I know some have done things to make the table more robust by adding linear rails to the outside of the table....why not just run a second c beam and run it off a slave belt? I know the second c beam is not an original idea but running 2 off a belt seems to be simple cheap solution.

 

Inventor CAD model of the C-Beam with 300x300 T-slot table, based around the STEP parts here: http://openbuilds.org/resources/step-parts-library.162/ I think some of the parts are out of date for the C-Beam build, based on the build video, but I can modify them once the hardware arrives.



Got my Arduino earlier, CNC Shield v3.10 is on the way, just need to get the DRV8825 drivers (thanks @Flash22) ordered then I can test the CAM/controller interface before committing to the mechanical build.

 

Mark
In the video you are using a 1/8 end mill to cut aluminum do you have a link to the bit you used?
Thanks
Bob

 

look for 3.175mm end mills, and/or ball end mills

 

Hi Paul, welcome to the forum.

I can't give "advice" as to whether or not the C-Beam plate maker would be the "right" choice or a "decent" choice for engraving.
We are experimenting here, it is all open source and we are adapting a base design to our own needs/wants as well as exploring the opportunities that we discover along the way.

I can offer a few thoughts on it though.
a) For small/fine work BACKLASH is probably your worst enemy.
b) Second enemy is probably resolution.
c) Third might be speed, by which I also mean acceleration and depth of cut (chip size).

For backlash, use anti-backlash nuts and build it TIGHT - assemble everything with NO slack/slop - OTOH don't go overboard on the pre-loading of bearings.
Resolution - IMO lead screws beat out toothed belts and the finer the pitch of the lead screw the better.
Investigate micro-stepping, some is good, more is almost certainly better (-:

Speed - tool deflection HAPPENS.
In large milling operations it is accommodated by taking roughing cuts to remove MOST of the material and finish cuts to produce fine surfaces and "to tolerance" dimensions.
Sometimes the finish cut is done twice on the same tool path, the second time being for tool deflection.
Acceleration - these are not RIGID machines and they are very light, accelerating and stopping the router too fast on the X-axis and the table on the Y-axis may produce shake that causes inaccuracies too large for what you are trying to do.
I suggest you look into low accelerations as well as low cutting speeds and thin cuts.

Again, these are merely my "thoughts" on the subject, maybe unfounded.
Hopefully others will weigh in and challenge them (-:

 

@R_B

A. Anti backlash nuts works wonders
B. Imho not a issue as long as the machine is built tight, square and file the tabs of the cast brackets, tbh the limitations are in the software used
C. This is mainly down to the large plate on the Y and the spacers and the way its affixed, replace the 6mm spacers for 3x 300mm aluminium bars to spread the load

Just my 2 cents, its all these ideas that make the machine better

ps. the c-beam can be used for pcb work and that needs good resolution

 

Thanks R_B and Flash22. Excellent input. Yep, I realize that we're experimenting here, but appreciative of all and any advice I can get. I decided to do a stock C-Beam build and then modify as needed to meet my needs. I just purchased a full C-Beam kit with steppers (thanks to Black Friday sale in the part store). I have quite a bit of machine build experience and stepper control/servo control design experience that should be helpful here. My past life was doing factory automation for a wide variety of machines and most recently doing gimbal servo systems for military apps... pretty different stuff but all goes back to the same principals. I have limited experience with aluminum structure / roller systems like these so I appreciate the advice.

I do know how to get more resolution and maintain torque on the steppers we're using by using an encoder feedback on the motor and controlling the phase currents more precisely in a closed loop fashion. This involves a more sophisticated motor control loop but it's isolated to just the motor. This is a fairly big project but something that my son and I will be experimenting with. We'll post our findings and designs as we go. Of course all the resolution in the world won't help if backlash and mechanical stiffness isn't kept in check. That will be the first order of business.

I will look into the bundle, and if it does not include anti-backlash nuts, I'll get some. I have a decent manual mill that I have squared very well so I should be able to setup and square the tabs and and the rail ends if needed. I will look into the Y mounting to improve the structure.

