MPCNC
Posted by Shank man
|
Re: MPCNC July 23, 2019 06:23AM |
... no "overkill"! -- if you look at microstepping and lookahead, the ArduinoDue should/could be much faster, to get better/faster/more accurate positioning 
Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]
Call for the project "garbage-free seas" - [reprap.org]
Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]
Call for the project "garbage-free seas" - [reprap.org]
|
Re: MPCNC July 28, 2019 05:38PM |
Thank you Shank man for Posting your progress.
I am still looking for a Control case, the one I found and 3D Printed is in fact a horrible 3D Printing Control case, it is so thin you can virtually see thew it.
And I do do want to waste my time with such a horribly Thin Control case, in fact do not believe it would survive very long and not protect the controller.
I have found some Anwenk 1/4"-20 x 15mm Furniture Screw in Nut Threaded Wood Inserts and have install an placed 32 of the them in the CNC main Deck.
Now I can fasten anything with a simple clamp that is 3D Printed or CNC milled using many type of material. First I still want to Find a Controller Case.
One NOTE: Because I decided to make my MPCNC so tall, the power cable on the Dewalt must be carefully positioned or it will cause the Z-Axis too bind on the X or Y max borders.
I decided, I need to get My MPCNC running. So for a quick fix I am going to quickly 3D Print this.
There is a few modification that I believe should be made while the good parts print, I can modify the pieces that want changes on.
[www.thingiverse.com]
Edited 9 time(s). Last edit at 07/28/2019 07:33PM by Roberts_Clif.
Computer Programmer / Electronics Technician
I am still looking for a Control case, the one I found and 3D Printed is in fact a horrible 3D Printing Control case, it is so thin you can virtually see thew it.
And I do do want to waste my time with such a horribly Thin Control case, in fact do not believe it would survive very long and not protect the controller.
I have found some Anwenk 1/4"-20 x 15mm Furniture Screw in Nut Threaded Wood Inserts and have install an placed 32 of the them in the CNC main Deck.
Now I can fasten anything with a simple clamp that is 3D Printed or CNC milled using many type of material. First I still want to Find a Controller Case.
One NOTE: Because I decided to make my MPCNC so tall, the power cable on the Dewalt must be carefully positioned or it will cause the Z-Axis too bind on the X or Y max borders.
I decided, I need to get My MPCNC running. So for a quick fix I am going to quickly 3D Print this.
There is a few modification that I believe should be made while the good parts print, I can modify the pieces that want changes on.
[www.thingiverse.com]
Edited 9 time(s). Last edit at 07/28/2019 07:33PM by Roberts_Clif.
Computer Programmer / Electronics Technician
|
Re: MPCNC September 22, 2019 08:49AM |
I was about to print final parts, discovered that after copying the G-code files from my computer to the SD card.
That MY PC incorrectly copied the file because of a USB driver corruption, caused file to make the 3D Print fail.
Failure appeared a lot like a layer shift on the print gave it the appearance of a under-extruded ghost image.
It also looked as though I had clogged the nozzle, but a quick check proved this to be incorrect diagnosis.
After re-installing the SD Card Reader USB Driver I again moved the G code to the 3D Printer this time with success.
I liked this case so well that I am posting the link here so everyone can see.
This slick little case could be modified for any project.
[www.thingiverse.com]
Made the Front plate without logo
Edited 2 time(s). Last edit at 09/22/2019 08:52AM by Roberts_Clif.
Computer Programmer / Electronics Technician
That MY PC incorrectly copied the file because of a USB driver corruption, caused file to make the 3D Print fail.
Failure appeared a lot like a layer shift on the print gave it the appearance of a under-extruded ghost image.
It also looked as though I had clogged the nozzle, but a quick check proved this to be incorrect diagnosis.
After re-installing the SD Card Reader USB Driver I again moved the G code to the 3D Printer this time with success.
I liked this case so well that I am posting the link here so everyone can see.
This slick little case could be modified for any project.
[www.thingiverse.com]
Made the Front plate without logo
Edited 2 time(s). Last edit at 09/22/2019 08:52AM by Roberts_Clif.
