• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.
[4]

[Newbie] Cnc Questions

[3]
[10] Like what you see?
Click here to donate to this forum and upgrade your account!

HEAVYMETAL87

Active Member
Active Member
Joined
Aug 6, 2016
Messages
47
Likes
3
#1
Hello everyone,

I am purchasing a PM727 and PM1022 mill and lathe. I was considering in the future making them where they could be used for both CNC and manual machining if that was possible. What would be the cost of making a basic, three axis CNC milling machine- roughly? I have a computer for it already, and had planned on using servo motors if I go this route.

Also, is there a way for a CNC machine to check to see if the cutting bit is still viable- so that if it breaks it simply stops trying to make the piece? Would a fourth axis in the form of a rotary table mounted so that the work piece was horizontal work in lieu of having a CNC lathe?

Sorry if any of this seems kind of dumb, I really don't know all that much about CNC- but it does look interesting.

Thanks all!
 

wrat

Active Member
Active Member
Joined
Oct 16, 2014
Messages
171
Likes
120
#2
Also, is there a way for a CNC machine to check to see if the cutting bit is still viable- so that if it breaks it simply stops trying to make the piece? !
I've worked with this feature years ago, but i've no idea if it's common today. It also takes some extra lines of code.
The idea is to monitor the spindle motor amperage. If something is cutting, there's amps being expended. If a cutter is snapped off, the idea is that it's basically freewheeling. Trouble is, WHEN do you monitor? You have to load up the cutter flutes, then tell the machine to start the monitor. Then, before the cutter is unloaded, turn the monitoring off. Otherwise, it'll think every cutter is broken before it touches the material. It can't be made automatic (in any scenario we could envision). Worse yet, if the cutter is just torn up badly, you still get amps and the unmanned machine will plow through the part like all is fine. During roughing, this might be recoverable, but during some finishing moments, maybe not.
In so-called 'high speed machining' (as if anything was ever 'low speed'), or meltdown machining, this is easier what with only one cutter (or a few) doing everything in small and easily predictable passes.
While it's true that in most cases, there's adequate coolant splash that an operator can hardly see if things are going great, anyway. And often there's enough noise to not hear if things are doing well. Still, he can usually catch things earlier than any monitoring system.

Would a fourth axis in the form of a rotary table mounted so that the work piece was horizontal work in lieu of having a CNC lathe?
Not generally. Certain moves, yes. But not for most of what you might want dependably round.


Wrat
 

JimDawson

Global Moderator
Staff member
Director
Joined
Feb 8, 2014
Messages
6,442
Likes
4,221
#3
Sorry if any of this seems kind of dumb, I really don't know all that much about CNC- but it does look interesting.
When on The Hobby Machinist there are no dumb questions. We were all there at one time, we are here to help. ;)

I was considering in the future making them where they could be used for both CNC and manual machining if that was possible.
Yes, that is very possible and highly recommended. My mill can be operated as a manual machine, or as a 2 axis with manual quill, or as a full 3 or 4 axis. When I CNC my lathe, I will keep full manual capability.

What would be the cost of making a basic, three axis CNC milling machine- roughly? I have a computer for it already, and had planned on using servo motors if I go this route.
I just replaced 4 mill sized servo motors on a CNC router, the parts cost was about $4000, including all of the cables and a new power supply. Add to this a CNC motion controller and software. Then there is the ball screws to replace the lead screws. I guess I would budget around $4000 for a 3 axis conversion on a mill. A lathe would be less because it is only 2 axis.

Also, is there a way for a CNC machine to check to see if the cutting bit is still viable- so that if it breaks it simply stops trying to make the piece?
Yes, it's possible, but would require a pretty sophisticated control system. Normally when the tool bit breaks the machine is just ''cutting'' air so no real problem. Really high end machines, $1,000,000 +, have systems that detect broken and dull cutters. This is not normally found on lesser machines. I thought about adding that to my mill, but I don't think it's worth the effort. It would be a lot of programming work and require a lot of experimentation to get it right. A better solution when running the machine ''dark'' (unattended) is to simply to mount a camera near the machine so you can watch the progress on your I-phone;), maybe even with a remote E-stop.

Would a fourth axis in the form of a rotary table mounted so that the work piece was horizontal work in lieu of having a CNC lathe?
In some instances, maybe. I have found that trying to do lathe type work on the 4th axis is not really practical. It is mostly to do with cutter geometry. Consider that a lathe tool bit is more or less a sharp point, where a milling cutter presents a relatively wide flat surface to the work. Each machine is pretty much designed to do a specific type of work. There are high end horizontal machining centers that do both very well, but again you are looking at $1,000,000 + machines.
.
.
 

