DIY Mill Turn Center

Show off your latest creations, whether 3D printed or not
User avatar
Windshadow
Printmaster!
Posts: 526
Joined: Mon Nov 30, 2015 6:35 pm
Location: Mid Coast Maine

Re: DIY Mill Turn Center

Post by Windshadow »

Impressive Progress well done :geek:
nebbian
Printmaster!
Posts: 175
Joined: Wed Sep 23, 2015 12:31 am

Re: DIY Mill Turn Center

Post by nebbian »

What voltage are you thinking of running on your motor? Also what is the max RPM you need it to go? This will determine your KV (rpm per volt). I can't help you much about the hall effect sensors, sorry, never dealt with them before.
User avatar
Generic Default
Printmaster!
Posts: 558
Joined: Mon Jun 03, 2013 6:56 pm
Contact:

Re: DIY Mill Turn Center

Post by Generic Default »

I was planning on keeping the RPM to a max of 4,000 RPM. A motor with ~168 KV like what I linked to would get up to over 6,000 RPM on a 36V battery, and then I'll reduce that RPM with a belt and pulley. I can change the pulley size to change the speed reduction, so I have quite a bit of range to work in.

From what I've read, the motors and controllers used in RC cars typically have hall effect sensors to help with starting torque at the low end, while multirotors and planes never run at low RPM so they run sensorless.
Check out the Tri hotend!
nebbian
Printmaster!
Posts: 175
Joined: Wed Sep 23, 2015 12:31 am

Re: DIY Mill Turn Center

Post by nebbian »

Would it be possible to add hall effect sensors to the output shaft? Then you wouldn't be limited to sensored motors. No idea if that's possible, just spitballing an idea.

http://mitrocketscience.blogspot.com.au ... t-for.html
User avatar
Generic Default
Printmaster!
Posts: 558
Joined: Mon Jun 03, 2013 6:56 pm
Contact:

Re: DIY Mill Turn Center

Post by Generic Default »

I think most of the motors that come with sensors have the magnets on the shaft and the hall circuit board on the stationary part of the motor. I've been researching a lot on this subject and I came across an open source BLDC motor controller than works with or without sensors, it looks like the best controller available and the price is reasonable. Plus it has field oriented control, which I may be able to use later for positioning and synchronized motion like threading and rigid tapping.
http://vedder.se/2015/01/vesc-open-sou ... urn false;


Update on the part making, I made the main milling spindle;
millingspindle closeup 1.jpg
millingspindle closeup 2.jpg
millingspindle closeup 3.jpg
millingspindle closeup 4.jpg
millingspindle closeup 5.jpg

I still have to do a few more operations on it, like ball mill some round slots for round keys that will transmit torque. The big red thing is a pulley for a timing belt. It's meant for a Gates Poly Chain belt with 8mm pitch, which looks like the best, stiffest, strongest timing belt on the market. RPM range for this milling spindle will be 0-10,000 RPM, with a clutched stepper motor to do rigid tapping. The whole thing is on a rotary axis and acts as the 5th axis of the mill, so all sides of the part can be finish machined in one setup. I'm also making progress on the linear ways and stuff so the machine will start getting assembled soon.
Check out the Tri hotend!
User avatar
Captain Starfish
Printmaster!
Posts: 962
Joined: Tue Dec 10, 2013 5:24 am

Re: DIY Mill Turn Center

Post by Captain Starfish »

ER25? You're not mucking about, are you :)

Be very interested to see how your spindle assembly turns out. Something I'd love on my machine is to be able to lock the tool spindle (or run it live) and have an A axis that ran like a lathe at speed with some guts behind it, or as a servo drive for a normal chucked A axis on the mill, OR as a trunnion/tombstone drive. But it all boils down to the combination of servo and spindle drive.
User avatar
Generic Default
Printmaster!
Posts: 558
Joined: Mon Jun 03, 2013 6:56 pm
Contact:

Re: DIY Mill Turn Center

Post by Generic Default »

Right now there are 4 main assemblies on the machine; Headstock, Tailstock, Turret, and Milling Column.

