Newbie- Is this the way to go?

[chrugel]

Hi
I've been collecting parts now for some time and getting close to begin the actual build, but i would like some thoughts and comment on my plans and Ideas.

What i want in the end is a cnc drill and mill capaple of drilling and milling pcb, milling stuff from wood to aluminium. Probly starting off with my dremel. I seen many here saying that it has much play so I might no be satisfied with that one, and it is freaking noisy!!! What about a BLDC Inrunner for model aircrafts?? Seen the in the KW range and they are good for at least 25K?

And as everybody elses this machine has a very tight budget and will be made with what i have lying arround. Budges estimate: 600$ give and take some

For the frame I'm going to us an old table with a steel frame, square 25mm x 25 mm tubing. Maybee if I'm able, I'm going to get a 1cm steel plate as base plate so i have a reference are from which i constuct the axes. Depends if i can get a hold of one cheap, dont think I can buy this on Ebay! I've read that it is good to have some weight to avoid resonace, at least to get the resonance frequency as low as possible.
The size of the table is 500mm x 600 mm. Because i live i a small condo it is going to be a moving gantry design to maximise the working area vs total size. I do not poses any advanced tools when it comes to the actual making of this machine, so my plan is to use the machine to improve it self when it is roughly finished.

The gantry is going to be made og aluminium, probably 4mm thick plate reinforced with solid 15 x15 mm square rods.

The slides and the screws is taken from a comercial cnc solder paste press.
My first misstake, why did i not just take the whole assembly!!!!!
A little luck after all, the slides are exacly as long as my frame!
The screws has a travel of about 1 cm pr rev.
My problem here is that the screws are 15 cm to long and made of hardend steel, how can I reetol them and fit them with new bearings?


I have decided to go for a DC servo solution using the UHU controller.
http://gsst.wikispaces.com/UHUget
The reason for this is that i have three DC servos that i think will fit my needs an if i like it should not be wery hard installing much moore powerful motors if needed:
Im probably going to use optical shaft encoders from US digital:
http://www.usdigital.com/products/s4/
Think these shuld do the trick! Have no clue about the CPR count i need yet .. . Probably in the area of 200 - 400 i guess . .. ?

X axis (The whole gantry)
DC servo from Bodine Electric
130V 1 A 2500RPM 1/8 HP
Probably driving the screw with a timingbelt solution with a shaft encoder on the screw, or the motor?????
Anyone got an idea of the ratio i would need here?? 5- 1 ?

The Y Axis
A DC servo from Japan Servo 127K1010
[URL="http://www.cnczone.com/forums/showthread.php?t=23730"][/URL]
(Thank you salahz for the info!)
Rated for 24VDC, 400ma, 2000RPM no load. LAB TEST BELOW:
RPM TORQUE (oz-in) Current
----- ----------------- ---------
2160 no load 400 ma
1740 30 2.2 Amps
1290 60 4.2 Amps
700 90 6.3 Amps
stall 100 7.6 Amps

The Z axis I've not yet decided but im thinking of another DC sevo from Japan Servo bu this one i hav very little info about.
There is a sticker on it that saysType: DS48BT25-2 EMF CONST: 6,4V/KPM iis this useful in any way?
I also have a bunch of small DC servo motors that only sasy RS-545PH with an optical encoder attached. Probably cant use the encoder since it is not incremental just the old tacko style.

The problem i see comming her is getting the motors attached to the pulleys since I've got axels in all shapes and sizes! An also the attachment of the encoders might be a problem since it is a very delicate instument!

I will post picture of my progress and of all the parts in this thread as the construction comes along.

I'm very new to this and I got lots of plans and ideas in my head, I might be way of on a lot of things so please coment!

I'm from the suthern parts of norway and would really like to come in contact with others of my kind.
I also hide behind this fact when it comes to my written english language, have me excused it is not my native language!!

best regards
Christian

[HomeMadeCnc]

www.mechmate.com its my next build, why bother reinventing the wheel.

Good luck,
Tim

[chrugel]

Originally Posted by HomeMadeCnc www.mechmate.com its my next build, why bother reinventing the wheel. Good luck, Tim

Thats so true, I've downloaded the drawings and will surtenly get some ideas from them! Cant exactly use them as a basis since I'm locked to whatever meterials i have lying around, but it is alway good to see what other have done.

Thank you Tim.

regards
Chris

[chrugel]

Hi

I might have frightned some people with all my writing in my first post. I realize now that probably not very many are going to read all of that. Anyway, here is something a little bit more fun, pictures!


This picture shows the X-axis and the Y axis motors I'm planning to use.



This picture shows my screw and slide. I have 4 slides and 2 screws.


This image show my thoughts for the Z axis, the smallest one i have quite a lot of, maybee using 2 to drive thr Z axis?