I don't have the resources to fund an expensive spindle, so I will initially look into one of the routers that others have used with success.

Thanks again for the advice and I'll let you know how the build goes.

Paul

 

I don't have my c-beam machine yet, but was looking at the Y axis and it looks like the bearing mount for the ball screw sticks up a little bit and looks like it runs into the Y table plate...is there any way to either machine out the Y table plate so that it doesn't it the bearing block or machine down the bearing block to give more clearance for the Y table plate to move over the top of it. If this is the case it looks like it might be possible to squeak a few more inches of travel. It was only a few frames in the video that Mark posted...so I could be totally off.

 

yep you are off it doesn't hit it's flush with the top of the c-beam....

 

I'm assuming this is the same plate (or one like it) used for the Y axis that the MDF screws to. If we are talking about the same thing, this is what I was seeing. The only difference was that it looked to me that the plate hung out over the v-wheels further than this pictures shows. The pictures I am seeing show the bearing block flush on one side but not on the other. Thats why I thought the plate edge would run into the bearing block.

 

Build Plate

Actually is this the plate in question. The v-wheels are set back from the edge on this plate.

 

sorry I was wrong I just checked...
the bearing block does stick above the c-beam and does limit the travel. you could cut the plate so that it does slide over the bearing block and just don't put the middle bolts in the plate front and back and you would gain a couple of inches of travel....
it's this plate that's in question...

Build Plate

once you are up and running just buy an extra plate and it would be no problem to notch it so that it doesn't hit....

 

iirc the above plate is the v2 (that what I run on the double Y setup) The C-Beam Comes with the universal build plate on the Y

 

Just to add the issue is not with the end plate but the clearance of the stepper at the back to gain the more travel

 

Maybe will double check the BOM and order an extra as one of my first projects. I was thinking instead of trying to get clearance by machining the bearing blocks. But not sure how much meat is on the bearing blocks in regards to the bearing and mounting holes...

 

I don't have mine yet so hard to see exactly what you mean but really only trying to maximize the travel of the v wheel to the very end of the rail. If I could gain only an inch in the front because of the stepper I could just split the difference and move the x forward half the distance of when I gained. Just have to see when I get it.

 

C-Beam™ End Mount

In this photo its hard to the height of the stepper. As long as the build plate can move over the top of the bearing block getting your mdf to clear the stepper shouldn't be to difficult hopefully. My main goal would be to get close to the same amount of travel from both axis. Gaining an inch or two of travel on the Y would increase the machines surface area quit a bit. If it can be done without buying extra parts even better ( even though I ordered two extra bearing blocks today for 25% off ). Just in case it doesn't work or I ruin one. . Don't even have it yet and modifying it already hahah.

 

I have to agree with @Flash22 if you jog the X as far forward (plate goes back) the MDF hits the Stepper, going the other way the build plate is only 2 mm above the C-Beam so it hits the endplate but I think this could be cut off to allow the build plate to go further forward.

but if you make the MDF spoil board bigger, it is just going to get even more flexible, I already can see the build plate deflect if I attempt to plunge too fast into the workpeice.

There are many solutions ideas to make the machine even better, all unfortunately add cost, Project logs • OpenBuilds C-Beam™ Machine Build • Hackaday.io there is a section close to the end of the posting about adding linears rails that looks interesting..

I am so happy with the Bundle, it is my first attempt in this hobby, but I have already got loads of projects to start, and my wife keeps adding them..

I love it, but it is going to cost me loads!!! lets keep that quiet though, Christmas is coming!!!

Michael

 

It wouldn't make it that much bigger Just utilizing the parts that it comes with so if you need the extra travel it is there. Most of the time the things that will be made on a machine like this will be a fraction of the machines table size. Just nice to know you are getting the most space for the $$$. For the money you can't beat this machine. The Shopbot Buddy uses a similar center drive with rollers for the out riggers. Something cheap and simple can be done. On a hobby machine like this nothing helps accuracy more than taking your time.

