Let me start out by going on record as saying I don’t think manual is ever better than CNC except in certain special circumstances that are really not a test of manual vs CNC, but of other factors. I’ll walk through the three special circumstances in a moment.
I’ll be the first to admit it is a controversial viewpoint. Machinists will spend hours debating the whole manual versus CNC thing and every related issue. For example, they will debate whether you can be a decent CNC machinist without having trained extensively on manual machines first. I’m not going to try to answer that one here, but rather, I want to focus on when or whether manual machining is a better answer than CNC for a specific job. As a backdrop, there seems to be a large audience that will say there’s no point in CNC if you only need to make one part and it is a simple part. I disagree, and here is a paraphrasing of the answer I gave on the LinkedIn forum when the same topic came up:
If you know MDI, CNC is just a manual machine with power feeds and DRO’s on all axes. If you argue against the CNC, in many ways you’re also arguing that the DRO or power feed on the manual machine doesn’t help.
In fact, the CNC is better than any manual machine I ever saw because it can do arcs without having to set up a rotary table and it has canned cycles that do all sorts of other things that have to be done manually on a manual machine. If you have conversational CNC on your control or as separate software, it can do even more. For many parts you can just call out the coordinates from the print into the conversational CNC and go. There is no need to go back to the design office and fire up the CAD/CAM. Plus, just think of all the tooling you’ll never need on the CNC–I mentioned the rotab, but taper attachments and all sorts of other things have no purpose with CNC.
You can do all the stuff on the CNC you could on manual, including rework like chasing threads. BTW, the article on chasing threads using CNC was very popular on my CNCCookbook blog.
That’s really the gist of it. You can do all the things on a CNC you can do on a manual machine (plus a lot you can’t), and they are faster on the CNC. When they argue you can’t just jump on a CNC and make a quick part, I think too many machinists are thinking about having to do a CAD drawing, then fire up a complicated CAM program, and finally post a g-code program before they can even get started. But if you know g-code well enough to get by with MDI (Manual Data Input, where you type in an individual g-code and the machine does that one g-code immediately), you really do have a very fancy manual machine with DRO’s and power feeds on all axes as described. BTW, if you don’t know the g-code well enough to use MDI, you’re missing out on all sorts of convenience. Check out our g-code tutorial and you’ll be there in no time.
What I find disqualifies the CNC versus the manual machine a lot of the time are issues that fall into three major categories:
1. The CNC machine is busy
Probably the number one reason to use a manual over a CNC is that the CNC machine is too busy making money on some production run, so you go back to the old manual machine to do the simple jobs or second ops. In this case, the manual machine is free money. Yes, you might do the job better or faster on the CNC (or maybe not if the other two reasons disqualify it), but you can’t even get started because the CNC is busy.
2. The Manual machine is cheaper and we can’t afford Toolroom CNC
Manual is cheaper, there’s no doubt about that. Heck, you’ve probably got an old Le Blonde lathe or Bridgie mill in the back that’s been there forever and was paid for a long time ago.
But putting that aside (after all, that’s hardly the fault of CNC or the manual), I don’t see how you can really do the job faster on the old machines if you also have a brand new CNC toolroom lathe (Haas TL series or a Romi, for example) ready to go.
I recently made an R8 tightening fixture that required an R8 taper. It was a one off (silly things are hard to find in R8 and essential to get a collet chuck properly torqued), and it took almost no time to punch up the taper on a Romi CNC Toolroom lathe and crank it out. Blued up near perfectly and I was very pleased. I could’ve done same on a manual lathe, but there’s no way the Romi wasn’t faster.
And here’s the thing, if you are comparing a low-end CNC to a highly used manual machine, the CNC is much more affordable than you may think. For example, a Tormach CNC Mill is a much closer match to an old Brideport than a full sized and much more expensive Haas. Check out our article on How much a CNC Machine Costs to get an idea.