Computer Programmer / Electronics Technician
|
Re: MPCNC March 21, 2020 02:45PM |
This second MPCNC build was kept small as space in my apartment is limited. Some improvements
were made over my 1st build. I left off the homing limit switches and made a waste board with more clamping points. All of the printed parts were done in black PLA+ at 300 micron layers and a 65% infill. I made a extra effort to ensure all of the tubes were cut to exact lengths. The overall build cost was the same at just under $285.00. It is a cheap CNC but the sacrifice is the amount of time it takes to build it. All told it took 9+ hours to assemble and then on top of that is the godawful amount of time it took to print the parts. Rather than trying to reuse the parts from my first one, I sold it on Craigslist, for a little under the build cost, to make room for this new one.
It took a long time to wrap my head around not needing end stops for homing but I finally came to the conclusion that they just aren't needed. Where ever you park the spindle is home after a co-ordinate reset.
One of the custom commands in the Smart Controller menu is to reset all co-ordinates. The co-ordinates need to be reset prior to starting all engraving jobs. They can also be reset by hitting the reset button on the RAMPS board or reset button on the Smart Controller or by powering the controller off and on. Oddly enough adding a G92 X0 Y0 Z0 command to the start G-code has no affect. The only real reason to use X,Y limit switches is for the auto squaring function but with this small of a build I can just manually jostle the axes back into place if need be. And then there is the added benefit of less wires to manage.
Much like printing jobs, some carving jobs take a long time to complete and just like with my printers I prefer
to load my G-code onto a SD card. Since I already had a Arduino Mega 2560 I bought
an inexpensive RAMPS along with a full Graphics Smart Controller and printed out a Prusa type controller enclosure to contain them. I like the enclosure design but the downside to using it is that there is no access to the USB port. So If I need to make changes to the firmware or swap out a driver, I'll need to partially dissemble the box to gain access. I found some black 4 conductor 20 awg PUR tool cord to use for the motor cables but ended up cutting them too short. I didn't figure in the loop back length needed for the cable chain. My intention was to distance the controller well away from the MPCNC so the cooling fan wouldn't suck in so much dust. Now I'll either have to extend the motor cables or add a fan filter.
The spoil board is made out of 1/2" MDF with 25, 5/16 T-nuts at a 95mm spacing.
I use bolts to hold down my work piece . A bit overkill but it eliminates any chance of the work piece from moving during the carving process. Bolting down the work piece at its opposing corners is enough to keep it in place. In the future I hold down clamps will be used in place of the bolts.
I use flat washers as spacers to elevate the work piece above the waste board when doing through cuts to prevent damaging the board. Making the spoil board took a bit of effort and expense and I want to preserve it for as long as possible. I realize this is contrary to the purpose of the board.
Somehow when buying all of the parts I overlooked the purchase of a ACME rod needed for the Z axis. In order to keep the build progressing I used a piece of M8 stainless steel threaded rod and a brass hex nut. These have served me well on my printer builds so I'll probably keep it this way. It works just the same. I only needed to change the steps in the firmware to accommodate this.
I've also switched from ESTLCAM to Shapeoko's Carbide Create CAM software to make the tool paths. The downside to C.C. is that it doesn't have a animated tool path preview although it is more intuitive with a nice library of files to choose from. But it does create the G-code with a .nc extension that my controller doesn't recognize. I have to change the .nc extension to .gcode to make it usable.
I use Oramask vinyl stencil material for masking on the engravings I intend to add paint. It is easy to apply and it leaves a nice clean edge when cut with a V-bit. A End Mill leaves it a bit ragged but good enough to prevent any paint from bleeding into the wood grain. Once the paint is dry the mask is easy to remove. This isn't Pinterest so I'll refrain from posting too many pictures of my still basic carves to keep the administrators happy.
were made over my 1st build. I left off the homing limit switches and made a waste board with more clamping points. All of the printed parts were done in black PLA+ at 300 micron layers and a 65% infill. I made a extra effort to ensure all of the tubes were cut to exact lengths. The overall build cost was the same at just under $285.00. It is a cheap CNC but the sacrifice is the amount of time it takes to build it. All told it took 9+ hours to assemble and then on top of that is the godawful amount of time it took to print the parts. Rather than trying to reuse the parts from my first one, I sold it on Craigslist, for a little under the build cost, to make room for this new one.
It took a long time to wrap my head around not needing end stops for homing but I finally came to the conclusion that they just aren't needed. Where ever you park the spindle is home after a co-ordinate reset.