The turret can be switched out for a tooling block (gang tools) for small parts. The milling column moves in both the X and Y axis, and has a dual worm drive swiveling head that is the spindle from the last post.

Right now the plan is that the Headstock spindle will be powered with a ~2kw BLDC motor for turning, and then a solenoid clutch can engage a large stepper motor that works as an indexer. When the stepper is engaged, the Headstock spindle can be rotated precisely for threading or whatever, it basically becomes a 4th axis. The milling column spindle is set up the same way, the stepper on it allows rigid tapping. If the BLDC motor controller has good field oriented control, and if I get good absolute encoders working, I may be able to get rid of the steppers and solenoids altogether, although I'd still want a spindle brake for milling operations.



For now, I really need to close in on a motor / power source / ESC combination! Originally I was planning on using LiPo RC batteries, but looking at prices and charge rates makes me realize that they would only be good for a few minutes of real machining before they needed to be recharged, and they only have a few hundred charge cycles in their lifetime. So regular DC power supplies it is!

I can get a 24v 1500w power supply for about $70 on ebay. Driving a 190kv outrunner motor, this would give about 4500 RPM at the motor shaft. After pulley gear-reduction, the main headstock spindle would be capable of about ~2000 RPM turning at ~2400 watts.

Doing the calculations with some conservative values (to account for losses);

~2000 RPM at spindle
~2000 watts from motor

Gives a spindle torque of about 9.5 newton-meters, or 85 inch-pounds. A 1 inch bar would have a radius of half an inch, so that means that a cutting tool could, in theory, apply up to ~160 pounds or ~72 kilograms of force upwards on the side of a workpiece during turning operations at maximum power. That's a hell of a lot of force for a small cutting tool. At this point I think it's realistic to say that this machine should have no problem doing steel and hard metals at a reasonable rate. That excess torque should allow for quick single point threading later on when I get encoders and firmware ready for it. The milling spindle will have a higher RPM capability but less torque, power will be similar.



Can you help me with electronics selection? I don't want to spend a ton until I have a working machine, so I'd rather buy cheap parts now for testing and evaluation, then upgrade to better stuff once it's proven to work.

My plan for now is to buy these in the very near future;


http://www.hobbyking.com/hobbyking/stor ... Motor.html" onclick="window.open(this.href);return false;
http://www.hobbyking.com/hobbyking/stor ... SC_V2.html" onclick="window.open(this.href);return false;
http://www.ebay.com/itm/12v-750w-24v-15 ... Sw~OVWu3CL" onclick="window.open(this.href);return false;
http://www.ebay.com/itm/HJ-Digital-Serv ... 1232034085" onclick="window.open(this.href);return false;


The servo tester is for manual control until I get a PWM signal from the main machine controller. Am I on the right track with these things, or am I making some huge mistake unknown to me that I will immediately regret?
Feedback needed please!
Check out the Tri hotend!
User avatar
Captain Starfish
Printmaster!
Posts: 962
Joined: Tue Dec 10, 2013 5:24 am

Re: DIY Mill Turn Center

Post by Captain Starfish »

Zero direct experience of speedies, sorry - but I'm interested as to the drivers behind the decision to go BLDC instead of a more traditional AC motor with VFD.
User avatar
Generic Default
Printmaster!
Posts: 558
Joined: Mon Jun 03, 2013 6:56 pm
Contact:

Re: DIY Mill Turn Center

Post by Generic Default »

A lot of it is size, weight, power density, and price.

The Turnigy SK3 motor I linked to in the last post is about the size of a soda can and weighs a kilogram, it claims to be able to output about 3 horsepower (2.3 kW). It costs 70 dollars, a suitable speed controller costs 30 dollars and a power supply costs 70 dollars (although that power supply can't drive the motor at it's full power rating). So for 170 dollars, I can get a really good spindle motor with upgrade potential.