3. My CNC doesn’t have the right options to compete with my manual machine
This one comes up a lot too. The easiest way to think about it is to consider the average production slant bed CNC lathe versus a proper toolroom lathe. Your slant bed is a little gang tool or other production oriented machine. It doesn’t have as big a spindle bore as your manual lathe or maybe there’s no 4-jaw chuck and no tailstock. That’s not a CNC problem, that’s a choice of machine problem. You can buy CNC Toolroom lathes that are set up the same as any manual lathe and they’re a wonder to use.
4. I have to fiddle with tests to get this part right
I can see where a manual machinist who is used to creeping up on some tolerance (like a press fit for a bearing) thinks it will be hard with a CNC. But guess what? We often creep up in CNC too, and the CNC is actually set up to help with the process. Most CNC lathes have elegant offsets that can be applied to hold tight tolerances that even allow for things like tool insert wear.
One of the reasons I got into CNC was I had a good machinist friend who encouraged me to put all of my manual machining activity aside and focus on getting my CNC mill up and running. His argument was that I would be so much more productive with the CNC that it was the most important productivity thing I could do for my shop. It didn’t take me long after getting the mill up and running to realize that he was exactly right. The most compelling argument for me in favor of manual is the cost argument. That, and I’ve got a bit of nostalgia for amazing old manual machines like the Monarch 10EE I have pictured. But if budget was no object, I’d set up a shop entirely with CNC machines. In fact, even if budget was an object, depending on the type of work I needed to do and the volume, CNC would still be my first choice.
Gentle readers, I’d love to hear your stories. Is there a case where manual beats the heck out of CNC that doesn’t fit one of my 3 categories? Tell us more about that case so we can understand. Maybe there’s more to the manual than power feeds and DROs on all axes of a CNC can satisfy.
Like what you read on CNCCookbook?
Join 100,000+ CNC'ers! Get our latest blog posts delivered straight to your email inbox once a week for free. Plus, we’ll give you access to some great CNC reference materials including:
- Our Big List of over 200 CNC Tips and Techniques
- Our Free GCode Programming Basics Course
- And more!
Just enter your name and email address below:
100% Privacy: We will never Spam you!
Bob is responsible for the development and implementation of the popular G-Wizard CNC Software. Bob is also the founder of CNCCookbook, the largest CNC-related blog on the Internet.
I’m retrofitting a circa 1990 Acra CNC mill with Sanyo-Denki closed loop steppers and drives that have 4000 P/R encoders. I expect its performance to be plenty better than the old Crusader M system with linear scales (which will be used in addition to the rotary encoders) and brush type DC servos that are less powerful than the steppers.
To start with there’s one part I will be making that is impossible to do on a plain manual mill, even with DROs. A manual operator would have to be an etch-a-sketch master to make this part, and do it time after time.
I imagine the molds used to make the original die cast part in the 1940’s were likely made with a hydraulic tracer mill but the patterns such mills traced were cut and filed by hand by master pattern makers.
The CNC cut 7075 aluminum replacements for the die cast originals will be superior in every way to the original, except in nostalgia.
On the other hand I have a Montgomery Ward 10″ lathe made in 1940 which I use for basic plain turning when I just need to turn something to fit. It’s simple and quick to chuck a piece into the Bison 3 jaw and carefully turn or bore until part A fits part B. With a carriage stop to set for the end there’s no possibility of a programming error crashing into the chuck. It’s just as easy to cut off too much with a CNC lathe as it is a manual one!
You missed a few categories. OK, they’re maybe mostly aesthetic, but they’re genuine.
1. Turning the cranks on a manual machine is a lot of fun. I have seen young men stand in line patiently for a turn at the old machines.
2. Turning the cranks is good exercise. I have been recovering from a severe lung problem, and the manual machine was a good start on the road back.
2. The manual machine represents a piece of history. If you’re making a reproduction or a model, there’s a kind of authenticity and a deep understanding of why things are the way they are you can achieve using original methods that you can’t have if you do it with the automatic tool. Ask the Woodwright.
I have a Tormach, and I use it to get things done. Love it, wouldn’t want to be without it. The Bridgeport and the heavy ten have their places in a crowded garage for other, very different reasons.
