[craynerd]

Hi all, don't post too often on here but read very frequently!

I've got a reasonable workshop and have built a small CNC machine/router, a CNC dividing head and a reprap 3d printer. All my projects are on my website diary but my CNC machine is specifically here on my site:
Raynerd.co.uk » Blog Archive » CNC Machine Build ? TEP

I don't have a great understanding of CNC like you guys on here, but I'm certainly not totally new to it.

I specifically have a job in which I'll be wanting to cut gauge plate steel, with a tiny 0.8mm end mill cutting tiny nibbling cuts to form a specific profile on the end of the bar. The bar will be tiny and in theory wouldn't need a machine area bigger than 3" x3"!! However, rigid, as vibrations with such small cuts will cause errors!

So my question is, has anyone any suggestions as to what to build? Money is a big concern as I know how much my other came to by the end of it!
Then, just to really make this confusing, I've always wanted to be able to cut 1/4" brass plate for clock frames, say 15" x 10" max (infact that is on the big side!!) - if I'm putting any money into it, is there a machine that will perhaps cost a little more but allow me to do both jobs??

Any thoughts or advice appreciated

Chris

[LeeWay2]

How long is that bar?
Sounds like something you might be able to do in a lathe with a milling attachment.

If you have a router already, you should be able to cut brass with it.

[Norman12]

Chris,

Mass and rigidity will be your friends.
An old out of spec iron surface plate may be a good foundation for an accurate DIY CNC. You can pick these up quite cheaply.
Fixed gantry, profile rails and good zero backlash lead-screws would be my recommendation.

[wizard]

Hi all, don't post too often on here but read very frequently!

I've got a reasonable workshop and have built a small CNC machine/router, a CNC dividing head and a reprap 3d printer. All my projects are on my website diary but my CNC machine is specifically here on my site:
Raynerd.co.uk » Blog Archive » CNC Machine Build ? TEP

Nice little machine.

I don't have a great understanding of CNC like you guys on here, but I'm certainly not totally new to it.

I specifically have a job in which I'll be wanting to cut gauge plate steel, with a tiny 0.8mm end mill cutting tiny nibbling cuts to form a specific profile on the end of the bar. The bar will be tiny and in theory wouldn't need a machine area bigger than 3" x3"!! However, rigid, as vibrations with such small cuts will cause errors!

Yes but on the other hand you are tailing an eight tenths of a millimeter end mill. That is very tiny so you reaction forces would be low. At least low compared to an 8 mm end mill. You might be able to get passable results with the machine you have given the right feed rates and spindle speeds.

So my question is, has anyone any suggestions as to what to build? Money is a big concern as I know how much my other came to by the end of it!

If this is the case don't build an entirely new machine, instead try to get what you have working. It sounds like you are doing engraving or fine detail milling given your tool size, this might result in acceptable work on the machine you have now. Itis worth a try and at worst you loose an endmill or two.

I'm not saying the machine you have now is ideal for this task. In fact I'd go so far as to say it is a long ways from ideal, but if cash is an issue it may save you a big outlay.

Then, just to really make this confusing, I've always wanted to be able to cut 1/4" brass plate for clock frames, say 15" x 10" max (infact that is on the big side!!) - if I'm putting any money into it, is there a machine that will perhaps cost a little more but allow me to do both jobs??

Those dimensions are large for most home sized milling machines. You would probably want close to 12" of travel on the "Y" which implies a big mill. Also a big factor here is the type of brass, some brasses and bronzes are very difficult. To machine well. Also a spindle fast enough for engraving may require step down belting or gearing for normal machining.

Any thoughts or advice appreciated

Chris

Well you can always to a CNC conversion of a Chinese mill. Selection of the mill might be a problem as one that gives you 10 + inches of "Y" travel is a big problem. Big as in large mill, remember that not only does the mill have to move the table ten plus inches it has to be able to support the part when doing that. Given the right design it might be possible to do half, flip and do the rest but getting the registration right might not be fun. Further Chinese mills for the most part come with rather slow speed spindles but that should not scare you off because there are many approaches to high speed spindle solutions. The bigger negative for you is that this path is expensive even as a DIY conversion.

Another approach would be a fixed gantry moving table router/mill build from scratch. You might be able to come up with something cheaper than a conversion mill but doing a really good job requires access to a mill at the very least. I'm not thinking your run of the mill light duty router here but rather something substantial enough that you can machine metals beyond Aluminum. This isn't a simple machine to design or build but a gantry that bridges a 12-16 inch cutting zone isn't impossible with modest equipment. You would also have the same issues with a high speed spindle suitable for engraving and a more general machining spindle.

In either case above good ball screws and other components will set you back some. If you intend to do a lot of work with 0.8 mm end mills I suspect that good quality ball screws would be required.

[awerby]

You could do the first project with a Taig mill. The spindle is fast enough to use that tiny cutter, although endmills like that don't usually have much length, so you'd need to make sure your bar wasn't too thick.