I could buy a 3 hp AC motor for around $200, but there are no VFDs capable of driving them off of regular 120V single phase power (like a wall outlet in the US).
For the same $170 I could spend on the BLDC motor setup, I could buy this;

http://dealerselectric.com/item.asp?cID ... &PID=24068" onclick="window.open(this.href);return false;


Which is less powerful a weighs a ton, plus it takes up a lot of space on the machine. It would probably double the weight of the entire milling column assembly and slow down Y axis movement.


Long story short, a BLDC setup using hobby RC parts is better in every way I can think of.
Check out the Tri hotend!
Polygonhell
ULTIMATE 3D JEDI
Posts: 2430
Joined: Mon Mar 26, 2012 1:44 pm
Location: Redmond WA

Re: DIY Mill Turn Center

Post by Polygonhell »

You want to be a little careful with brushless hobby motors, there are usually two issues, one is cooling and the other is that the power claims are usually only valid across a small RPM envelope.
User avatar
Generic Default
Printmaster!
Posts: 558
Joined: Mon Jun 03, 2013 6:56 pm
Contact:

Re: DIY Mill Turn Center

Post by Generic Default »

It bugs me that they don't give good specifications for RC parts, I know they overrate all of the numbers a bit too. The VFD and AC motor I linked to are supposed to be constant torque across the RPM range, which means that the power would be linear with the RPM.

The ESC's used on RC helicopters typically have a control mode called "Governor Mode" where the ESC attempts to keep the motor RPM constant by varying power to the motor. I'm assuming it's some type of PID algorithm. This is perfect for CNC machines, since a change in cutting force will not affect the RPM.

If the machine starts taking a heavy cut, the RPM will stay constant, but the power draw will increase until it reaches the maximum allowed by the ESC. I'll have to measure this with a current/voltage/rpm/power meter once I start taking cuts, then compare it to predicted values. Even if the BLDC motors only perform at a fraction of their rated values, it's still better than a massive AC motor mounted to the machine.
Check out the Tri hotend!
Polygonhell
ULTIMATE 3D JEDI
Posts: 2430
Joined: Mon Mar 26, 2012 1:44 pm
Location: Redmond WA

Re: DIY Mill Turn Center

Post by Polygonhell »

While it's true helicopters run constant RPM I'm not sure how good the closed loop velocity control is, large changes in required power are usually transient for helicopters and you can certainly hear helicopter motors struggle to keep up with rapid changes and LION batteries can supply ungodly current for short bursts.

I'm interested to see how it goes, I looked at using one for a project awhile ago and couldn't get definitive answers on expected performance with real world loads.
I know people have used "cheap" AC treadmill motors on Mills with good success, but those are usually ~1.5hp.

I assume you realize that 2.3KW is 21A at 110V.
User avatar
Captain Starfish
Printmaster!
Posts: 962
Joined: Tue Dec 10, 2013 5:24 am

Re: DIY Mill Turn Center

Post by Captain Starfish »

What's missing is a torque/rpm chart. Is it going to be like the cheapo chinese spindle I have which cracks a sad at anything under 5000RPM and doesn't really have that much torque, rather, the claimed power is coming from the 24000 RPM more than anything else. Or is it going to be like an old fashioned "Oh my goodness that thing is huge" motor and drive which has power aplenty from DC through to it's topout at around 6000RPM? Or is it going to be like a stepper, all grunt at low speed but stuff all speed?