HEAVYMETAL87

Active Member
Active Member
Joined
Aug 6, 2016
Messages
47
Likes
3
#4
Holy cow! A conversion can cost up to $4000.00?
Is it possible to do one for less- or is there some surprise expense I am not aware of- aside from the servos, power supply, mounts- what other parts are required to do the conversion?
 

JimDawson

Global Moderator
Staff member
Director
Joined
Feb 8, 2014
Messages
6,442
Likes
4,221
#5
.............aside from the servos, power supply, mounts- what other parts are required to do the conversion?
When you are talking servos, you are talking some real money. The word servo implies a full closed loop system, where you have feedback from a positioning device via some type of encoder, and normally today when used in this context means BrushLess DC Motors. BLDC motors and drives are expensive. Just to put it into perspective, I did some major upgrades to my machine and the total was about $7500. And I already had the ball screws, and the X and Y axis motors & drives. I also went with full on industrial hardware, the motion controller card and associated electrical hardware was about $3000, then add about $750 for 1 micron linear encoders.

Cheap/ accurate/ fast: Pick two :grin:

Having said that, going with an open loop stepper system is much much cheaper. You still really need the ball screws for the X, Y , and Z axis, but they are not terribly expensive from an Ebay vendor.

Below is an example of a 3 axis stepper motor kit, if I were going to cnc your mill this is the specifications of kit I would use. This is just the first one I came to in a search, there may be a better price. I have used a lot of Wantai hardware and have been happy with the performance.
http://www.ebay.com/itm/Free-ship-w...701345?hash=item33bd7210e1:g:4KYAAOSw3YNXY6~h

The above stepper kit also comes with a breakout board and cable that is compatible with Mach3

Below is an example of a ball screw kit, this is only an example. I have no idea if any of these would work on your machine.
http://www.ebay.com/itm/Anti-backla...803347?hash=item3d07cbd2d3:g:KlUAAOSwAYtWNvOL

Mach3 CNC controller software is about $200, CAM software is about $150, then there is the CAD software. The good news is that Fusion360 by Autodesk is free to hobby users and is a full professional 3D CAD/CAM package.

The rest of the parts, mounts, and other stuff can be built in your shop.
 

HEAVYMETAL87

Active Member
Active Member
Joined
Aug 6, 2016
Messages
47
Likes
3
#6
Honestly sir cheap and accurate sound like the two that matter to me the most.

I really don't mind slow all that much, mostly because it's going in my garage, which will be air conditioned, likely have a computer in there anyhow, so I think that I would be fine if it's not going to get anything done in much of a hurry. Just as long as it is a good piece when the job is done. :grin:

Is it true that a manual mill converted to be CNC/manual will never perform as well as a dedicated machine? I am not trying to build parts for NASA here, just live steam engines and every day little projects.

Which is easier to convert by the way- the lathe or the mill?
 

Metal

Active Member
Active Member
Joined
Sep 30, 2015
Messages
232
Likes
115
#7
Heavy: dont let the cost dissuade, you, you have a mill, you can make many of the parts for the conversion WITH the mill that you already have! :)

you have two options
Everyone will suggest servo's, these are a closed loop system that can self sense their location, they are also expensive
Steppers only turn (usually 1.6 degrees) when sent a signal, if they can't turn due to load, the software will think that they did and you will get inaccurate parts.

I've done both now, I would highly suggest steppers, oversized for the job and geared down with a timing belt, if things go bad you can get bad parts, but they are /dramatically/ easier to setup, and cheaper
 

JimDawson

Global Moderator
Staff member
Director
Joined
Feb 8, 2014
Messages
6,442
Likes
4,221
#8
Is it true that a manual mill converted to be CNC/manual will never perform as well as a dedicated machine?
Not true at all. It all depends on what quality of hardware is used to do the conversion.

I'm not sure which machine would be the easiest to convert. There are less parts to contend with in the lathe conversion. But I have only worked on mills and routers thus far. The lathe is on my list, as is my surface grinder.
 
Last edited:

HEAVYMETAL87

Active Member
Active Member
Joined
Aug 6, 2016
Messages
47
Likes
3
#9
The machine that I was looking at converting is the PM727M milling machine.
It's basically another Rong-Fu- 7x27, about 700 pounds etc.

I take it that with CNC instead of learning to manually machine and go a little bonkers from that, it's CAD that makes you a little bonkers with CNC?
 

