have said yes but now I have changed to no.

I'll explain in depth later, gotta make Tim leech's coffee

.

[oldtiffie]

The OP heading is: "Is manual milling really a prerequisite for CNC work?".

I'd have changed "milling" to "machining".

I am firmly in the "no" camp.

There's a big difference- at least for most people- between the theoretical concept and the actual implementation. The old example is reading a book on how to ride a bike. You may know how, conceptually, but 'til you actually do it, it's just an abstract.

I agree with Doc - up to a point.

Once the concept is grasped the rest is pretty well tedium and more of the same.

"Learn by doing" is all very well so long as it does not become a "monkey see - monkey do" type of rote learning.

That is the "if I do this then that happens/happened" way without knowing or needing to know why something did or didn't happen.

I can "feel" a machine by putting my hand on the table, quill or carriage when it is under load. I don't need to use the hand-wheel/s if needs be. Same applies to CNC-ing.

A life-time of machining is not and should not be necessary to learn all that is needed.

I've seen some 15 and 16 year-old kids come out of their 3 or 4 year mandatory pre-Apprenticeship trade trade training that "hit the road running" when they started their 5 year Apprenticeship/s - with more Trade School. Many were up to Ist. Class Machinist standard by the time they were 18 or 20 or so. Many had "gone on" to do specialist tool-making and job-planning/scheduling, "Inspection"/Metrology etc.

All of those Apprentices had to have a pretty good theoretical and applied maths, physics etc. and Shop Practice back-ground that they used to good effect.

Copy lathes, capstan and turret lathes, gang-milling, gear shapers and hobbers etc. and presses and dies etc. were pretty boring once they had been set to work and proved.

CNC-ing whether direct/cold coded or coded from a cad/cam requires a great deal of good creative analysis and application as well as a very sound knowledge of machine and code - particularly when "proving" and/or "de-bugging" or "tweaking".

Non-CNC is pretty well all mono/uni axis work in a single plane where-as CNC can do all of that as well as work in 3 or 4 - or more axis - simultaneously.

I think that a good CNC machinist will beat a good manual machinist over time.

[lazlo]

Is manual milling really a prerequisite for CNC work?

No. As frequently pointed out on PracticalMachinist, most CNC operators are button pushers who would't know which end of a manual mill to drive.

[Rusty Marlin]

No. And I base this on my own expirance. I was hired as an Applications Engineer by a VMC manufactuer right out of engineering school. And learned CNC programing, set up and operation by doing and with a lot of help from my boss. I had no milling expirance before I started, but in 4 years of nothing but CNC VMC work I could make a machine standup and dance! Full macros programing with feedback from a Renisha probe, creative setups and all the tooling my little heart could desire, what a way to learn!

I now run manual machines at home, because that's what I can afford. If I could afford a CNC VMC, min 15 hp and CAT or BT40 tooling and a 4th axis I would have one.

The CNC machining helped tremendously with going "backward" to manual machines, just as I belive that if I had been a manual machinist first that would have help with going to the world of CNC's. I was also spoiled rotten having 35 HP cat 50 toys at my disposal.

The only downside I've encountered is the need for more creativity when setting up odd shapped parts in a manual machine, especially when one doesn't have an XY rotary table. The CNC machine makes cutting parts EASY and a mchinist can lose a little of the inventive spark because programing in complicated countours is a snap, vs setting them up on a manual machine.

[J Tiers]

You folks will have a long and interesting argument about it, but with machining, as well as with so many other types of work, the answer is ALWAYS that YES you DO need to do the thing the simple way before you graduate to automation. At least to programming. And the work may be simply a basic "exposure to" with a few basic projects completed. For button pushing, it should be not in the least required.

It isn't even a serious question.

In engineering (USA engineering, not UK type "engineering"*), it is often much easier to do a simulation on the computer instead of running through a long (perhaps weeks of work) manual calculation. You tend to get the answer in minutes or hours instead of days or weeks. And that is if the solution is even practical by hand methods.

So, why do the engineering courses still have the students run through the methods of "manual" solution etc?

because the modeling is never perfect, the problems are usually somewhat unique, and without an understanding of the methods, and more importantly, a decent prediction of the probable results (at least qualitative results) the use of simulation leads to mistaken results.

Long ago some psychologists had people use a calculator that gave intentionally wrong results. Some folks accepted the results as-is. A few, who knew what the answer generally ought to be, noticed and handed back the "defective" unit.

The same general concept should work for CNC. At least until the distilled knowledge of the centuries is put into the interface program, and the machine mentions the problems to the operator/programmer.

* In the UK engineering seems to mean mechanic's and machinist's work

[oldtiffie]

Originally Posted by beanbag
Is manual milling really a prerequisite for CNC work?

No. As frequently pointed out on PracticalMachinist, most CNC operators are button pushers who would't know which end of a manual mill to drive.

And there you have it.

A lot of "Machinists" are pretty well process-workers who do the same job/s or variations of it - day in and day out.

Same old "Start" and "Stop" button-pushing with some other mundane auto-feed and repetition work in between.

They are as much machine attendants or machine janitors as some who only load/unload a single or several automatic/CNC/NC machines.

Not all "non-CNC" machinists can do it all. Some can do some of it very well and some cannot do much at all very well.

[Carld]

Hmmm, I would say that the requirements for each, CNC or manual are different. I don't run or know how to run a CNC but in the shops I have worked that had CNC machines there was a problem. If a worker was trained in CNC and tried to use a manual mill they would use to fast a feed and rpm for the cutter in a manual application. I watched as one turned an endmill red and snap off from to much feed and speed and then asked me, "what is wrong with this mill", and I said, "it's not the mill it's you". He didn't have a clue as to what to run the cutters at on manual machines. He stayed on the CNC from then on.

I think it would be easier for a manual machinist to go CNC than a CNC machinist to go manual. For one thing on CNC you plan your part machining progression in the program. If your doing it manual you have to plan the progression in your mind and maybe write it down or memorize it and most the time I memorized my "program to make the part" where the CNC has a memory to store the "program".

As said, the CNC can do many things at once where a manual machine seldom if ever works in two axis's at the same time manually.

Another thing is that not all people that run CNC machines can program them. The one big advantage to CNC is you can hire a programmer or two to do the CAD/CAM stuff and then hire operators that are not machinists to run the parts thereby reducing costs. Why would a CNC shop hire a shop full of programmers/operators? I sure wouldn't.

[madman]

With the programs like mastercam that create tool paths tool depths of cuts speeds feeds everything is covered. I think computers are replacing the old school guys who ran conventional equipment. I ran 4 axis cnc machinery for about 20 years and myslef i like the old stuff (with power feeds of course) but i recently added a taig Mill to my home shop collection, I am also thinking of a Novokon 200 mill in ther future if I ever have enough money.
I have to agree with John