Making a few bucks with the CNC

@Jason: been reading some about weight on the Z axis. How do you think a water cooled spindle plays into that? Though I am thinking like you - something to grow into!

I do wonder about the optimim spindle for an entry level machine.
 
Leo's machine has a little of that weight on the Z challenge.

Mine's a 3/4" lead screw that is 6tpi - with a 425oz stepper on it. I think it'd lift me up 7" in the air if i could fasten a seat to it. :p
 
Also this video is real interesting. It is about a hold down table for drag knife work. However if you jump to the end the CNC he is using is really interesting..
https://www.youtube.com/watch?v=fOrkvPF0pro

If you watch the video you can see that this is obviously a cottage business as there are a couple of young uns running around.

The website is but doesn't talk about the machine.

May well be more money is drag knife work than many realize. A fun watch anyway..
 
SPINDLES

My machine has a air cooled 2 hp 220v Perske ER-16 Router Spindle. It is not a hand held router.

I use 1/8 shank and 1/4 shank cutters 90% of the time. I really wish I could use a 1/2 shank.

It is true that a LOT of CNC routers use a hand held router as a spindle. But consider this: Routers are purpose built. A hand held router is purpose build to be used occasionally - not constantly. As an engineer I know that machines are designed around "design intent". A hand held router is a brushed motor and the bearings are not constant service designed. Those are the failure points. The CNC Router spindle is an induction motor (no brushes) with constant use bearings. That is a huge difference.

Can you use a hand held router on you CNC machine - YES - of course you can. IS it the best choice? Not in my book.

In my mind the jury is still out on the Air Cooled vs Water Cooled - but I am leaning towards water cooled - only because of the noise factor.

I do work with industrial machines and I have worked with a CR Onsrud twin spindle twin table 25 hp columbia spindles - they were air cooled. The 15,000 RPM horizontal machining centers I work with are air cooled. There are not many water cooled machine spindles that I see in the industrial - except maybe the Studer S33 precision grinder I just bought that is capable on sub micron tolerances - but that is a different story.

I will select a ER20 Water Cooled spindle when I eventually go to the 48 x 48 machine.

The ER series is an industrial series that is precision to less than .0002 and you can buy extra precision to less than .0001. Certainly more that is needed for most of our applications.

As Jason (beamer) said - I have a counterweight on my machine. My spindle is 15 pounds and I have leadscrew but the design of the machine with a small stepper on the "Z" axis caused lost steps more times than I want to remember - so I counter balanced the "Z" axis. I think better design and stronger stepper would be better than a counter balance.


CUTTERS

This is my world - cutters. It is what I have been doing for 40 years - BUT - in the metal cutting industry.

So Wood is a different material and cutter geometry is a "little" different, but VERY similar to non-ferrous materials like brass and aluminum.

Onsrud is a GREAT cutter - I will take nothing away from Onsrud. I have used them a lot - mostly cutting aluminum.

I may present a bit of a different viewpoint on cutters, but that is because of my background in manufacturing in the metals industry.

I know I have gotten it wrong a bunch of times, but that experience has taught me what is wrong.

**CHIPLOAD**

Chipload is going to be the buzz word when is comes to Cutters and speeds and feeds. That is the biggest question most people have.
The variables here are: material (notice I did not say wood), number of flutes, Cutter material (mostly carbide), tool diameter

I have a spreadsheet to calculate the RPM and feedrate based on chipload. I have a chipload matrix based on tool diameter and material on the spreadsheet. I also have reference to the places I derived the chiploads from - it's not just my opinion. There is also a picture of cutter showing what RPM and FEED and # flutes means. The spreadsheet works - but I want to do more to it to have more cross section to chipload and to make it more user friendly and understandable.

If anyone wants the spreadsheet I will email it but please know - it is not finished.

I wish I could post a spreadsheet here - but that does not seem possible.

**BRANDS**
There are a lot of brands out there. I don' t like and I don't use "cheap". I like great quality but I don't want to pay more for it than I need to. I know a lot about cutters so I shop to find what is going to work best for me and the lowest possible price. After all I am a New England Yankee. I get SGS cutters from http://www.carbidedepot.com/Dynamiclanding.aspx?CategoryID=4243 They have the best prices I can find for the quality cutters I want. I get all my straight and ball nose cutters from them.

Onsrud is a GREAT cutter and they have single flute which helps to get the chipload where you want it. I don't use Onsrud ONLY because of price. I can get what I want for less money.

Beckworth, MCLS, Bits n Bits, Centerion, Burchett, Think & Tinker Precise Bits, Melin, Garr, SGS - Just to name a few other brands I use.