One of the custom commands in the Smart Controller menu is to reset all co-ordinates. The co-ordinates need to be reset prior to starting all engraving jobs. They can also be reset by hitting the reset button on the RAMPS board or reset button on the Smart Controller or by powering the controller off and on. Oddly enough adding a G92 X0 Y0 Z0 command to the start G-code has no affect. The only real reason to use X,Y limit switches is for the auto squaring function but with this small of a build I can just manually jostle the axes back into place if need be. And then there is the added benefit of less wires to manage.
Much like printing jobs, some carving jobs take a long time to complete and just like with my printers I prefer
to load my G-code onto a SD card. Since I already had a Arduino Mega 2560 I bought
an inexpensive RAMPS along with a full Graphics Smart Controller and printed out a Prusa type controller enclosure to contain them. I like the enclosure design but the downside to using it is that there is no access to the USB port. So If I need to make changes to the firmware or swap out a driver, I'll need to partially dissemble the box to gain access. I found some black 4 conductor 20 awg PUR tool cord to use for the motor cables but ended up cutting them too short. I didn't figure in the loop back length needed for the cable chain. My intention was to distance the controller well away from the MPCNC so the cooling fan wouldn't suck in so much dust. Now I'll either have to extend the motor cables or add a fan filter.
The spoil board is made out of 1/2" MDF with 25, 5/16 T-nuts at a 95mm spacing.
I use bolts to hold down my work piece . A bit overkill but it eliminates any chance of the work piece from moving during the carving process. Bolting down the work piece at its opposing corners is enough to keep it in place. In the future I hold down clamps will be used in place of the bolts.
I use flat washers as spacers to elevate the work piece above the waste board when doing through cuts to prevent damaging the board. Making the spoil board took a bit of effort and expense and I want to preserve it for as long as possible. I realize this is contrary to the purpose of the board.
Somehow when buying all of the parts I overlooked the purchase of a ACME rod needed for the Z axis. In order to keep the build progressing I used a piece of M8 stainless steel threaded rod and a brass hex nut. These have served me well on my printer builds so I'll probably keep it this way. It works just the same. I only needed to change the steps in the firmware to accommodate this.
I've also switched from ESTLCAM to Shapeoko's Carbide Create CAM software to make the tool paths. The downside to C.C. is that it doesn't have a animated tool path preview although it is more intuitive with a nice library of files to choose from. But it does create the G-code with a .nc extension that my controller doesn't recognize. I have to change the .nc extension to .gcode to make it usable.
I use Oramask vinyl stencil material for masking on the engravings I intend to add paint. It is easy to apply and it leaves a nice clean edge when cut with a V-bit. A End Mill leaves it a bit ragged but good enough to prevent any paint from bleeding into the wood grain. Once the paint is dry the mask is easy to remove. This isn't Pinterest so I'll refrain from posting too many pictures of my still basic carves to keep the administrators happy.
|
Re: MPCNC March 21, 2020 03:30PM |
Still procrastinating, have not finished the layout on my controller case. Have all the parts and know how I want them laid out on my case.
Including wiring to the 8 pin mic sockets and plugs, similar to the CR-10 3D Printers.
Still procrastinating testing how to add another stepper motor driver to my ramps 1.6 driver board so I can have a extruder tool for 3D Printing.
And was waiting for warmer weather to test with the Dewalt Dw660 as it is quite loud inside, though I have tested printing a sketch in pencil with a lot of fine detail.
Then I want to place a shed not too far from the house where this project can reside, using a OctoPrint w/cam for year around use to build all sorts of projects.
And maybe I am overthinking this, as My 3D Printer has dual Z stepper motors so I could place my Dual Stepper XY in series and have E0 / E1 for 3D Printing.
As long as I wire them to turn in opposite directions as one spins clockwise the other counter clockwise. An do not use Dual End-stops on the XY Axis.
Edited 9 time(s). Last edit at 03/21/2020 04:19PM by Roberts_Clif.
Computer Programmer / Electronics Technician
Including wiring to the 8 pin mic sockets and plugs, similar to the CR-10 3D Printers.
Still procrastinating testing how to add another stepper motor driver to my ramps 1.6 driver board so I can have a extruder tool for 3D Printing.
And was waiting for warmer weather to test with the Dewalt Dw660 as it is quite loud inside, though I have tested printing a sketch in pencil with a lot of fine detail.
Then I want to place a shed not too far from the house where this project can reside, using a OctoPrint w/cam for year around use to build all sorts of projects.
And maybe I am overthinking this, as My 3D Printer has dual Z stepper motors so I could place my Dual Stepper XY in series and have E0 / E1 for 3D Printing.