People think manual machines are so slow because no one knows how to set them up to make them quick. There are so many simple tools to make that can instantly be used on to the machines to make simple jobs quicker. I make no claim that I can produce 10 of the same little complex shafts, or machine 5 little bearing housings quicker than a CNC, I never could, and no one ever will. I also make no claim to ever compete with the radiuses and curves that can be produced nowadays with sweeps and ruled surfaces ( Australian terms for 3D milling lol ) But, I could make simple jigs, shafts, pins, bushes and repair jobs faster than a CNC will on a WELL SET UP manual machine. Most manual machines in most shops are treated like shit. No tooling, nothing set up, machine is trashed, cant turn straight, slides all worn beyond repair. But if you step into my shop, or my old manual one off/prototype/ repair area at my work ( a CNC production shop ) then you’ll see the ways I set them up. A few things I do to my machines:
-LATHE. Obvious, quick change toolpost with at least 10 holders, all set up rigidly and tightly so they don’t move and ness with your centre height. Set them to centre height, tighten them as tight as possible and engrave ROUGH, FINISH, BORE etc on them and don’t touch centre height. Rough turn, finish turn, chamfer, thread, part off, groove, knurl, and a tool to hold 5/16th HSS bits. Also I make up my own design of boring bar holders. Very similar to a CNC lathe boring/drilling pod. Standard 32mm or 40mm bore and sleeves to slip in and adapt to the size of standard boring bars. ( 8,10,12,16,20,25,32 and 40mm for our metric country ) The larger bored sleeves ( 16mm and up have a slot running down them for the grub screws in the pod like holder to lock to the bar. The smaller ones have their own grub screws intergrated to lock the bar. The way it is orientated insures the bars are all exactly on centre height, and takes about 15 seconds to change out to a different bar. I would like to have 3 of these pod things set ( only have one at the moment.) Chuck spiders for getting faces to run true quickly with seconds ops. Centre drill holders, standard centre, reduced ( slender centre ) pipe centre, two drill chucks all set up right next to the machine. All corresponding to the tailstock taper so you dont have to change sleeves. A good smooth 3 jaw chuck that runs true, a nice smooth 4 jaw ( also have two dial indicators at the machine or atleast with in eyeshot ) a spindle mountable collet chuck, equipment to turn between centres ( you never know ) and a faceplate. Bore polishing sticks ( to wrap emery paper around ), ALL the spanners and allen keys ( aussie term for hex keys ) you need for EVERYTHING on the machine you need to adjust, ALL the torx keys and allen keys for ALL the indexable tools that are standard on the machine. A simple handle to make tailstock drilling easier. Work light properly installed and positioned, good quiet coolant pump and clean coolant. Of course a digital readout ( should be first in this list! )
MILL- READOUT, you can get away without one on a lathe but forget it on a mill! Get as many toolholders as possible ( don’t waste money though, you don’t need 10 ER40 collet chucks on an R8 light duty Bridgeport ) but don’t cheap out. A few good different sized facemills with good quality inserts, 2 drill chucks, a few collet chucks ( on a bridgy an ER40 and maybe two ER20’s ) with collets at the machine and the wrench with it! A 1″ plain stub arbor with collars for your slitting saws etc, taper adaptors ( one MT2 and an MT3, go bigger on a bigger machine of course ) a fly cutter or two, a nice boring head with good boring bars, I even made an arbor to accept the old HSS shellmills that I have a ton of ( got them for free ) and I use them on aluminium. Even make an arbor to hold your dial indicators ( the interapids or whatever is popular in America, here in Australia, mitutoyo test indicators are very popular!
Key your dividing head so it is aligned when you put it on the table every time, same as its tailstock. IMO you don’t need to do it to a vice, because a vice can be bolted down and aligned in 3 minutes and you wont be wondering, is my vice really lined up. Have dedicated bolts to hold down your vice, dividing head and tailstock, three jaw chuck, rotary table and any other jigs you may have. Tee nuts are cheap, machinists are not so have heaps on hand. A good clamp kit or two for the table AND rotary table and also the wrenches for them, and the vice spanner, and chuck keys for the dividing head and three jaw. Have a few 123 blocks at the machine ( not popluar here, we just use ground blocks, if your lucky enough for the shop to have them haha ) a honing stone, set of paralells. Also every wrench and allen key you need for the machine and tooling. Have a spot drill at each mill too I guess would be an idea too ( might have to buy a couple now! ) I think on a mill you can get away with one mag base and just a test indicator, but if you do any repair work you’ll probably need a standard dial, while your at it, get a separate base for it!
