How powerful of a motor do I need?

[MeltManBob]

I'm looking into designing and building my own spindle motor along with designing and building my own CNC machine but I am just learning so I apologize ahead of time for any stupid questions and yes for the time being I am pretty set on the idea of making the motor and not purchasing one since I see this are part of the fun!

The things I would like to use the CNC machine for would be wood, plastics, aluminum and possibly some steel. I put steel in there because some of the motor parts would use it but I am not expecting to machine very large parts out of metal and I don't expect it to be super fast at doing it. I kind of see this like the idea of getting a small pickup truck with a V6 engine, you can haul some light loads pretty well and you could even get away with hauling loads that you should really have a larger truck for but at the expense of the truck struggling and burning through gas.

So what I am looking for is some information on what RPM ranges this machine would need to operate at for each material and approximately what sized motor I would need. I'm guessing that a power rating wouldn't be accurate but maybe a Kv rating since from what I understand power varies with RPM.

I should mention that my intention for powering the spindle motor is possibly to convert household 120/110 to DC and drive the motor with that but I am open to any advice. Originally I was thinking a computer power supply but everything I've read about motor building suggests that getting more torque lowers the Kv which is why I am thinking of using household voltage so I can recover the RPM's.

Anyway thanks in advance!

[cheetahcnc]

Since you are starting from scratch, there are VFD's which take 110vac input and output three phase variable speed power. This way you can easily vary your output RPM. Depending on the desired max RPM, it would be wise to underdrive your spindle to gain torque for the lower RPM's, if you do not plan on using a gearbox. Without size and weight restriction requirements, it is difficult to come up with a HP recomendation

[MeltManBob]

Well I can say that my plan is to make the machine no larger than 2'x2' if that helps. As for the VFD I'm guessing that's the same thing as an Electronic Speed Controller which I plan on trying to build myself. Basically I'm starting with the mindset of I will build everything and as I learn I will decide what I don't want to spend my time on but the electronics are definitely something I want to try.

If I remember correctly I read somewhere that aluminum is cut between 7k and 12k RPM's and acrylic between 15k and 20k, correct me if this is wrong.

Also I should add that I realistically I don't expect to cut anything much thicker than .25" in one pass through wood or plastic and at most .0625" in aluminum unless that too modest. Steel would be either cutting patterns out of sheets or very slowly shaping bar stock. Hope that helps a bit to describe what I'm looking for, thanks for the quick response.

[ger21]

It's extremely unlikely that
a) you'll be able to cut steel with this machine
and
b) that you'll be able to build your own spindle.

Unlesss of course you already have a metalworking lathe and mill.


Cutting steel will require very low rpm, and most likely coolant, which doesn't mix well with wood. If you look at all the build threads here, you won't find anyone cutting steel, except for maybe a short test cut, or some engraving.

[dkirtley]

You can't just attach an RPM speed to a material such as 10,000 rpm for aluminum. You have to use the linear cutting speed for the material and then look at the diameter of the bit (or the workpiece if on a lathe) and see what speed the cutter is travelling along the workpiece. A 1/8 in bit has to spin much faster than a 3/4 in bit. That is why the teeny tiny drills that dentists use crank out 100,000 rpms and beyond.

I recommend picking up a copy of Machinery's Handbook. It has the cutting speed information for many different materials. I picked up an older edition on Amazon for a few dollars. Some stuff may be out of date but I don't think that drill bit sizes have changed enough in the last few years to justify getting the "latest edition." It has lots of good information but not the most exciting reading material.

The next part of how much power you need to cut depends on how fast you are removing material. People routinely cut steel with little lathes and mills that are run with 150 to 300 watt motors. They just don't do it very fast. You can go with big iron mills and lathes with 5 HP motors and beyond. They can remove material somewhat faster. I cut some steel the other day and was using much less than a hundred watts of power -- Me pushing a hacksaw. Needless to say, it didn't go very fast -- but it worked.

[MeltManBob]

Haha using a hacksaw! Right now a lot of the things I know I would use the machine for would be acrylic, I'm also saying wood because I usually make things out of wood although it's typically furniture. I'm fine with cutting steel at very slow speeds, nothing I am or would be working on would be that time sensitive.