But (as has been pointed out) the brass clock frames would need a much bigger and heavier mill. If it's just a cut-out job, could you do it with a plasma cutter?

Andrew Werby
www.computersculpture.com

[Bob La Londe]

Well you can always to a CNC conversion of a Chinese mill. Selection of the mill might be a problem as one that gives you 10 + inches of "Y" travel is a big problem. Big as in large mill, remember that not only does the mill have to move the table ten plus inches it has to be able to support the part when doing that. Given the right design it might be possible to do half, flip and do the rest but getting the registration right might not be fun. Further Chinese mills for the most part come with rather slow speed spindles but that should not scare you off because there are many approaches to high speed spindle solutions. The bigger negative for you is that this path is expensive even as a DIY conversion.

Spindle speed is an issue for lots of things, but as has been mentioned flipping the part might be an option. I did a mold for an associate that required me to stack plates and machine both sides of some of them. I started with a base plate with a centering register (pocket) and two alignment pins centered based on the register. Then every plate was predrilled to match the alignment pins. Then every plate was machined on the jig. Those that needed to be machined on both sides were simply flipped and clamped down.

Another approach would be a fixed gantry moving table router/mill build from scratch.

That is an excellent option. A fixed gantry can be very rigid and given the shallow profile parts he can even add central angular bracing with only modest difficult.

You might be able to come up with something cheaper than a conversion mill but doing a really good job requires access to a mill at the very least.

I disagree. Not in general, but in principal. Often cheap machine tools are used to make better machine tools for those who have time, but not money. Yes it would be a huge pain to layout and cut pieces to make a larger gantry on a smaller gantry, but it can be done.

I'm not thinking your run of the mill light duty router here but rather something substantial enough that you can machine metals beyond Aluminum.

Even for aluminum rigidity is an issue. I make a molds in aluminum and currently use a couple small desktop mills for it. I see issues anytime I get about 1/8" (3.175mm apx) cutter size and take modestly heavy cuts. I can get good cuts by taking care with my lead in and lead out, and buy taking lighter roughing cuts and always making a nice finish cut to complete. I do some light steel work on one of my small mills, and I take the same approach. Lighter than calculated cuts, and always a finish cut.

This isn't a simple machine to design or build but a gantry that bridges a 12-16 inch cutting zone isn't impossible with modest equipment.

No, but its not rocket science either. The draw back again is that your envelope in Y (or) will be more than twice your travel. There is atleast one commercially made cast iron gantry design sold as a mold cutter that uses this format though. If I was into foundry work I would probably make one myself. As it is I do plan to make some braced fixed gantry machines out of aluminum at some point to replace my desktop machines for most shallow work.

You would also have the same issues with a high speed spindle suitable for engraving and a more general machining spindle.

High speed spindles are invaluable tools for small cutter machining. For wood you can use a wood router with very good results. For aluminum you really have to run coolant at those speeds, so a wood router might be dangerous with its open architecture drawing atmosphere through the electrical components of the motor so a water soluble coolant is definitely out. Even with some form of light oil the heat from heavy cutting will kill the router fairly quickly. (I know I have killed a bunch of them.) This leaves you looking at the sealed water cooled spindles. I have one that turns 10000-40000 rpm, but I don't know how well it will last yet, and it was farily expensive even as a cheap Chinese motor. Go with a domestic brand and the price of the spindle dwarfs the price of the rest of your build. If you do a lot though it might be worth it.

In either case above good ball screws and other components will set you back some.

Lubrication is as important as initial quality for longevity. Perhaps more so. Depending on what you have access too, and depending on how important feed rate is to you... you can get very good accuracy and repeatability with quality acme screws and double spring loaded acme nuts. (Not pinch nuts. Double nuts. ) Better than aerospace requirements if you take the time to map your screws. Ball screws though require only lubrication for the life of the screw. Even with ball screws if you want best accuracy and repeatable performance you need to double nut them. They will however turn much easier (and faster) than an acme screw. My fastest stepper driven acme machine stalls at about 80ipm (2000mm/m) under cutting loads. My cheapest ballscrew machine stalls at over 600ipm (15000mm/m) and cuts just fine at upto 150ipm (3800mm/m).

If you intend to do a lot of work with 0.8 mm end mills I suspect that good quality ball screws would be required.

Required? No. Highly recommended... absolutely. They are also much less likely to bind under heavy load. They are far superior in almost every respect, but its is not impossible to get quality results with other components.

[wizard]

Spindle speed is an issue for lots of things, but as has been mentioned flipping the part might be an option. I did a mold for an associate that required me to stack plates and machine both sides of some of them. I started with a base plate with a centering register (pocket) and two alignment pins centered based on the register. Then every plate was predrilled to match the alignment pins. Then every plate was machined on the jig. Those that needed to be machined on both sides were simply flipped and clamped down.

I have no problem with flipping nut it is an issue with high volume milling. Good design plays a part here too.

