• Scam Alert. Members are reminded to NOT send money to buy anything. Don't buy things remote and have it shipped - go get it yourself, pay in person, and take your equipment with you. Scammers have burned people on this forum. Urgency, secrecy, excuses, selling for friend, newish members, FUD, are RED FLAGS. A video conference call is not adequate assurance. Face to face interactions are required. Please report suspicions to the forum admins. Stay Safe - anyone can get scammed.

My next CNC project - a PrintNC router

DavidR8

DavidR8

Scrap maker
Administrator
Moderator
Premium Member
I've been jonesing for a CNC router for a while now.
Two main criteria
1) it had to fit on a 2' x3' table...my welding table.
2) it had to be able to mill aluminum with decent results.
Looked at all the options from OpenBuilds, to Scienci to DIY.
On a DIY CNC Facebook group I ran across the PrintNC design. Scaleable, open source, (I'm a big proponent of open source), fully supported by a user/builder base.
https://threedesign.store/

The preferred system for the machine is LinuxCNC. I'm complete Linux noob let alone LinuxCNC. GRBL is another option but I wanted to go with the tried and true. I found a cheap used Dell desktop tonight and installed Linux and LinuxCNC.
Well I tried to...until I ran into a problem.
Remember that user base? One kind fellow walked me through command-line inputs to delete disk partitions like I was an old hand.
And it worked. I have the control PC all setup.
On to the building of the machine.
 
Last edited:
@David_R8 Did you look at using your CNC plasma unit as a CNC router? If/when I get to doing mine, that is kind of my plan but seeing as I know little of the road you have been down, not sure if it is reasonable.

My reasoning for thinking it would be an easy (relatively) conversion is this.....
X-Y hardware and software would be the same and already done
G code for spindle speed is used for plasma on/off and could become router up/down. Router does not have to turn on and off, just move up and down, therefore the same GCode could perform that function.
Watching DAC's videos, he converted the Gcode, then went in and modified the commands slightly to accommodate his unit and project. Do you need to do this too or is the different software you are using getting around this?
I would think replacing the torch holder you are using with a base for a trim router and then a height mechanism would be the only hardware mods required. A small air cylinder would be a relatively easy lift mechanism but then you are adding an air source requirement. Air solenoids are easy to work with though. I am picturing fixed depth though....maybe varying the height within a project will get a little more complex but then the stepper motor is the best option. Since CAM programs vary heights very precisely anyways, it seems easy (remember, coming from me who knows nothing about actually doing any of this, ...o_O)
Or maybe you just want to keep wood and metal separate..?

I agree with you 100% on open source technology. I think that was the dominate reason for going with FreeCAD over Fusion 360 when I decided it was time to learn to use a CAD package. Since Autodesk has altered their original marketing strategy to limiting hobby users to 1 year, I wondered if after 1 year, they will start charging, or reducing the hobby level features till users in that bracket have to purchase something. They may be counting on those that have dedicated the time to learn it, unwilling to leave it, a common marketing principle, I believe. Programs such as Firefox, OpenOffice, GIMP rival their commercial equivalents in every aspect, from features to support. Don't get me wrong...if someone uses a program lots, it may be well worth it to purchase but I tend to drift in and out on interests so don't want to purchase software that is seldom used or used for a month and sit unused till the license expires....but now I am straying from this thread.....

Whatever you decide to do, I am looking forward to another well documented, researched, build project thread by David. :D

Thanks
Regards,
Shawno
 
I can't picture long term support using Discord. You will have trouble finding past solutions, so will have to ask each time. And hope someone answers. That is why I stopped using FreeSwitch for voip, and dropped out of anything IRC or mailing list supported.
 
gerritv said:
I can't picture long term support using Discord. You will have trouble finding past solutions, so will have to ask each time. And hope someone answers. That is why I stopped using FreeSwitch for voip, and dropped out of anything IRC or mailing list supported.
Click to expand...
There's also a wiki that is updated regularly.
Discord saved my bacon last night when I ran into problems with the software install. Access to folks who have built or are building the same machine is a key for me.
 
ShawnR said:
@David_R8 Did you look at using your CNC plasma unit as a CNC router? If/when I get to doing mine, that is kind of my plan but seeing as I know little of the road you have been down, not sure if it is reasonable.

My reasoning for thinking it would be an easy (relatively) conversion is this.....
X-Y hardware and software would be the same and already done
G code for spindle speed is used for plasma on/off and could become router up/down. Router does not have to turn on and off, just move up and down, therefore the same GCode could perform that function.
Watching DAC's videos, he converted the Gcode, then went in and modified the commands slightly to accommodate his unit and project. Do you need to do this too or is the different software you are using getting around this?
I would think replacing the torch holder you are using with a base for a trim router and then a height mechanism would be the only hardware mods required. A small air cylinder would be a relatively easy lift mechanism but then you are adding an air source requirement. Air solenoids are easy to work with though. I am picturing fixed depth though....maybe varying the height within a project will get a little more complex but then the stepper motor is the best option. Since CAM programs vary heights very precisely anyways, it seems easy (remember, coming from me who knows nothing about actually doing any of this, ...o_O)
Or maybe you just want to keep wood and metal separate..?