For the 3D stuff - my most common cutter is a 1/8 ball nosed tapered end mill from Burchett In this video cutting HDU https://www.youtube.com/watch?v=BdBpddb0jbs I also have 1/16 ball nose tapered and 1/32 ball nose tapered. These cutter are NOT inexpensive, but I have found the cheapest on the market I can find. They just do a job that no other cutter can do.
 
There was a question on the Vectric user forum this morning about ball nosed cutters.

I just thought it was fitting for this thread:

THE QUESTION WAS
I have a question about the use of Ballnose bits. These are used in the finish pass on 2.5D and 3D cutting. On occasion used in other places.
My question is when and why do you use 1flute 2 flute 3flute or 4 flute.What are the advantage or disadvantage of the different flutes types.
If I where as I am just starting to gather my bit collection this would be a lot of bits if I only get 1/32 thru 1/4. So if you would lets keep this
to Ballnose/Roundnose for now I will ask about the others later. I am putting together a small book of information for myself. So I will not
have to keep asking. LoL



MY ANSWER WAS:

OK - BALL NOSED - but in reality it is for ALL of them

The number of flutes is significant.

More to the point - the space between the flutes - or gullets.

The gullets allow clearance for the material being cut - chips - room to move out of the cutter and get flung out of the cut area.

How much gullet space you need depends on the chipload. Chipload is how much material is being cut as the cutter advances in feedrate in one revolution of the cutter. It is generally specified as thousandths of an inch and is PER FLUTE.

SO - if you programmed chipload is .020 and your cutter is 4 flutes - then the cutter will advance .160 per one revolution. That times the spindle RPM will give you feedrate in IPM.

NOW - you need to think about 2 things 1) can your machine move that fast? 2) Is the gullet space big enough for the chips to evacuate?

1) Can your machine move that fast? I don't know - I don't know what you machine is capable of.

2) Is the gullet space big enough? Not at a .020 chipload on a 4 flute cutter.


What you should be trying to achieve is the optimal chipload for the material you want to be cutting.

The ways to adjust chipload is with RPM, FEEDRATE, Number of flutes.

In most cased with Brass, Aluminum, Plastics, Wood, HDU, PVC you want no more than 2 flutes, though you can safety cut brass with 4 flutes. Some of the tapered ball nose cutters are 3 flute and certainly workable options. Four flute and higher cutters are for steels where the chipload is low.

Single Flute Cutters on soft materials and high chiploads are a great option - plenty of gullet for chip evacuation.

I don't recommend that you go out and get a set of cutters!

If you are asking this question now, it is because you don't know about cutters. There are a LOT of variations out there and a LOT of options.

I highly recommend that you get ONLY what you need for the job at hand until you have a better understanding of cutters.
 
When I first stated running my AndI CNC it was buying the Onsrud 1/2" cutters the do a good job. But I got to know the owner of C.L.Baily pool tables. He was running two twin head twin table Andi's and then bought one like mine. He got me using 3/8" cutters from Vortex and what an improvement that made. Only bits I would use after that and they were cheaper than the Onsrud. But I did buy them 10 at a time.
 
Really great information thus far. Thanks, guys.

How does one decide on the machine to get, leaving initial cost out of it? It seems to me that one would decide on the hardest material one expected to mill regularly and choose a machine that had the torque or whatever to handle it. But how do you measure torque and where would you get a source of information giving material resistance? Hope that isn't too confusing.
 
First - decide on your footprint and cutting envelope.

The materials you want to cut will determine the rigidity of the machine you'll need. I wouldn't worry too much about torque, really. A proper cut recipe doesn't need much torque really. It's the machine's rigidity that's most important. If you want to mill anything harder than dense woods, I strongly suggest you look at machines with real linear guides (good hardened steel rails of some kind, with linear bearings designed for that purpose). Anything lighter than that (v-bearings on aluminum, skate bearings on black pipe, etc) will not be as rigid.

You can get the job done with the lighter weight machines. Don't get me wrong. Just about any machine can cut brass and aluminum. They just flex so much you have to go incredibly slow, use a lot of air or lube and babysit the things like crazy.
 
As I recall, my ShopBot had lots of flex and I turned my back on at my own peril! Especially with the small amount of aluminum parts I did. Babysat everyone of those will a container of cutting fluid and an acid brush painting the tool path with every pass. It got the job done and repeatability was acceptable, but it was painfully slow.

So the marriage of the appropriate cutter and speeds trumps torque. Rigidity rules. CNCparts makes a machine with vee bearings on hardened steel on a wide rail that looks rugged. But that price tag bumps my budget by a factor of 3.5. I think I would want to have a market and product that would justify that first! Maybe a machine to grow into someday.

That said, market and product has not yet been chosen. Still doing due diligence in that area.
 