As long as I wire them to turn in opposite directions as one spins clockwise the other counter clockwise. An do not use Dual End-stops on the XY Axis.
Edited 9 time(s). Last edit at 03/21/2020 04:19PM by Roberts_Clif.
Computer Programmer / Electronics Technician
|
Re: MPCNC - New Primo July 31, 2021 01:47PM |
I chose the Primo version for my third build. I really like the redesigned parts and think they give the machine a sleek and refined look.
With this new design there are a few differences between the Primo and the Burly.
The Primo uses 10mm GT2 belts instead of 6mm ones and integrates the belt tension adjusters into the corner pieces.
All 608 bearing idlers have been replaced with 20 tooth GT2 idlers reducing the bearing count from 53 to 45.
The conduit spacers (AKA: Spacer_Corner_*_Burly) are no longer needed to assemble the corners.
Despite of having the same footprint as a Burly, the Primo has a slightly smaller work area. It has a overall footprint of 610mm x 610mm but only has a 290mm x 270mm x 77mm cutting area. This is due to the redesigned corners.
My Burlys were built using 84 Oz stepper motors and this Primo was built with 61 Oz ones. There has been no noticable performance difference using the smaller motors. After approximately 70 hours of cutting time there have been no axis skips and the motors don't heat up much beyond room temperature.
All of the Black-PLA and Blue-ABS parts were printed on a Piper printer whose design happens to be based on a MPCNC. Everything was printed with 0.25 layer height and a infill set at the recommended 45% with the exception of the core that has a 70% infill.
Once the printing was finished the Piper was cannibalized for its 608 bearings and stepper motors. The rest of the parts were trashed. The oddball printer had served its purpose.
This now one piece core is a substantial print and took 32 hours. Mine was printed with 0.25 layers and a 70% infill. A couple of hours could have been shaved off had I sliced with a variable 70-30-70 percent infill setting.
I found some 4-pole, M12 receptacles and some 3-meter 90-degree single ended cables from AutomationDirect.com and figured they would work great for this application. I remixed an ancient Minion Controller enclosure from Thingiverse to mount the receptacles. Also the cooling method was changed to use a single 60mm x 10mm fan instead of two 50mm x 5mm fans. Two 50mm fans seemed a little excessive to me. More fans will only suck in more sawdust.
With this new design there are a few differences between the Primo and the Burly.
The Primo uses 10mm GT2 belts instead of 6mm ones and integrates the belt tension adjusters into the corner pieces.
All 608 bearing idlers have been replaced with 20 tooth GT2 idlers reducing the bearing count from 53 to 45.
The conduit spacers (AKA: Spacer_Corner_*_Burly) are no longer needed to assemble the corners.
Despite of having the same footprint as a Burly, the Primo has a slightly smaller work area. It has a overall footprint of 610mm x 610mm but only has a 290mm x 270mm x 77mm cutting area. This is due to the redesigned corners.
My Burlys were built using 84 Oz stepper motors and this Primo was built with 61 Oz ones. There has been no noticable performance difference using the smaller motors. After approximately 70 hours of cutting time there have been no axis skips and the motors don't heat up much beyond room temperature.
All of the Black-PLA and Blue-ABS parts were printed on a Piper printer whose design happens to be based on a MPCNC. Everything was printed with 0.25 layer height and a infill set at the recommended 45% with the exception of the core that has a 70% infill.
Once the printing was finished the Piper was cannibalized for its 608 bearings and stepper motors. The rest of the parts were trashed. The oddball printer had served its purpose.
This now one piece core is a substantial print and took 32 hours. Mine was printed with 0.25 layers and a 70% infill. A couple of hours could have been shaved off had I sliced with a variable 70-30-70 percent infill setting.
I found some 4-pole, M12 receptacles and some 3-meter 90-degree single ended cables from AutomationDirect.com and figured they would work great for this application. I remixed an ancient Minion Controller enclosure from Thingiverse to mount the receptacles. Also the cooling method was changed to use a single 60mm x 10mm fan instead of two 50mm x 5mm fans. Two 50mm fans seemed a little excessive to me. More fans will only suck in more sawdust.
|
Re: MPCNC May 26, 2022 02:26AM |
your picture is nice. thanks for sharing. Go BBQPP goEdited 1 time(s). Last edit at 05/26/2022 02:27AM by Juditolleson.
Sorry, only registered users may post in this forum.