That is an excellent option. A fixed gantry can be very rigid and given the shallow profile parts he can even add central angular bracing with only modest difficult.

A fixed gantry of suitably rigid design should work well.

I disagree. Not in general, but in principal. Often cheap machine tools are used to make better machine tools for those who have time, but not money. Yes it would be a huge pain to layout and cut pieces to make a larger gantry on a smaller gantry, but it can be done.

I have to disagree, at least in this context, when we are talking a gantry machine built out of steel and suitable for machining metal more difficult than aluminum.

The need for a mill isn't a huge negative here though, just take the gantry to a suitable machine shop. The machine effort would be a small part of the overall cost.

Even for aluminum rigidity is an issue. I make a molds in aluminum and currently use a couple small desktop mills for it. I see issues anytime I get about 1/8" (3.175mm apx) cutter size and take modestly heavy cuts.

Which mills?

Do realize that this is an issue on any machine no matter its size, if you operate beyond its capacity you loose accuracy. However I still think it is possible to build a gantry machine that effective deal with the issue.

I can get good cuts by taking care with my lead in and lead out, and buy taking lighter roughing cuts and always making a nice finish cut to complete. I do some light steel work on one of my small mills, and I take the same approach. Lighter than calculated cuts, and always a finish cut.

That is pretty common practice.

No, but its not rocket science either. The draw back again is that your envelope in Y (or) will be more than twice your travel. There is atleast one commercially made cast iron gantry design sold as a mold cutter that uses this format though. If I was into foundry work I would probably make one myself. As it is I do plan to make some braced fixed gantry machines out of aluminum at some point to replace my desktop machines for most shallow work.

I'm of the opinion that somebody could do well selling a gantry casting for DIY gantry machines.

High speed spindles are invaluable tools for small cutter machining. For wood you can use a wood router with very good results. For aluminum you really have to run coolant at those speeds, so a wood router might be dangerous with its open architecture drawing atmosphere through the electrical components of the motor so a water soluble coolant is definitely out. Even with some form of light oil the heat from heavy cutting will kill the router fairly quickly. (I know I have killed a bunch of them.) This leaves you looking at the sealed water cooled spindles. I have one that turns 10000-40000 rpm, but I don't know how well it will last yet, and it was farily expensive even as a cheap Chinese motor. Go with a domestic brand and the price of the spindle dwarfs the price of the rest of your build. If you do a lot though it might be worth it.

The expends of spindles is real. If you have the right skills DIY might be the answer. For engraving type work a small spindle with an ER type collet would be a winner.

Lubrication is as important as initial quality for longevity. Perhaps more so. Depending on what you have access too, and depending on how important feed rate is to you... you can get very good accuracy and repeatability with quality acme screws and double spring loaded acme nuts. (Not pinch nuts. Double nuts. ) Better than aerospace requirements if you take the time to map your screws.

True when new but wear does introduce issues. The thing here is that ball nuts can be easier and even more cost effective.

Ball screws though require only lubrication for the life of the screw.

This however isn't rue at all. Ball screws are even more sensitive to lubrication than acme.

Even with ball screws if you want best accuracy and repeatable performance you need to double nut them. They will however turn much easier (and faster) than an acme screw. My fastest stepper driven acme machine stalls at about 80ipm (2000mm/m) under cutting loads. My cheapest ballscrew machine stalls at over 600ipm (15000mm/m) and cuts just fine at upto 150ipm (3800mm/m).

That speed can be significant for today machinng needs. In the end though is the expense hugely different.

Required? No. Highly recommended... absolutely. They are also much less likely to bind under heavy load. They are far superior in almost every respect, but its is not impossible to get quality results with other components.

We could probably debate this but I'm of the opinion that ball screws are a requirement for such fine work. In the end it costs less and there is less screwing around.

[Skyfire]

Hi guys. I just saw this thread and the title interested me because I'm building such a small, rigid CNC milling machine right now. The original purpose is to show how to build the industrial level machines with a small but real model because many coversion here based on benchtop mill. They are all not of something real for CNC. So I get the idea to build a benchtop CNC but following the industrial protocal.

So my aim is just fit to craynerd's need. small but made of casting iron structure to handle steel cutting works.

weight about 120KG(260lbs);
travel range XYZ: 8"*5"*8";
linear rails and ballscrews for all axis;
micro stepper drive or servo optional;
milling spindle power 0.75-1HP, speed range 200-3000 or 5000RPM; electric spindle 1.5kw 24000RPM exchangeable;
etc.

And the most important than the obvious features above, accuracy under 0.02mm(0.001") typical error 0.01mm(0.0005")for all axis in whole travel distance just as the industrial CNC.

I'm building it now and have a thread since long time ago start with cast mouds. I want to build the best benchtop CNC in te world for now. maybe you guys can read my thread and may found something interesting.

Welcome for chat. Thanks.

This is my thread:
http://www.cnczone.com/forums/vertic...nning_end.html