T Bredehoft

Active User
H-M Supporter-Premium Member
Joined
Dec 27, 2014
Messages
2,161
Likes
1,510
#10
For what it's worth I ran two different mills which had been converted to CNC, one vertical light weight, one horizontal, heavy weight, I programmed them by punching in machine code, wrote my own sub routines, etc, used tool offsets, etc. for one-off pieces, never did production (more than one piece) on them. I ran those machines for over 10 years, taught others to use them, did lots of tool-room work on them. The management bought Mazak lathes and mills, I ran them, too, but preferred my old manual CNC machines. Yeah, tool changers were great, but the machines were no more versatile than my to manuals.
 

TomS

Active User
H-M Supporter-Premium Member
Joined
May 20, 2013
Messages
1,485
Likes
465
#11
+1 on what Jim said about a stepper driven system. For reference my conversion cost about $2,000, not including the price of the mill. Your cost should be a bit lower as I have a NEMA 42 4200 oz. in. motor and driver on the Z axis. Properly set up a stepper system will produce very accurate parts. You just need to realize you are converting a hobby manual mill into a hobby CNC mill. Adding CNC components doesn't make it more accurate.

Good luck with your project and keep asking questions. We are here to help.

Tom S.
 

JimDawson

Global Moderator
Staff member
Director
Joined
Feb 8, 2014
Messages
6,442
Likes
4,221
#12
The machine that I was looking at converting is the PM727M milling machine. It's basically another Rong-Fu- 7x27, about 700 pounds etc.
Nothing wrong with that. It's a good base for a home sized CNC platform. You can run anything on that as a CNC that you could as a manual machine. Adding CNC is not going to increase it's capacity, but it will certainly increase it's usefulness.

I take it that with CNC instead of learning to manually machine and go a little bonkers from that, it's CAD that makes you a little bonkers with CNC?
Once you shift your thinking from manual feeds and speeds and get comfortable with the machine running itself, then you get to go bonkers figuring out the process. Workholding, order of operation, tool selection, etc. As you gain experience with both CAD and your machine, the parts you make will become more complex, and you will probably spend more time figuring out how to machine the part than actually machining the part. I did a 4 axis part the other day that I spent hours working out order of operation and tool selection, and about an hour doing the actual machining.
 
Last edited:

jbolt

Active User
H-M Supporter-Premium Member
Joined
Dec 3, 2013
Messages
1,225
Likes
1,006
#13
I have had both manual and CNC mills, my current mill is CNC only, no hand wheels. I have become very comfortable running the machine as CNC or manually with an xbox controller as a pendant.

As for the lathe I have never found a reason to convert one to CNC as I don't know what I would use it for. Then of course there would be the problem of wanting an auto tool changer, air chuck and bar feeder!$$$
 

bpratl

Active User
H-M Supporter-Premium Member
Joined
Sep 5, 2012
Messages
940
Likes
53
#14
The only reason that I converted my lathe to CNC was to be able to cut metric/ASE threads with out changing gears and have the ability to machine tapers/balls without special fixtures.
I started with an ELS conversion on Z axis only and then converted the X axis for full two axis CNC.
 

HEAVYMETAL87

Active Member
Active Member
Joined
Aug 6, 2016
Messages
47
Likes
3
#15
Thank you everyone, I really do appreciate it.

The reason that I wanted to do CNC was for repetitive stuff mostly- or things that I would value being precise and what not. Part of the reason why I didn't want to break the bank on it too much. At the same time though, if I decide to go full Kool-Aid (as my old Drill Sergeant used to say) with this whole thing, then I would have at least gotten my feet wet enough to know what I was doing. I would love to build a router at the bare minimum, I had an idea with a friend of mine who is a programmer to make custom tool trays where someone would just select the tools on a website, lay out how they wanted them arranged through Flash, and then if they ordered it some how convert the order into a format the CNC software would recognize.

I take it that the bells and whistles like tool changers and the like is where the DIY realm ends and the nightmare begins?

I wonder if it would be possible to attach a 3D printer head to an R8 taper some how- and use a CNC mill as a monster 3D printer.
 

JimDawson

Global Moderator
Staff member
Director
Joined
Feb 8, 2014
Messages
6,442
Likes
4,221
#16
I had an idea with a friend of mine who is a programmer to make custom tool trays where someone would just select the tools on a website, lay out how they wanted them arranged through Flash, and then if they ordered it some how convert the order into a format the CNC software would recognize.
I'm a little confused about what you are suggesting here. :confused: But in software almost anything is possible

I take it that the bells and whistles like tool changers and the like is where the DIY realm ends and the nightmare begins?
That might be considered an advanced DYI project;) I have a mill sitting out here with a 24 station tool changer on it. I wrote the software to run it.