SLOTTER: A good selection of bars and HSS bits, like, as many as you can make, buy, find whatever. Make up sleeves for them to adapt them all to a standard size ( I used 32mm ) Then mill up a toolblock that bolts on to the slotter ram which the sleeves bolt in to, much like the pod system on the lathe I mentioned above!. You know have a quick change slotter!
A clamp kit, big paralells, all the wrenches etc, and two dial indicators for the X and Y movements ( or get a readout, you don’t need one but could be nice to put one on I guess )
A nice big rotary table ( unless your table is a built in rotary table ) and get a three jaw for it. Have some nice cutting oil in a squirt bottle ( have one on the lathe too for screwcutting, and while we are at it, on for the mill for tapping, why the hell not! )
I have probable missed heaps, and I hav’nt even gone into surface/cylindrical grinders yet, but I don’t have them yet, but when I do you can be sure ill set them up just as well. This makes me quicker for one offs and it works for me.
Zak, I still don’t see anything you’re suggesting that can’t also be done on a CNC. I run a QCTP just as you describe on mine with all the tools on center. So what makes your manual faster?
When designing and building prototypes, it is sometimes easier to drill, mill, and fit manually (designing on the fly) without having to measure setup, and program.
Richard, it may seem odd, but you can operate on the CNC without having to measure, setup, or program. Just treat it as a manual machine with DRO’s and power feeds on all axes. Whatever you can do on the manual, you can also do on the CNC in that respect and there are many more things the CNC can do that a manual won’t.
Hope you are still reading these posts.
I think the manual / CNC line has been pretty well blended now, especially when you factor in conversational toolroom machines.
Sometimes I think surface grinding may be the final bastion of manual dominance, especially for 1 off pieces, or stamping die component repair/ sharpening.
I do a fair amount of valve body machining on unbalanced, asymmetrical cast valves, 100 to 1000 lbs. Even with counterweights I would not want to subject a nice CNC machine to such abuse, and though I have had to run Prototrak lathes with cast 4 jaw Chuck’s mounted on them, I was queasy about it even with the max spindle speed set to a safe limit.
Once in a while, I run into IF work where it is easier to reach a spot with my boring bar set to an angle rather than tooling my way out of it, and a lot of my business is emergency repair for power plants, so I’d call it an outlier niche.
In closing, I think you are on target for about 98 percent of the situation, maybe 99 depending on programmer proficiency. That being said, best to remain open and flexible, take the best route you can in the current environment you find yourself in.
There’s also the matter of accuracy and precision, on a manual the accuracy and precision is dependent on the mesuraing tools the machinist uses, on a cnc it’s dependent on hardware, so take for example a part that requires two holes spaced within .00005 inches of nominal and are 30 inches apart, an ordany cnc set up will have divisions of an inch down to .0005-.0001 well we now need less then half that, on a manual we can just use or order an gauge block and a dti of the necessary precision and accuracy, on a cnc mill we now have to either upgrade or send the part out, assuming any one has a mill that tight, most mill’s that size are definitely not. Then there is times when reactive machining is required, such as placing tight toleranced chamfers on material that has a too much variance from part to part, to do so on a cnc will require special fixtures and a probe cycle, on a drill press it takes a depth stop and get’s the job done almost as fast, and as a bonus the chamfer tool will drag the part into alignment. Then there’s one offs, and simple corrections where just setting up the part leaves it it ready for its only operation, at that point there’s no need to consume a cnc Mills time, it’s bad economy to wast that time on a cnc, however it’s right up a manuals streagths. Also there are times when geometry is more important then dimensions such as common with scientific instruments, while you may get by with a cnc for a while, having gear driven jigs will allow the wear of the machine to improve accuracy of the form, and diminish the dimensional accuracy as opposed to cnc where the wear of a machine just diminishes overall accuracy including the form. All this being said, it has been proven time and time again that a hybrid approch, where a skilled machinist has access to the lead screws of the machine as well as a cnc is the most versatile and cost effective method, to diminish or ignore one method over the other is a great way to lose money.