The last two pictures shows the power for all the motors exept the X-axis. I'm planning to power this one with a 500VA 2x 55 v TOROID transformer. I have to remove som of the secondary windings because 2x55V AC is to much when i need 130DC.

Anyway this is all for now.

Regards
Chris

[chrugel]

Hi again!

Now things are starting to come to come together, I got a deal with a metal-worker where I work to cut out the aluminum pices i need! He actually were villing to do it nearly for free!
I have also got a hold of the pulleys and timingbelt, I decided to use a gear-ratio of 2.4 - 1 from the motor to the ballscrew. This would give me a top speed of about 270 IPS. My problem no is that I'm going to have to order the encoders and I have no idea what CPR to use. The way I see it now I think I'm going for 360 in the Z axis and 500 on the X and Y axis.

I've also been looking for software and I think I have decided to go for TurboCNC for G-Code to Machine. Since i have no experience with 3D design I'm still looking for a simple free 3D cad software.
So fare I've done all my designing in the very advanced cad software package paintbush! I am at least going to find a BMP to G-Code program!

I've been thinking a lot about Spindle designs lately and think I'm going to try and make two spindles. One with rpm range of 10k to 20k and 1500W and If my plan works i hope to be able to create one in the 30k to 45krpm and at least 1500W.
The slow one I'm thinking of using a Plunge Router directly coupled to the spindle.
The basis for the highspeed design is going to be the brushless inrunner design found on this page: http://www.cncathome.com/spindles.html
But i think i might try ceramic abec7 bearings from this place:
http://www.edsbearings.com/prices.html
The drive system I'm thinking of using an old Bosch plunge router wich has a 1400w 28k rpm motor. For the really high speed setup i planned to use a 2-1 timingbelt setup.
Does anyone have any idea if any of this stands a chance to work?

I soone hope to begin the assembly of both the dc controllers and the whole machine it self.
Her are my highly advanced paintbrush drawings, 1 pixel is 1 mm.
The moving Z assembly is not included yet since I've not yet come up with a way to do this.

regards
Chris

[chrugel]

Hello again!
Progress report with pictures:

Anyway i think I'm redy for the initial testrunns soon. Little sceptical concerning the configuration of the motorcontrollers since my motors are quite a bit oversized. And the parameterization of the controllers probably involves some oscillations.
Initial runns with just a dc power on the motors shows tha i probably have to do something about the speed, at half the rated voltage i get moore than 300ipm. And if I short the motor when it is moving fast the whole machine moves on the floor! I think I might have undersized the machine a bit to my motors, at least the one on the Y axis.

I'm also going to try and build my own spindle, a slow one for milling aluminum. 5k - 10k rpm and 1,5 - 2 hp.
ER20 chuck and collets have been aquired along with a monster several kw 24V powersupply for the RC BL outrunner i intend to use. I just have to give it a try!

Best regards
Chris

[Lurkepus]

Hi
maybe we could meet up some day i kind of live just down the block form you ( Larvik ) . i am planing to build my first cnc mill ( JGro) and it would have been nice to meet someone who have been trough the prosses of building one ..

i also got a limited budget on this build so it will be just a smal plywood mill ...

if you want to get in touch pls send me an PM and we can exhange contact info

Best Regards

Stig

[LUCKY13]

The little servos marked "6.4v/KRPM, have you guys run these any at all ?

What RPM will they run ?
How muhc power you think they have?
How much voltage have you put to them?



I am setting up a X1 mill and I know where some of these motors are with encoders. Plus I am building a very small mill with only a few inches of travel. I would like ot use these motors if I could find out a little more about them.


Jess

[FandZ]

Wow, your progress is pretty impressive, especially given the fact you are not using any cad to design it. I'd be completely lost. I've used solid works and auto cad. I hated auto cad and liked solid works.

I've been using google sketchup now for a few months, Sketchup isn't true cad but you can make accurate 3d models with it fast and it's FREE. You can even get gcode from it with a step or 2. I know I'll have rocks thrown at me for saying it but I'm liking sketchup so much that I haven't used solid works in over a month. Also there is a ton of support. If you can use paint brush you can use sketchup. It is also a very good training tool to get you used to thinking 3d cad like.

For CNC software I plan on using Mach 3. Everyone raves about it and I think it is rightly so.