So for arguments sake if I built a 1HP motor then I should be able to cut wood and acrylic fairly easily? Better yet here is what would be acceptable to me in terms of cutting rates, 12 IPM in acrylic or wood up to 1/2" thick, 12 IPM in aluminum up to 1/8" thick, and 6 IPM in steel up to 1/32" thick. Would these expectations be hard to achieve? With these expectations is there a rough estimate of a minimum sized motor I would need?

One thought with designing this motor is that when I wind the stator I will be using small gauge wire as strands for a larger gauge which I could have all of the strands terminated into some sort of switching mechanism so that based on what I'm cutting and the speed desired I could have the wires arranged differently. Since more turns of wire creates more torque but a lower RPM/V then if I had say 48 turns of 29ga wire I could have the switches connect the wires to be 24 turns of 2 strands which is the same as 24 turns of 26ga wire for when I need to cut steel at a low rpm. If I wanted to cut at a fast RPM I could have them connected as 6 turns of 8 strands which is the same as 6 turns of 20ga for less torque but more RPM's.

As for building my own spindle I would most likely purchase the shaft and bearings and attempt to make the outrunner housing for the magnets that would attach to the shaft. If I purchase those parts and they are of decent quality then what else would be difficult to do to make the motor? Maybe I forgot to add but my plan is to have the motor shaft attach directly to a tool holder although I haven't decided on either a collet or chuck. Is this a completely stupid way of doing things?

[dkirtley]

Not to dissuade you from tinkering, but that sounds like a motor that only Rube Goldberg could appreciate. Electronic control of motors is a much more normal approach. Many of the RC groups have a lot of info on wind your own stuff. They are making some monster brushless motors. They have some places where you can buy pre-made cores and you machine the shafts and design your own windings. Pretty cool stuff but way beyond anything I would spend time with. Life is too short. I don't know if it is a stupid way to do things or not. People find entertainment in many ways. I'd be the last one to judge anyone on that. There are lots of people that make their own spindles for CNC.

I have no idea on the inch per minute but if you are just powering through and cutting it in one pass, you are going to need some serious power behind it. You will also need some hefty support for moving it around and to support the workpiece. If you are going to be cutting metal plate extensively, you might as well fork out for a plasma cutter to mount on it. It would probably come out cheaper than building a one off piece with that kind of power. Another alternative would be just to buy a mill and retrofit it for CNC. Do you already have the machining capabilities to make this stuff? Are you building to entertain yourself or make a living off of it? If I were going to depend on one for my livelihood, I would buy off the shelf components that I could get replacements for so I don't have to keep jobs waiting while I piddle around making replacement tools.

If I were building a spindle, I would build around one of the ER collet holders that come on a straight shaft and machine it to suit. They can be had for $15-$50 from Hong Kong depending on capacity.

"Just being able to cut" is not a real valid criteria for motor selection. More important is duty cycle and cooling. If you are going to be cutting a lot, most commercial routers used for wood and plastic are in the range of 2-3 HP. You can use a little $20 laminate trimmer from Harbor Freight if you don't mind replacing them when they burn out.

Just depends on your goals and what you want to play with.

[MeltManBob]

Thanks for the info! This is not something to make money off of, just for personal use and the learning experience. I get a lot of random ideas that I wish could turn into projects but are never developed simply because of the inability to make small parts and/or accurate parts. I decided to go the DIY route because a different project I'm working on requires most of the parts to be cut out of acrylic and to be repeated accurately multiple times not to mention that is just for the prototype which means I will be making many revisions. I just don't want to spend that much time trying to make the parts by hand over and over again. I would rather spend my time designing something, putting it together and testing it to figure out how to revise the design.

I checked into getting the parts laser cut which would work for this project but by the time I pay for all of the parts for the prototype I would have already spent about $200. I'd rather invest the time and money into making the machine that I can use over and over again at my convenience than throw it away at a machine shop.