James, color me VERY skeptical. Just the first example where we’re going to be better off hitting half a tenth tolerances with the manual leaves me scratching my head. The CNC may very well have thermal compensation and laser mapped leadscrews that tell the machine down to such tolerances exactly what’s going on. The manual machine certainly does not. At best it may have a DRO.
The part that has too much variation to do the tightly toleranced chamfers is another interesting example. I believe you’re admitting the CNC can do it faster with the probe, but not much faster. But why does this part exist? Why didn’t we make a skim pass to clean up the issue before trying to chamfer? Why can’t we set it up so we have both the probe and the cutter operating at once so we can call a probe cycle and the whole thing will act exactly as your depth stop?
You want to avoid wasting expensive CNC time, but I see two trends that argue against that being a priority. First, you can buy a very decent CNC for what it will cost you to buy a decent manual machine. What if instead of the venerable Bridgeport in the back of the shop you instead had a Tormach CNC Mill? Now the CNC time is cheap. Meanwhile, as we compete with the labor costs in China versus here, we see the real expense is the operator. How can we minimize that cost? Why, if we had a CNC it’d be straightforward because of all the things it can do to help.
The accuracy argument I believe is also going to go the CNC’s way. You need to employ techniques like what Renishaw uses for its RAMTIC manufacturing:
Very clever stuff for solving down-to-earth problems with tolerances on relatively cheap machinery. Or take even the use of wear offsets on CNC lathes. Yeah, I can sorta kind do things on a manual lathe that’s remotely similar, but it is so much more labor intensive. With the CNC, I can take a test cut, measure unwanted taper, and easily program it to compensate for that taper.
The list goes on. I think the way to lose money is to be unwilling to consider new alternatives.
Bob you may be skeptical, that’s fine, please do explain how an binary optical incoder with a finite resolution is going to compete with a grade aa gauge block mounted directly to the table and indicated with an analog device that has no effective resolution only extreme magnification or better yet use a laser directly? And i have seen RAMTIC manufacturing, the pit fall is that the computer cannot tell anything outside of the divisions of the supplied encoder, humans use analog, and are therefore indefinitely precise depending of course on avaliable magnification. As to shape generation vs interpolation, shape generation jigs as I said, wear into higher comfority, meaning that a ten thousand dollar machine can run longer and faster then a hundred thousand cnc, but that tenthousand machine can only make one shape, it’s a trade off that make sense in certain production machines, when was the last time making a standard screw was cheaper then buying it? A helix is a common example, another is your surface plate, telescope lenses etc etc. It’s expecially when dimensions don’t matter just speed and geometry that cnc starts to lose, but if a critical eye towards the short coming’s of cnc is generally cast, then these shortcomings can be overcame in the next generation,(I have high hopes for STEP control to do a lot of this) if the short comings are not addressed then we as a whole waste more money to get less. PS when was the last time you saw an excavator bucket loaded in a cnc to have its bearing’s trued and line boared? Also, the varying thickness is an issue in sheet metal, higher tolerance stuff like aerospace mixed with low tolorance stuff like sheet metal where the thin stock can vary quite a lot, and skimming down the parts doest work out, a face mill will cause deformations if its ran fast enough to beat manual, and a fly cutter goes to slow to beat the manual (we tried the fastest was actually a laser cutter then manually ream and countersink because of loading times)… There are just times when the process is too simple for cnc, and too awkward for enclosed. Lesson and point? Buy a good manual, one of the hybrid cnc ones, it will expand your shops abilities up to your machinists abilities. Plus the conversational controls are mostly more intuitive.