What Jason says is true, now that I've upgraded mine to harden rails/linear bearings, things are a whole lot more rigid. I still have some flex depending on how hard I'm pushing the cutter, but have found that adding slower/lighter final pass or two cleans things up quite nicely. Looking at the Shapeoko machines and a few others that use the extruded parts, the end plates just seem to be a bit light to me, been curious what kind of production can be had from them and still get good cutting results. It's not that they won't do the job, they just may need that extra pass or two and won't buy you the efficiency that you're looking for for high output use and day in/out production. I don't intend to use my machine for day in/out production though, but being able to set it and let it do its thing would be nice.
 
Rigidity is a key component as well as the power of the drives.

Reliability, repeatability is what you really want.

Some people out there will set up their machine and let it run overnight while they are sleeping.

A 4' x 8' machine can do large cabinet parts that cannot be done on a small machine. It can also do 100 smaller items in one setup and run unattended while you do something else - like sleeping.

It depends really on what you would want to do with it.

If money didn't matter, servos and glass scales would be great. But money does matter - it always does.

CONSTRUCTION

There are levels of machines from shop built to small commercial to industrial.
I spend my daily live in the industrial machine market. I buy machines and put them into production.

One think I do is to compare machines side by side on a spreadsheet to do a weighted comparison.
I will give each feature a number from 1 - 10 and add everything up in the end. Highest number wins.

The rigidity of the machine itself in the important feature. A benchtop machine will have the rigidity build it and does not need a solid rigid base. BUT - if the base is in fact part of the machine then a welded square steel tube base is a great feature.

The rails that hold the carriages are a weak point. Industrial machines have mostly linear guides - some machines have boxway construction. I found some IGUS slides that are similar to boxway construction and plan to do some testing some day after house remodel is done. One of the popular slides is the HIWIN slides. They are VERY similar to the industrial linear slides on the industrial machines. My choice would be HIWIN.

The round rod with inexpensive ball roller is OK, but they do have some play to them and the fine dust can get into the balls and gunk them up.

The construction of the carriages is also important. The gantry needs to be proportionally strong enough to support the weight of the spindle and overcome the cutter forces.

I downloaded the Joe's machine. This is a nice hobby machine build for someone that cannot design their own but still want to build. I do NOT like the rails. It is just not possible to take IRON pipe and have anything precision about it.


DRIVES

I will break the drives into two things - motors and linear actuation (screws, chains, belts, rack and pinion)

**Motors**

Servos with encoders - this is an induction or brushed motor with a encoder mounted to it. The motor turns and advances the carriage to a set distance. The encoder records the distance and feeds back to the controller the position. If the distance is not reached the motor keeps turning until it is reached. There are safety measures.

Steppers - this is an induction type motor that will rotate a small portion of one rotation per electrical pulse sent to it. The pulses are sent in extremely rapid fire succession making the motor appear to be rotating. The pulses and stepper motor are calibrated to a given distance per a given number of pulses. As an example - 1000 pulses equals 1 inch of travel. If the carriage travels 1/2 inch and stalls but the controller send all 1000 pulses - the controller believes that the actual distance traveled is 1.00 That is called lost steps.

Stepper motors are rated in oz-in holding force. That is a study in itself of which I am not overly well versed in. But I know a little bit. Steppers also some in several form factors (sizes and configurations). Most popular is NEMA 23. Next most popular is NEMA 34

Power can be done in the 100 oz-in range up into 2000 oz-in.

For a light duty machine 12" x 12" a 90 or 120 oz-in NEMA 17 or 23 is maybe going to be OK.

For a 4' x 8' machine with heavy service a Servo is most likely a better choice but a NEMA 34 at 1500 - 2000 would also be OK


SOOOOO - SPEED - is also a consideration. I can get NEMA 34 1000 oz-in to move a lot faster "reliably" than I can get NEMA 23 100 oz-in motors. WHY does that matter? Well - if I were just cutting profiles in 1/2 plywood maybe not - unless I was getting paid by the hour - then faster cutting means more dollars. It also matters in 3D cutting. Generally 3D cutting takes a lot of time due to the surfacing and the small stepover rates to avoid sanding. It would be nice to cut 3D at 600 IPM without loosing steps. 3D cutting can easily tie up a machine for 18 - 24 hours of constant cutting. Reducing that to 6 hours can mean a lot.



**Linear actuation**

Lead Screws, Ball Screws, Rack and pinion, chain drive, belt drive.
Industrial machine have precision ground ball screws.

Lead Screws are probably most common on our smaller machines and rack-pinion becoming more popular.

I have two lead screw machine and they are OK. I am skeptical of a lead screw on a 8' distance. I would be concerned about whipping unless it is a large enough diameter screw.