I gave the cutting rates that I did because it seems like I've been able to cut at about that speed by hand with a dremel but I could be remembering wrong. As for the motor design I have been posting on an RC forum but I finally decided to post here because CNC is the end goal and I figured the people here would be better suited to help me figure out what it would take to do what I want like figuring out the motor parameters I should shoot for.

For example if you guys here were able to say something like I would need a minimum of 1HP at 5000 RPM's and the ability for the motor to spin upwards of 30k RPM's to do what I've mentioned then I could run with that and seek help actually designing the motor from the RC guys. I say minimum to mean a realistic minimum, I'm not expecting anyone to say do this and I promise it will do what you want, just a good idea of the standard I would at least have to meet to probably accomplish my goals.

Something else to add is that part of my reason for choosing a CNC over say a laser cutter which would do what I want, is that I view a CNC machine as being a bit more versatile. It's hard to say what my needs will be down the road beyond the few idea's I have at the moment but once I get this built and I can actually develop my ideas I will find out if it would be worth my time and money to say build a plasma cutter or laser cutter.

The only machines I have available to use is a decent table saw, drill press, and a wood lathe. They are actually my grandfathers in his garage shop and I'm sure he has some other tools but those are the ones that come to mind that I'm sure I would use. Also my grandfather is a master carpenter retired from the navy as pattern maker or mold maker from back when all of that stuff was done by hand so I'm sure he will be helpful.

Edit: The switching mechanism I mentioned probably would be electrical not mechanical.

[RomanLini]

Just buy a router with a decent speed control built in, many people use the Hitachi M12VC. It goes down to low enough revs to do most plastics and aluminium and uses standard collet sizes so you can buy standard cutting tools.

Then when you need an improved level of performance and/or improved power at lower RPM just add a Super-PID (external) speed controller to that router;
http://www.cnczone.com/forums/diy-cn...ontroller.html

[ger21]

So for arguments sake if I built a 1HP motor then I should be able to cut wood and acrylic fairly easily? Better yet here is what would be acceptable to me in terms of cutting rates, 12 IPM in acrylic or wood up to 1/2" thick, 12 IPM in aluminum up to 1/8" thick, and 6 IPM in steel up to 1/32" thick. Would these expectations be hard to achieve? With these expectations is there a rough estimate of a minimum sized motor I would need?

12ipm will result in melted acrylic, and burned wood.

For wood, I'd consider 100ipm a minimum target speed, and 200+would be preferable.
Acrylic can be cut at similar speeds.

While acrylic can be cut at 12ipm, with a very low spindle speed (2000?), wood will usually require 8,000-10,000 rpm minimum to achieve a good cut quality. Actual rpm will vary with tool diameter.

As was mentioned several times, buy a $125 2hp wood router.

But again, don't expect to cut steel with it.

[MeltManBob]

Again I appreciate the help and advice but I am pretty set on making my own motor even if it ends up sucking and I have to go out and buy a router to get the CNC project going.

As far as the cut rates I mentioned they were just minimums to get the point across that I don't have very high expectations for a fast machine although that would be better!

ger21- From what you said about cutting rates and spindle speeds it appears that the wood would be the limiting factor between the wood and acrylic. Basically it sounds like even though the acrylic could be cut much slower than the wood, since the wood needs a faster spindle speed to produce a good cut then I should just stick with that as my lower speed range since the acrylic could also be cut at that rate. I don't have a specific need to cut acrylic at 12 IPM, faster is better!

One thing I can think of to ask at this point would be what would be the benefit of having a spindle that runs faster than the 8k-10k you mentioned for wood? Is it simply that an increase beyond 8k-10k would mean faster feed rates? Also I should mention that I've been thinking of using 1/8" cutting bit so that not as much material is wasted since I have some parts that I could actually get 2 parts from if the cutting width is 1/8" or less. Thanks again!

[ger21]

One thing I can think of to ask at this point would be what would be the benefit of having a spindle that runs faster than the 8k-10k you mentioned for wood?

If you use bits smaller than 1/8", than the 10K rpm will limit your maximum feedrate. I use 1/16" bits at 175ipm, but use about 15,000-18,000 rpm.