Originally Posted by chrugel Hi again! Now things are starting to come to come together, I got a deal with a metal-worker where I work to cut out the aluminum pices i need! He actually were villing to do it nearly for free! I have also got a hold of the pulleys and timingbelt, I decided to use a gear-ratio of 2.4 - 1 from the motor to the ballscrew. This would give me a top speed of about 270 IPS. My problem no is that I'm going to have to order the encoders and I have no idea what CPR to use. The way I see it now I think I'm going for 360 in the Z axis and 500 on the X and Y axis. I've also been looking for software and I think I have decided to go for TurboCNC for G-Code to Machine. Since i have no experience with 3D design I'm still looking for a simple free 3D cad software. So fare I've done all my designing in the very advanced cad software package paintbush! I am at least going to find a BMP to G-Code program! I've been thinking a lot about Spindle designs lately and think I'm going to try and make two spindles. One with rpm range of 10k to 20k and 1500W and If my plan works i hope to be able to create one in the 30k to 45krpm and at least 1500W. The slow one I'm thinking of using a Plunge Router directly coupled to the spindle. The basis for the highspeed design is going to be the brushless inrunner design found on this page: http://www.cncathome.com/spindles.html But i think i might try ceramic abec7 bearings from this place: http://www.edsbearings.com/prices.html The drive system I'm thinking of using an old Bosch plunge router wich has a 1400w 28k rpm motor. For the really high speed setup i planned to use a 2-1 timingbelt setup. Does anyone have any idea if any of this stands a chance to work? I soone hope to begin the assembly of both the dc controllers and the whole machine it self. Her are my highly advanced paintbrush drawings, 1 pixel is 1 mm. The moving Z assembly is not included yet since I've not yet come up with a way to do this. regards Chris

[chrugel]

Hello Jess,

So fare i have tested this motor wit 30V with no problem, i have not measured the rpm but i believe it is in the 3k - 4 k rpm range noload at this voltage due to the 6,4v/krpm rating. this means 1krpm at 6,4V and 2krpm at 12,8v and so on.
I loaded it at 24 v and at 3 amps i did not have the guts to load it any more.

This motor will be my Z axis drive motor.

beswt regards
Chris

Originally Posted by LUCKY13 The little servos marked "6.4v/KRPM, have you guys run these any at all ? What RPM will they run ? How muhc power you think they have? How much voltage have you put to them? I am setting up a X1 mill and I know where some of these motors are with encoders. Plus I am building a very small mill with only a few inches of travel. I would like ot use these motors if I could find out a little more about them. Jess

[chrugel]

Hi FandZ,

It is and have been quite fun and i have learned a lot so fare! It would most certainly have been a lot easier if I knew A CAD program well enough to take advantage of it. A lot of things i do i have to redo a little bit later on because it comes innto conflict with something else later on. I also see now that when I am to do it over a lot of things will be done in different ways!

I was not aware google sketchup were any good for this sort of designing, you might just have opend up a whole new world to me! thanks!
I'm now going to put some effort innto learning sketchup.

I'm going to use EMC2 mostly because it is free, and that i trust linux moore than windows to run a realtime system. I think I got EMC fairly under control, i have made the config files and I'm now soon ready to do my first test-runs. The only thing missing is a screw, since this is a closed loop system with the rotational sensors on the screws i need the mechanical system complete before i can do any testing.

best regards
Chris

Originally Posted by FandZ Wow, your progress is pretty impressive, especially given the fact you are not using any cad to design it. I'd be completely lost. I've used solid works and auto cad. I hated auto cad and liked solid works. I've been using google sketchup now for a few months, Sketchup isn't true cad but you can make accurate 3d models with it fast and it's FREE. You can even get gcode from it with a step or 2. I know I'll have rocks thrown at me for saying it but I'm liking sketchup so much that I haven't used solid works in over a month. Also there is a ton of support. If you can use paint brush you can use sketchup. It is also a very good training tool to get you used to thinking 3d cad like. For CNC software I plan on using Mach 3. Everyone raves about it and I think it is rightly so.

[FandZ]

I've been designing my CNC completely in sketchup. If I were you, I'd at least make a mock up in it so you could avoid any future roadblocks. It sounds like your almost finished though.

When you get a chance can you post a picture of yoru x axis from the right corner. I'm interested in seeing how you are driving your Z axis.

Here's a link to the second version of my CNC design. Just to give you an idea of how usefull sketch up can be. I have a third now but it hasn't changed much since the v2 versian. http://www.cnczone.com/forums/showthread.php?t=69385

Also there are a ton of sketchup tutorials on youtube to get you going. I think I may also have some tutorials for Solid works too if you wanted to try the demo. I'd really start with sketchup though.

Originally Posted by chrugel Hi FandZ, It is and have been quite fun and i have learned a lot so fare! It would most certainly have been a lot easier if I knew A CAD program well enough to take advantage of it. A lot of things i do i have to redo a little bit later on because it comes innto conflict with something else later on. I also see now that when I am to do it over a lot of things will be done in different ways! I was not aware google sketchup were any good for this sort of designing, you might just have opend up a whole new world to me! thanks! I'm now going to put some effort innto learning sketchup. I'm going to use EMC2 mostly because it is free, and that i trust linux moore than windows to run a realtime system. I think I got EMC fairly under control, i have made the config files and I'm now soon ready to do my first test-runs. The only thing missing is a screw, since this is a closed loop system with the rotational sensors on the screws i need the mechanical system complete before i can do any testing. best regards Chris