Analog is skilled art. My father was an Analog Aerospace Machinist. Some of his work went to the Moon. It is going to break my heart to convert his personal Clausing Lathe to CNC.
There are always several ways to do a single job. In the end, it comes down to a full understanding of the machines at hand and having a full understanding of what is being made. After that the discussion is academic.
I wish I had the luxury to stay Analog in my machine shop. I also wish I could grow my own organic food. But at the end of the day, someone has to feed the cat, wash the dishes and cook the store bought potatoes. Time is also needed for the business end of the shop, to balance books, to do the design work, and to do the regular reading to continue growing and keep my mind from atrophying.
My Father had a whole support team behind him to machine those parts. Unfortunately he never acknowledged his support team.
CNC will allow me to be the support team while CNC does the machining.
When I get to the point where I am building corner cube reflectors to place on the Moon, then my shop will be at a point where it will be profitable enough to hire the support team to do the needed analog work. … Optics is a whole other fascinating field ….
Analog is definitely a beautiful art, I started my apprenticeship under my father with analog too and moved to a aerospace shop for the past few years, no machines older then five years and everything start’s on a cnc and half of it still ends on manual, but analog and cnc don’t need to be exclusive, they just are for reasons that are beyond me, they’ll include features in a mill or lathe that %99 of the time will never be used yet they make no allowances for analog, it’s a shame; but if your converting why not use through motor spindles on your servos and attach handwheels? They also sell tooling turrets that bolt right into the cross slide, allowing the machine to be used fully manual/analog when needed plus you can switch without removing the part! It seems to be a good comprise, at least in my experience. FYI Optics are a hobby of mine, it’s always amazing to see a simple jig with no real means of measure built in turn out a functionally perfect lense or mirror.
James it seems your accuracy argument rests almost entirely on the perceived limitations of optical encoders to measure position as well as the perceived limitation of how small a move the machine may be commanded to make.
Let’s look at those particular issues more carefully. First the encoders. Bottom of the line encoders will have 4096 positions per revolution. But, the question then becomes, how far does the axis move in a revolution and therefore what dimensions does that division correspond to? That varies from one machine to the next.
Let’s try the low end of the market–Haas. They quote 2.5 microns accuracy and repeatability–0.000039″. Already it’s capable of your earlier example of half a tenth from that perspective. How about a higher end machine? A garden variety CMM is a CNC that’s twice as accurate.
Position to the standards of the measurement is another issue. BTW, most CNC’s can position more accurately just by switching them to metric. They go from 0.0001″ to 0.001mm which is 2.54 times smaller.
Putting a handwheel on the servo doesn’t really help. You can lay on all the AA Gage blocks you want (and you will need a relative measure like that as you won’t be able to see the divisions on the handwheel precisely enough to matter), but unless you’ve parked your mill in a carefully temperature controlled environment or taken other steps to compensate, just breathing on it wrong will change dimensions at these scales.
Is the analog human really capable of infinite resolution? No, not on any kind of productive basis at least, AA gage blocks or not. Can they do it in a carefully controlled situation with lots of experience and just the right equipment? Sure.
But if we have to have a specialized situation (grinding mirrors or line boring excavator buckets), my work here is done. Forget the old Bridgeport or Clausing in the corner. Get a modern low end CNC mill and lathe for the toolroom. You’ll be more productive and it’ll pay for itself in no time.