If it works it's gold, as you say GD it's got all the advantages.
stonewater
Printmaster!
Posts: 345
Joined: Mon Jan 06, 2014 1:24 am

Re: DIY Mill Turn Center

Post by stonewater »

these guys do frequency drives small to large. their smallest is 1/8 horsepower.

http://www.kbelectronics.com/Variable_S ... rters.html" onclick="window.open(this.href);return false;

I use 8 of them where I work to control the speeds of our pitching machines. we have the pastime machines from abc batting cages http://www.battingcages.com" onclick="window.open(this.href);return false;


Tom C
MAX V1
325 MM carbon arms trick trucks effector mount LED ring heat spreader
Corvair750
V6 Hotend
Robo 3D
Flashforge creator
http://WWW.TeslagenX.com" onclick="window.open(this.href);return false; - Bedini experimenter kits, books, DVD's
User avatar
Generic Default
Printmaster!
Posts: 558
Joined: Mon Jun 03, 2013 6:56 pm
Contact:

Re: DIY Mill Turn Center

Post by Generic Default »

Small update, bigger one coming next week.

This is the new main spindle pulley design. It's threaded to fit on the back of the spindle, and it has 6 0.25" round key slots to keep it in place and transmit torque. The round keys can be adjusted and preloaded in place with set screws at an angle to the keys, so screwing in the set screws presses the keys against the spindle. This is to take out backlash, and possibly help with centering the pulley.

The belt is a gates Poly Chain, it's carbon fiber reinforced with a blue nylon tooth surface. I carefully measured the belt tooth profile and drew an accurate conic rho curve for the pulley mating profile; there is no backlash and there are at least 20 teeth in contact when the belt is on the pulley. This is the best I can do with 3d printed parts, I think it will work nicely. The green pulley is just under 4 inches in diameter, this should give more torque and a maximum speed of around 2000 RPM at 25 volts with a 149kv motor. If the initial tests are good, I'll get a full 3kW 48V power supply unit for twice the RPM.

I have the machine frame mostly done at this point, once I get the carriages finished I'll get pictures of the whole thing.
green spindle pulley 1.jpg
green spindle pulley 2.jpg
Check out the Tri hotend!
User avatar
Captain Starfish
Printmaster!
Posts: 962
Joined: Tue Dec 10, 2013 5:24 am

Re: DIY Mill Turn Center

Post by Captain Starfish »

Clever work on the keys - will adjusting the set screws also affect axial angle as well as runout, though?

Can't wait (as I'm sure you can't yourself!) to see it all together and running.
User avatar
Generic Default
Printmaster!
Posts: 558
Joined: Mon Jun 03, 2013 6:56 pm
Contact:

Re: DIY Mill Turn Center

Post by Generic Default »

overview1 june 24.jpg
Just a tease for now, I have to finish a few technical papers before I can show the rest of the machine.

I ended up going with an Azteeg X3 Pro board for now to control it, I want to use the high end smoothieboard in the future whenever it comes out.
It's running repetier host and has successfully turned and threaded parts. I need to do some major firmware modification to get this thing to run the way I want.

In repetier host (and firmware), you can jog the X, Y, and Z axes with the GUI panel. You can also work the extruder forward and reverse, and switch between extruders with the T0, T1, T2, ect command. However, you can only control one at a time. This is a problem because the B and C axes (rotational) need to be able to move simultaneously for certain cuts. The tool turret also needs to be able to home to a default zero position so the tools can be made with offsets. As far as I know, you can't cheat the firmware by zeroing the hotends/extruders in repetier firmware.

Can anyone help me out with where to start with firmware mods? My goal is for all 8 of the stepper motors to be treated like independent axes so all of them can be homed or moved simultaneously. Rather than switching to another axis by command "T3", I would just do it as part of the G1 command, like "G1 X0 Y0 Z-1 C45 B90 E10" or something.
Check out the Tri hotend!
User avatar
teoman
ULTIMATE 3D JEDI
Posts: 1783
Joined: Sat May 24, 2014 5:43 pm

Re: DIY Mill Turn Center

Post by teoman »

I think you need to look at the kinematics file.
When on mobile I am brief and may be perceived as an arsl.
User avatar
Captain Starfish
Printmaster!
Posts: 962
Joined: Tue Dec 10, 2013 5:24 am

Re: DIY Mill Turn Center

Post by Captain Starfish »

You're a horrible person for dropping just that tease pic and getting us all excited. MOAR!!! :)

Re the firmware and driver board: this sounds a bit like guys fitting extruders to CNC routers (just the other way around). Yes, you can do it. Yes, it kinda works. Eventually. But it's a kludge that's far from optimal and every time you go to use it the headaches will outweigh the advantages.