Rack and pinion does not have any whipping issues and is a direct drive system. These systems and also mounted up and somewhat away from the dust. They also allow the underside of the machine to be used for long parts through the table

I don't really have much to say about the other systems - but they should be looked at and considered.


SERVICE

If you can fix your own machine, this may not be really important. If you need help to fix it, then you need some local support.


SUPPORT

This you will need. At least phone support to walk you through a fix if needed. This can be from anywhere in the world.
 
How much do you need to worry about balancing the components so the machine is less likely to tear itself apart? Seems like you could get into a train wreck in a hurry here... Similar to how beginner metal lathes often have belt drives instead of gear drives so when you crash the cutter into a piece of metal it doesn't rip everything apart.
 
On the rack and pinion drives there is a spring mechanism that disengages the gear in the event of a crash. The gear turns but the slide goes nowhere.

Unlike industrial machines there are no failure sensors on any of these machines that cause automatic shutdown. When they malfunction - it is not pretty. Much like a kickback on a tablesaw.
 
On the rack and pinion drives there is a spring mechanism that disengages the gear in the event of a crash. The gear turns but the slide goes nowhere.

Unlike industrial machines there are no failure sensors on any of these machines that cause automatic shutdown. When they malfunction - it is not pretty. Much like a kickback on a tablesaw.

Only sometimes a LOT more expensive :rofl:

Dang you Leo you have me studying the China Machines now. Looks like you can get a lot more machine than building and for less money.
 
I built limit switches for the ShopBot that shut it down when they were hit. I did that after crashing it, no real damage, but causing me to recalibrate the whole thing. Not fun.

There is something to be said for building your own. At least you know how it works and what it take to get it going again. Darren is showing that.

Jay, what machine are you specifically looking at?
 
Was less worried about crashing to the edge, but more if your spindle or axis motors were a lot stronger than the frame was capable of handling if you could torque the system apart that way..

Very interesting discussion...
 
OK. thought process for machine acquisition to this point.

Machine working area size 26" x 50". Weird, I know, but based on previous experience, the space I am willing to dedicate to it and the size of projects I expect to run.

Rack & pinion. Rugged, solid construction. Steel with some 8020 extrusions. Vee-bearing on hardened steel edges. Again, my experience with the ShopBot influence this.

Materials to mill: wood, plywood, plastics, foams, composites, and non-ferrous metals.

Material hold-downs: clamps, screws, tee-slots, and vacuum.

Software: TBD, but translatable to other machines for the purposes of sub-contracting out.

Still to decide: motors and whatever else that hasn't hit my radar screen thus far.

Any suggestions about finding small job shops locally. Thus far I have only found machine shop services to milling metals. It occurs to me that what I am looking for would not be advertising in a costly fashion. But how do I find them? I envision using my machine for R&D, proof on concept, and production language. Actual production runs would be done by a small job shop with minimal inventory on hand. That has storage and tax implications I don't want to deal with. BTDT. My main efforts would be to sales and new product development.
 
I sold my ShopBot because I was moving in 2004. Now I think they are overpriced. Four years ago I joined Joe's CNC and have watched builds there. He has a new Titan machine he prototyped and is now testing. Don't know yet whether there will be plans, kits, sub-assemblies or what. Not including controllers, likely ~$4K for a machine ready to assemble.

My information gathering matrix is getting well populated! Thanks. Lots of research to do, but keep it coming. I am collecting links, resources, suggestions, opinions in every area that pops up.

Which brings me to the comment you made about signs being the best money-maker. Your thinking? What other markets might pursue? I noted boxes, scrolls, Corbels, Wreaths, Intarsia, Engraving, picture frames, and cabinet engravings.

I want to decide on products and markets before I decide on the machine.

Carol I spent all evening on the 3D sign forum Leo listed and looking at these two sites http://www.qcrouter.cn/ and http://www.salecnc.com/catalog/ I need to do a lot more studying but both look like workable machines at a Great price
 
crashing happens.... all my good crashes have been with me driving the machine manua

Limit switches help if you overrun the machine's cut envelope. They don't save you from a vise or clamp or other such thing getting in the way. Running within it's capacities are what seem to keep my machine predictable. I only have miscut parts when I am pushing it too hard - too fast -- too deep a cut -- something like that. Or if a cutter breaks...


Ryan's question of balance is a good one - you don't have much to 'balance' really - but you can get sympathetic resonance and harmonics in the machine that will affect cut quality. This is why it goes back to rigidity. The tighter your machine runs, the less affected it is by those kinds of things. The resonance can be setup just by the nature of the cut - a series of short little steps at the right frequency can get things rattling good!
 
Top