Bob, with you’re hass example you are mixing resolution with precision, there is an actual gap in the rotation that the computer cannot perceive, it can only tell when it has began and ended, and can halt the rotation at the start or end of said gap, that is why even cheap machines can have such a tight accuracy, but they still cannot correct tighter then their native resolution. Second you’ve made a grave assumption, these are limits of my (former) company’s custom built .5 to .8 million dollar 60×30 inch machines, literally the best money can buy and the latest greatest tech, however when you scale up the positioning you end up scaling up the error, and then you start into the comprises, this tech will hold your tolerances better but you can forget hogging steel, or this that will still repeat better and tackle HRSA’s but it cannot provide smooth interpolation smaller then or faster then etc etc. Get critical. Analog can do what digital cannot, yet there are plenty of tools that convert analog to digital, if you happen to be an car or airplane mechanic that is, not so much for our machines, even just last year when our latest machine arrived we could not rely on the cnc, the best that Renishaw and Japan had to offer, we still, in production, had 50% or our components relying on a manual process. I gave you the reasons a top rated, top of the line cnc job shop still used manual, why would keep trying to argue against the simple science that led us there? Why not ask questions and open yourself to a different veiw? Our machinists all had years of specific training, and engineering degrees! And our shop attracted a growing number of industry’s our bids would often compete with over seas, with a excellent profit no less! And if you’d like the data for a production setting using analog and manual, I’ll tell you this much, we seated them in the manual while the cnc ran the lower toleranced features, and our reject rate was 1 in 10’000. Over ten years, all ten of the rejects came from one man with a drinking problem. And even then it was not because he was drunk, but where he set down his drink.
James, what I’ve said about accuracy and repeatability stands–they’re the numbers Haas quotes. You are correct that we cannot measure between the encoder pulses, but it’s a red herring–the figures include that consideration. There is no confusion with resolution, as Haas is stating their spec for accuracy and repeatability. The actual resolution is even more accurate to achieve those numbers.
What’s the name of the Job Shop you are referring to that did so much manual machining? Surely if the results are so miraculous, someone must have done a case study on it?
Encoder pulses? That’s an interesting subject. The resolution of the machine can infact be lower then the accuracy and precision and a grey code, or absolute encoder will be, because of both the Encoder pulse and the torque (stopping power) of the servo. It stops at the pulse, the line, in a optical encoder it’s a very well defined line, that is much thicker then the stop zone(in testing circumstances), it’s not the entirety of the lines that the make up the stop zone(an incremental however has a tendency to skip a count now and then so it does need more divisions then it can actively utilize), it’s also worth mention that accuracy and precision numbers are only achieved at similar velocity and mass ratio as the test, and the velocity is very low, and the mass is the table, no parts or vices etc and that brings up problems of true motion with curves although most angular issues have been solved-ish. I went independent of my former shop, to produce my own line of survival gear, my funding fell out I have been reliant on work that they have thrown my way, rather than risk P.O. ING the owner by brining in his shop to an discussion about the short comings of current cnc, I’ll let him decide whether or not to post, but that is in your words a red herring, as this is a discussion about the merits of manual machining, or more appropriately when the limits of current positional tech exist and or rather when manual is the best option. It’s not a back handed thing to say cnc has its limits, repair shops are just one example of an entire industry that would do better without it more often then with, although hybrid cnc/manuals is the repair shop credo. It’s a fine science that requires you to take the fan boy that’s squealing in delight in all of our hearts and tell him to shut it, the hard way is better here or there, sadly. I hate manual milling. With a passion, but I have pushed cncs past its limits, and cost a lot of man hour’s and money because of it.
“I differ with you on this point. I am sure there are many benefits of using a CNC machine such as:-
They can be used 24 by 7, only needs to be switched off for occasional maintenance.
Less skilled/trained people can operate such kind of machineries without much hassle.
Unlike manual machines, CNC can be updated by improving the software.
And last but certainly not the least, the cost is slowly coming down.
How about this . jobs for american boys and girls that dont require degrees or certificates in machine tool.. Sure, not getting parts out the door faster or making complex parts but dammit jim…..what are people going to do with their hands!? Everyone thinks about the bottom line but noone thinks about giving their fellow americans jobs anymore. A quick buck. Sure, someone has to write code and push buttons, but juat not as many people. We put men on the moon with simple tools, but the machines are winning
Joe, I fear if we decide to subsidize jobs that are too uncompetitive, things will not end well.
The truth is, we can do both. We can create jobs and we can automate. The amount of training required to run a CNC is not that huge. We could get a lot of folks started with some simple vocational programs. Heck, a lot of folks start without even that. We live in an age where you can learn an awful lot online.
The machines are not the enemy. Policies that make it easy for companies to ship the jobs overseas to escape taxes and fatten their coffers are the enemy.