I wonder if it would be possible to attach a 3D printer head to an R8 taper some how- and use a CNC mill as a monster 3D printer.
Yes, absolutely! You need a 4th axis available to do it. I'm going to do this with both my router and mill.
.
.
 

HEAVYMETAL87

Active Member
Active Member
Joined
Aug 6, 2016
Messages
47
Likes
3
#17
Jim, sorry about that. I was kinda going on a tangent.

A few folks I know including myself really like shadowed tool boxes. The idea was to set up a small business where people could design and order custom inserts for their boxes made out of plastic that would be made on a CNC router. The plan was to figure out a way to make the website and the CNC software work together.

That still sounds totally badass though- having a tool changer like that. Then again though, it would also imply a powered draw bar.
What's the forth axis if you will forgive me being dense?

If I have plans for something as a PDF, can that be imported over to the software and used as a start point- or would I have to basically reproduce it all?
 

JimDawson

Global Moderator
Staff member
Director
Joined
Feb 8, 2014
Messages
6,442
Likes
4,221
#18
A few folks I know including myself really like shadowed tool boxes. The idea was to set up a small business where people could design and order custom inserts for their boxes made out of plastic that would be made on a CNC router. The plan was to figure out a way to make the website and the CNC software work together. If I have plans for something as a PDF, can that be imported over to the software and used as a start point- or would I have to basically reproduce it all?
There are PDF to DXF converters, so yes that is possible. To get the scaling correct you would need a couple of accurate dimensions. I do this all the time. There are also methods to import a picture and scale that. It requires a bit of work, but once you get the hang of it, it's pretty easy.

That still sounds totally badass though- having a tool changer like that. Then again though, it would also imply a powered draw bar.
upload_2016-8-26_11-49-33.png



What's the forth axis if you will forgive me being dense?
Not a problem, we are here to help educate. :) X, Y, and Z add up to 3 axis. The stepper drive on the feeder of the print head requires an axis also. So you need a 4th axis to operate it. Mach3 will handle this.
 

HEAVYMETAL87

Active Member
Active Member
Joined
Aug 6, 2016
Messages
47
Likes
3
#19
Wow.

Well that is convenient. That tool changer you built is too awesome for words. I was just thinking of something like the Tormach ones.

A lot of the plans have literally got dimensions everywhere on them- so that would work in my favor as well.
 

JimDawson

Global Moderator
Staff member
Director
Joined
Feb 8, 2014
Messages
6,442
Likes
4,221
#20
I didn't build that tool changer. I just put a new motor on it, built a new gearbox, designed the controls, and wrote the software.
 

HEAVYMETAL87

Active Member
Active Member
Joined
Aug 6, 2016
Messages
47
Likes
3
#21
Jim,

Are there kits out there for tool changers (hobbyest ones) and fifth axis attachments for smaller mills?
I'm going to take the plunge, but after I have both machines.
 

JimDawson

Global Moderator
Staff member
Director
Joined
Feb 8, 2014
Messages
6,442
Likes
4,221
#22
Here are a couple videos of styles of tool changers that could easily be built in the home shop.



I am going to assume that you mistyped 5th axis and actually meant 4th axis.;) However, a 5th axis is not out of the question either. A 5th axis normally adds tilt to the 4th axis. Great for making turbine wheels from a solid billet.:)

Note the tool changer on this one also

Yes, 4th axis kits are available. Google 4th axis kit.

Or it's just more fun to build one.

http://www.hobby-machinist.com/threads/4th-axis-build.40642/

Once the 4 axis drive electronics are installed, it can be used for both a rotary axis and a 3D print head. All that's required is one more stepper driver and wiring. Add another stepper drive, and you could have a 5 axis machine!:grin: I think Mach3 will handle 6 axis.
 

JimDawson

Global Moderator
Staff member
Director
Joined
Feb 8, 2014
Messages
6,442
Likes
4,221
#24
Wow! That's quite a piece of equipment. Never seen one of those before. Looks really expensive.:eek:

A PM25 would be OK as a CNC platform. Yes, the lighter weight would affect the maximum material removal rate, and thus the accuracy/stability at higher speeds. As you go down in weight, the ability to take a cut that will load the tool bit properly is decreased, and thus the tool life suffers.
 