 

I have to agree if you want more travel then just swap out the c-beam's for the 1000 mm units and add an extra y c-beam but then you are going to up north of a 1000.00 dollar machine....
I have an ox already and I can tell you that I have never been limited by the cutting area.
I haven't cut anything on my c-beam machine yet waiting on another power supply but I can tell the table is plenty strong for what it's designed for....
Bob

 

I thought about going bigger but decided that I wanted to keep my "shop" small and manageable. I have actually owned two other larger format cnc routers (4'x8' and 2'x4') and I found bigger isn't always better in terms of practicality. Sure its nice to toss a sheet of plywood down and go to town but not having any room in your work area because of it really blows. The C-Beam Machine is really the wrong format for a large machine. I'm just ocd about certain things and if I can't have at least 12" x12" workable area it will bother me. I wouldnt want an 18" x30" maxhine. It would have to be 24" x24"... It's odd...I know but I don't fight it anymore. Besides for purchasing material having "even" sized beds makes things much more simple. 12" x13" I can handle...11" x13" gives me brain damage. I think I can get my extra inch of travel from machining down the front bearing block...then I will have my minimum 12x12 work area. Time to call the men in white coats, I know

 

I bet if you talk to mark he would change out the 3 extrusions you would need longer say 600mm instead of 500 and you would have it. It wouldn't cost much more I bet under 50.00 to do it that way....
I kind of think the perfect size would use 750mm all the way around, but then you would need 2 y axis rails, that would bring the price up about 200.00 I'm not sure it would be worth it, unless you have something to cut the requires that size....

Mark if you are reading this I would of ordered a 750 kit, might be an idea for a drop down menu on the order page....
I know the rails come in 1500mm lengths so 250,500,and 750 would be cool sizes pretty much a size for everyone....

Bob

 

I think you loose about 3/4 of an inch travel going forwards. If you want to access this I'd buy a spare cbeam end plate and mill the top to give sufficient clearance, probably 1/8 to 3/16. You will need to mill a recess in the bed to clear the stepper. I'll see if I can get a picture of the bed on mine, to give you an idea of what is required.

Brian

 

Already ordered two end plate/bearing blocks just in case there is a little to be gained from the back side where the steppers mounted...can't wait! Hahah

 

I added the STL files to GrabCAD.com for the 3D-Printed 'press-in' End Caps I made for Extruded Aluminum that came with my OpenBuilds - C-Beam Machine - Plate Maker ... You can download them for free at: GrabCAD - CAD library ... I have printed them in PLA and ABS filament ... They are a tight fit so they will not easily come off once pressed-in; however if needed you can sand off the sides of the areas that fit inside the extruded aluminum to make them easier to press-in and take off.
Let me know if you have any ideas for 3D-Printed extras to add to the C-Beam machine build? ...
BTW ~ If you haven't ordered the OpenBuilds C-Beam Kit it is GREAT!! The most organized and well put together kit I have ever received... very well done!! ...
OpenBuilds C-Beam CNC Machine ...
CHEERS

 

3D-Printed End Cap STL files for pressing in the ends of the Extruded Aluminum that came with my C-Beam Machine CNC Kit
... GrabCAD - CAD library
. . .

 

Hey guys,

Here is an update of my Cbeam machine enclosure build. Still have a some important things to add, but it is getting there!

 

I'm looking to purchase a spindle rather than use the Colt router that I have. I see lots of these on Amazon, ranging from 400 W air cooled to 2.2 kW water cooled. I know some of you are using these. Would the 400 W units be too small to cut aluminum? Your recommendations would be appreciated.

FYI, Galaxy Products no longer makes the collets for the PR20EVSK, and have exhausted the remainder of their inventory. I was looking to purchase the 1/8 inch collet BRT-1250.

 

I've got one of the 400 watt brush less spindles. It will cut aluminium. I've cut 10 mm thick aluminium with mine, wasn't quick though.