I agree with you 100% on open source technology. I think that was the dominate reason for going with FreeCAD over Fusion 360 when I decided it was time to learn to use a CAD package. Since Autodesk has altered their original marketing strategy to limiting hobby users to 1 year, I wondered if after 1 year, they will start charging, or reducing the hobby level features till users in that bracket have to purchase something. They may be counting on those that have dedicated the time to learn it, unwilling to leave it, a common marketing principle, I believe. Programs such as Firefox, OpenOffice, GIMP rival their commercial equivalents in every aspect, from features to support. Don't get me wrong...if someone uses a program lots, it may be well worth it to purchase but I tend to drift in and out on interests so don't want to purchase software that is seldom used or used for a month and sit unused till the license expires....but now I am straying from this thread.....

Whatever you decide to do, I am looking forward to another well documented, researched, build project thread by David. :D

Thanks
Regards,
Shawno
Click to expand...
I thought about it briefly but the CNC plasma cutter is not sufficiently rigid to deal with the forces of a spindle.
There is the potential of using the CNC router as a plama cutter though. :)
 
Big step today. I submitted the order for all the parts.
Linear rails
725mm HGR20 QTY (2)
425mm HGR20 QTY (2)
250mm HGR20 QTY (2)

Ball screws
825mm 1610 Ball Screw QTY (1)
525mm 1610 Ball Screw QTY (2)
300mm 1204 Ball Screw QTY (1)
HGW20CC QTY (6)

Motors, driver and board
4 x Nema 23 Stepper Motors
4 x DM542 Stepper Drivers
Parallel Port Breakout Board + DB25 Cable
5 x LJ8A3-2-Z-AX M8
360w Switching Power Supply

Shaft Couplers:
XB25*30 8*10 QTY (1)
XB25*30 8*10 QTY (2)
XB25*30 8*10 QTY (2)

Cable Chain 25*57 QTY (2)

Spindle + VFD Package
2.2kw Water Cooled Spindle + VFD
10m Water Pipe + Flange Type 80mm Spindle Mount
ER20 Collet Set
220V
 
Yeah looks like a big step forward - I need to measure my ball screws lengths first and then I submit my own order for RF 30 CNC.

I am also planning to use Linux CNC. Mach3 feels legacy and a bit obsolete without any future support. Mach4 Hobby feels a bit pricey for what you get. Linux CNC is free, has excellent support and you do not need a motion card to start with - can upgrade later on.
 
I went with linixcnc , moderate learning curve but great community and wait for it..... source code.

I use a Mesa 7i76E (Ethernet interface ) board which has the driver support plus field I/O . No regrets so far


Sent from my iPhone using Tapatalk
 
Thanks @Tom Kitta it's a big leap for me as I'm not Linux/LinuxCNC experienced at all. And the learning curve is pretty vertical.

@kevin.decelles I went with a parallel port breakout board because I was/am completely bewildered by the Mesa documentation. I may move to a Mesa board as I get more comfortable with LinuxCNC and how the thing works. In theory it's a simple machine but the ancillary wiring to make everything run and be safe is a bit daunting.
Attached is the 'recommended' wiring diagram.
 

Attachments

I also plan to upgrade to Mesa board later on. The motion board simply takes the motion load off the PC which is not designed to easily do the "motion" part due to being a multitask oriented system. To put it the other way, PC OS switches loads on the CPU from one task to the other all the time - this is OK for humans as the fact that one app runs for 1ms and then another runs for a 1ms is not a big deal - we cannot notice. BUT motion control needs to send thousands of signals every second... That is a LOT of signals - and if few are miss timed due to task switching you may have a problem.

There is an app in CNC linux that determines what is the worst case scenario for PC lag & adjusts your timings to that - BUT that may not be too grainy enough and is not a guarantee - you may have occasionally "worse" then worst case. There are no guarantees.

Mesa board or any other motion controller is real time & removes all these issues - motion board only does motion - no task switching.
 
Same as you BOB via parallel port. I am just getting the RF30 on a base now, then I take it apart a bit to see length of ball screws I need. I think I should go with same length as the old screws are or next bigger offered. You can always cut shorter but making it longer is not possible. I only need two screws for X and Y as Z is controlled by the fine feed control on the quill. Not sure how well Z will work but everyone else seems to go with the same setup.

This is my first CNC so I hope I do not screw up too much.
 
Tom Kitta said:
Same as you BOB via parallel port. I am just getting the RF30 on a base now, then I take it apart a bit to see length of ball screws I need. I think I should go with same length as the old screws are or next bigger offered. You can always cut shorter but making it longer is not possible. I only need two screws for X and Y as Z is controlled by the fine feed control on the quill. Not sure how well Z will work but everyone else seems to go with the same setup.

This is my first CNC so I hope I do not screw up too much.
Click to expand...

You and me both! :D
 
Another big step today.
Picked up the 2"x3"x.125 wall tubing and a roll of PETG 1.75mm filament for the 3D printed motor mounts.
Next step is to modify the parameters in the Fusion 360 file so that the radius of the 3D printed parts will match the steel.

Edit: modified the F360 file and downloaded all the assembly tool stl’s for printing.
 
Last edited:
Deburred and painted the steel today.
34d7a0fb71c7c6f3e1b5dce4d0872568.jpg
 
You must log in or register to reply here.
Back
Top