JimDawson

Global Moderator
Staff member
Director
Joined
Feb 8, 2014
Messages
6,442
Likes
4,221
#26
Taking shallower cuts is the answer. Keeping the feed rate up to keep the chip load in a reasonable range.
 

hanermo2

Active Member
Active Member
Joined
Mar 26, 2013
Messages
36
Likes
17
#27
Fwiw..
A CNC conversion, when done really well on a lathe, cost me about 8k€ in parts at wholesale rates.

That is for industrial quality, with industrial components, somewhat equivalent to a 100k € lathe, so..
3 axis, 4 Mhz, CSMIO-IPs, 0.2 micron step rate, 32 mm ballscrews, C axis at 30.000 steps/turn (0.013 degrees) with 2.5 kW servo spindle at 90 Nm torque.

Belt drives x,z, 30 mm wide,HTD8, AC servos 220V, 750W, 1:2, 20.000 steps/turn at 5 mm screws for 0.25 um res.
1:3 at spindle, HTD8,1:3.

My interest is for high end quality, and very high resolution..

I have 2 toolchangers, not yet implemented, with TC2 a 70 kg servo turret, 8 tools.
TC1 s a 4-way front flat turret.
Lathe is a 12x24 light industrial.

Glass scales will be put on, hopefully with feedback, hopefully with submicron res.
 

JimDawson

Global Moderator
Staff member
Director
Joined
Feb 8, 2014
Messages
6,442
Likes
4,221
#28
@hanermo2 we would like to see pictures of you setup. Sounds like a very high quality conversion. Is the lathe stable enough for submicron machining?
 

hanermo2

Active Member
Active Member
Joined
Mar 26, 2013
Messages
36
Likes
17
#29
No public pics.
It´s a Q of $$, and commercial concerns, and I am sorry.
This may change..

The lathe is not good enough for submicron features (feature =/= accuracy).

Since about 2005, I got about 1 micron mechanical resolution, with ghetto-solution gecko drives, treadmill servos w. US digital E7 encoder 10k counts, and HTD5/15 belts at 1:3.
That was still with acme screws.
Today the stuff is 2-6x better.

I need to do major finishing work, re: wiring MPG and ENC threading, shields wrap-up, before I can do pics.
It really must look industrial, with metal shields, et al.

Major work is required, eg. the saddle plate must be replaced again or milled to accept the toolchangers.
Its about 15 mm too high for the 4-pos toolchanger, at 50 mm thick, 200 mm wide, 400 mm long, by memory.

I can make 1 micron steps on turning.
E.g. the lathe can make (need to) a plug gage with 1 micron steps, etc.

It is not accurate to 1 micron, no.
The steps are 1 micron in size.
Measuring the steps, I can find out eg which one is 50.000 and which are -0.001 and plus 0.001.

Actual accuracy, as-is, today, is about 0.01 mm.

I may, likely will, get better resolution with ground screws,
This is a Q of money, and it´s an ongoing project.
Until other stuff finishes, I cannot justify the ground screws, at this time.

The x axis is 14x more stiff than stock, with 3/4" ballscrew, used to be dual constrained in tension.
The X-screw itself is crap, a roton / mcmaster part I bought back in 2002, when I started.
It needs changing, and causes problems.

It can, and will, make small movements when properly loaded in an incremental direction.
But it has lots of slop, error, etc etc..

Changing to a stock , my import stuff, 25 mm / 5 mm, screw at back, will likely:
+remove most slop.
+remove the spring in current system.
+Allow accurate positioning.
I expect about 2 um-1um-0.5um resolution, and about 3-4 um accuracy over a 50 mm area.
Cost of screw, new, retail, about 120-140€ iirc, and I have some.
About 20-50 hours work.

A ground screw would give 0.1-0.5 um resolution, and 0.5 um accuracy, mapped.
I have one C0 screw, but it only has == 100 mm movement.
Trying to source another.
A custom ground hiwin screw is an option, 20/2 mm, about 800$ ex-works, with 200 mm movement. C3.

I want the longer movement so (gang) tools can work front or back, as needed, since I have a servo spindle.
 

JimDawson

Global Moderator
Staff member
Director
Joined
Feb 8, 2014
Messages
6,442
Likes
4,221
#30
With glass scales for feedback you don't need to worry about the ballscrews. The motion controller will automatically compensate for for any backlash or screw error. It is best to use double nut ballscrews so you can preload for zero backlash. This is how my mill is set up and it will consistently position to +/- 1 micron. I'm using 1 micron magnetic scales.
 
[6]
[5] [7]