Hate to say it but, for something this complex, I'd be looking to move away from printer drive boards and go for something like a kflop, a mesa board on a machine running LinuxCNC, or even a good old parallel port style BOB (eg UCNC100 or 300) running into dedicated step/dir or servo drivers.
User avatar
Generic Default
Printmaster!
Posts: 558
Joined: Mon Jun 03, 2013 6:56 pm
Contact:

Re: DIY Mill Turn Center

Post by Generic Default »

I considered doing the mach3/linux cnc thing but I wanted to end up with something a little more future-proof. All of the parallel port stuff is going extinct, and it's kind of a pain to have to have a computer wired up to the machine all of the time just to run it.

Right now I'm maxing out the number of stepper drivers on the board, my machine would be a bit faster if there were a few more stepper drivers for extra functions. Servos are too expensive for now, and I'm getting around 50 IPM rapids with this machine without adjustment or optimization (or full stepper current). For a machine with a small work envelope like this, 50 IPM is fast enough.

The longer term plan is to modify a GUI and host software so the machine can be controlled from a touch screen tablet, like those matterhackers ones. If the new smoothieboard is as good as it sounds, it will have all of the features I need to make a working production CNC machine that can crank out parts all day unattended. At least that's the goal, I still have to get tool changes and bar feeding sorted out.
Check out the Tri hotend!
travelphotog
Printmaster!
Posts: 398
Joined: Tue Jul 15, 2014 12:30 pm
Location: Huntsville, Alabama (Marshall Space Flight Center)
Contact:

Re: DIY Mill Turn Center

Post by travelphotog »

You might want to give a look at Planet CNC controllers. They are meant for CNC milling and turning. Their MK3 board has 9 Axis control and has everything from servo to stepper control for axis movement. I know of a few lathes that run their controller. I run 3 of their systems, one on each of my mills and will be running my 4th mill on a MK3 also. Their boards must run their software though, but I have found it far better than MACH3 and it is updated regularly. The developer is fast to respond to questions and they are very full featured controllers with USB and Ethernet connection. Also have the option for running code from SD card. It is a 32bit ARM controller with a coprocessor. Might not be what you are looking for and they are not cheap, but it might be something to consider and look at for ideas. http://planet-cnc.com/turning/
http://planet-cnc.com/product/cnc-usb-c ... k3-9-axis/
Just my 2 cents worth.
http://713maker.com/ Custom aluminum and carbon fiber hot end mounts for the Rostock Max and Orion.
User avatar
Captain Starfish
Printmaster!
Posts: 962
Joined: Tue Dec 10, 2013 5:24 am

Re: DIY Mill Turn Center

Post by Captain Starfish »

Yeah the KFlop gear is similar - designed to do multi axis machining centres from the controller board, the host interface is really just the UI and data source for trickle feed.

They've already done a lot of the work you have ahead of you.

Still, if you can get the smoothie / azteeg to do the job then there'll be a warm glow of accomplishment at the end of the process you simply won't get from plugging in a controller and pushing the "go" button :)
User avatar
Generic Default
Printmaster!
Posts: 558
Joined: Mon Jun 03, 2013 6:56 pm
Contact:

Re: DIY Mill Turn Center

Post by Generic Default »

Another update on this thing. I haven't gotten that much done in the last few weeks on it. It can turn, thread, and mill parts in an ER40 collet or a 5 inch 3 jaw chuck. The spindle is eccentric by about 0.00015" or 3-4 microns because I used cheap non-precision bearings, but that's actually better than I expected. Static rigidity at the spindle nose looks to be about 40,000 to 100,000 pounds per inch of deflection, which isn't as good as full sized machines. I blame the thin-walled aluminum body tube for this floppiness.

The C axis works (precise indexing rotation of the workpiece in the spindle) with a belt drive on the back, which has a pin-engaged clutch to switch between the brushless DC motor for turning and the Nema 23 stepper motor for indexing and threading. Eventually I'd like to switch away from this system and have a direct drive for turning and threading and a clutchable strain wave gear for indexing and heavy milling/braking.

The B axis is the swiveling milling head, it's driven by a double worm gear where the worms are fast travel leadscrews on opposing sides of the rotating shaft gear. It's a 45:1 gear down using a 3d printed gear out of PLA. It can't be backdriven unless the gear teeth shear off, which is a realistic concern since it's printed plastic! So far there seems to be no wear or backlash and the rigidity is pretty good. I've been testing strain wave gears as a replacement for the worm gear since the worm gear may not be capable of larger torques from heavier milling operations.

The turret (shown below) has 8 tool stations for fixed turning tools, drills, and boring bars. It may be compatible with live tooling in the future because I have a through-turret bearing shaft that can be powered, but right now there is no motor hooked up to it. The turret is indexed with a tapered geneva mechanism for 16 stepper rotations per full turret rotation. It can't be backdriven and there is no backlash since the locking pin is tapered and preloaded slightly. Repeatability and rigidity of the turret seem to be good so far, but it's unhardened 1018 steel so it won't last forever.
turret1 july 16.jpg

overview1 july 16.jpg

This is just a small impeller I was making as a test. It's easy to program this stuff by hand.
impeller1 july 16.jpg



I have run into a few problems though. The Azteeg X3 Pro board has most of the hardware I need to make this machine work, but the firmware is limiting me. I need to be able to switch tool offsets without switching fake extruders, so I need to modify the firmware to allow tool offsets since it has a maximum of 5.

I also need to change the controls so that the firmware treats all axes as independent so they can be moved simultaneously. I need to be able to zero my fake extruder axes so the milling X and Y axes can have a home position, and also touch off milling tools to get the radius from the B axis. Then with more firmware modification I can make the swiveling B head maintain a fixed position relative to an observer while the B axis rotates, if this makes sense.





Pretty much the entire machine is scratch built, and the only parts that ended up not working right where things powered by fast motors. I didn't plan that part very well, so I had to go through several design iterations to make it work. If I had to do it again I would use sensored BLDC motors with larger keyed drive shafts and metal pulleys/gears, along with better ESCs. I should have put time into designing a gearbox/pulleybox to drive the spindles.
Check out the Tri hotend!
User avatar
Windshadow
Printmaster!
Posts: 526
Joined: Mon Nov 30, 2015 6:35 pm
Location: Mid Coast Maine

Re: DIY Mill Turn Center

Post by Windshadow »

Well Done GD :ugeek: Blody Well Done! :ugeek:
User avatar
Captain Starfish
Printmaster!
Posts: 962
Joined: Tue Dec 10, 2013 5:24 am

Re: DIY Mill Turn Center

Post by Captain Starfish »

What he said. More photos and videos of the beast in operation please, but what you've shown us so far is a thing of beauty.

How are the plastic guides holding out so far, considering smooth motion in the 'allowed' degree of freedom vs stiffness in the others?

Rather than the strain wave harmonic drive style, also worth looking at hypocycloidal eccentric gears. Nice smooth curves = relatively easy to machine the rotor and the ring can be done with some steel rod stubs. Considering putting one together to rebuild my el cheapo chinese 4th axis which is currently belt driven off a NEMA23 stepper and which I can backdrive by hand - even if it's only a bit of 8mm rod in the chuck.

Question: it looks like your X is driving the A spindle back and forth along the rails, how do you keep the toolstock synched if, for example, you're just using a dead centre in it? Slaved secondary lead screw on the tailstock, does it clamp to the same one, or what?
Post Reply

Return to “What are YOU making?”