Has anyone had experience of this CNC engraver/drill/mill

Hydroid

Senior Member
Do your parts look like this?
Hi Rick, I couldn't get my photo to upload on the last message. This time, it seems to have uploaded just fine???

This is what I got in the box:


Springs.jpg

The extra nuts are similar to the supplied drive nuts, just with a shorter threaded section...




I'm just afraid I'll run out before I go.:)
I'm certain you're not alone in that thought :)


I've added homing to my machine on the X and Y axis. It's a little more elaborate than a simple switch though. The schematic is at the 5:18 mark in the video.
I'll have to have another look...


The Arduino Uno has a limited number of pins so instead of a home and two end limit switches for each axis, it uses one pin and two switches. The switches are wired in parallel. I tried soft limits but I got into some funky state where I couldn't jog the machine so I turned them off. I wouldn't worry about home and limit switches yet....
Still reading the wiki, over a coffee or two - lots of info :)


The motors don't have enough power to do serious damage
Good thing for a beginner!


Keep in mind GRBL v9 has the Z-limit and the spindle enable pin swapped.
Not sure what you mean. I hooked up the supplied relay board to control the spindle motor and connected it's 'IN' pin to the shields 'SpnEn' pin and I can start stop the motor with 'M3' and 'M5' commands through GRBL...


When you get into using the Flatcam program to produce your g-code, you will be able to specify your feed rate.
Once I get the wiring all soldered, heatshrunk and loomed - nice and neat - that's next on my list. Download and play with Flatcam.

Thanks for all your assistance, I appreciate it!

Regards, John.
 

Rick100

Senior Member
Hello John,

The extra nuts are similar to the supplied drive nuts, just with a shorter threaded section...
They may be intended for use as anti-backlash nuts by using long screws and nuts. I tried a similar setup on my X axis without the springs. I'm not happy with it so I ordered the ones from Banggood. This is my second attempt at ordering anti-backlash nuts. The first ones I got from Ebay are too tight to use. I would just run the machine in your present setup and see what results you get.

Not sure what you mean. I hooked up the supplied relay board to control the spindle motor and connected it's 'IN' pin to the shields 'SpnEn' pin and I can start stop the motor with 'M3' and 'M5' commands through GRBL...
The GRBL version on your Arduino must have been compiled with the VARIABLE_SPINDLE constant in the config.h file undefined. Then it reverts to the normal layout. It makes sense now that I think about it since the software was probably compiled for that machine. I compiled my own version and left it turned on in case I wanted the variable speed control in the future. I'm not using the relay board. I initially used it but had a problem when it crashed my PC when the motor off command was sent. I may have had the USB cable too close to the motor wires though. Did the Arduino come with software already installed?

Thanks for all your assistance, I appreciate it!
Glad to help,
Rick
 

Hydroid

Senior Member
GRBL Loaded

They may be intended for use as anti-backlash nuts by using long screws and nuts. I tried a similar setup on my X axis without the springs. I'm not happy with it so I ordered the ones from Banggood. This is my second attempt at ordering anti-backlash nuts. The first ones I got from Ebay are too tight to use. I would just run the machine in your present setup and see what results you get.
That was my plan. I'm going to get out my dial indicator once it's all done and do some experimenting to check the accuracy of the beast. As long as it's close, it'll do what I need (I think)...


The GRBL version on your Arduino must have been compiled with the VARIABLE_SPINDLE constant in the config.h file undefined.
You've exceeded my knowledge base when it comes to Arduino - easily done :)
I re-read that section on the wiki (the link you provided) and at the start it says:

For Grbl v0.9 with variable spindle PWM ENABLED: (NOTE: The Z-limit and the spindle enable pin are swapped, because we had to access the hardware PWM on D11 for variable spindle PWM output to work.)

I'm assuming that what you said about variable speed (PWM) being disabled on my Arduino must be true...


Did the Arduino come with software already installed?
Yes, GRBL 0.9 was already installed. I just downloaded 'Universal G Code Sender V1.0.9., plugged in the USB cable, set the correct port baud rate and it worked. Good thing, one less thing to figure out :)

I looked at your video again WRT the optical stops to get the home position set. I can make sense of the schematic. It's hard to see in the video, but at one point I think I saw a blocking slide which I assume activates one of the limits as a "coarse stop" and then the other shaft mounted optical stop is used to "fine tune" the zero? I'm assuming too that you wrote some kind of macro to slow down / stop / reverse and re-advance the drive in order to achieve the zero? I've seen those exact optical stops on eBay in my travels, but have never seen the shaft mounted wiper. Is that something you made up?

Regards, John.


Glad to help,
Rick[/QUOTE]
 

Rick100

Senior Member
I looked at your video again WRT the optical stops to get the home position set. I can make sense of the schematic. It's hard to see in the video, but at one point I think I saw a blocking slide which I assume activates one of the limits as a "coarse stop" and then the other shaft mounted optical stop is used to "fine tune" the zero? I'm assuming too that you wrote some kind of macro to slow down / stop / reverse and re-advance the drive in order to achieve the zero? I've seen those exact optical stops on eBay in my travels, but have never seen the shaft mounted wiper. Is that something you made up?
Yes, when both sensors are blocked you are at home. I didn't have to write macros or any code though. All the homing behavior is built into GRBL. You can adjust some of the settings to fine tune it. The collars with the wipers and optical stop mounting blocks (the yellow parts) are designed in Tinkercad and 3D printed on a Folgertech I3 clone I built from a kit. It's construction and electronics are very similar to the engraver. You can even exchange some of the parts.

Good luck,
Rick
 

Hydroid

Senior Member
Yes, when both sensors are blocked you are at home. I didn't have to write macros or any code though. All the homing behavior is built into GRBL. You can adjust some of the settings to fine tune it. The collars with the wipers and optical stop mounting blocks (the yellow parts) are designed in Tinkercad and 3D printed on a Folgertech I3 clone I built from a kit. It's construction and electronics are very similar to the engraver. You can even exchange some of the parts.

Good luck,
Rick
Oh, if it's part of GRBL then it should be easy for me to figure out... That's some "Too much coffee" humour... :)

Hmm, a 3D printer eh... Something else that I don't have.... yet.... Shh, my wife's at work and can't see this... :)

Rick, do you recall where you downloaded Flatcam from. I got the wiring all squared away so i went onto the internet to try and find it. I find all sorts of links to it - on various different websites, but eventually they all lead back to a site called "bitbucket.org" and when I click the "Download the repository .zip file from Bitbucket" link, I get a page that says "Error 404 - Oops, you've found a dead link".... :(

John.
 

Rick100

Senior Member
Hello John,

Rick, do you recall where you downloaded Flatcam from. I got the wiring all squared away so i went onto the internet to try and find it. I find all sorts of links to it - on various different websites, but eventually they all lead back to a site called "bitbucket.org" and when I click the "Download the repository .zip file from Bitbucket" link, I get a page that says "Error 404 - Oops, you've found a dead link".... :(

John.
I just got the 404 error also but I think that's where I originally downloaded it. I followed links from Flatcam.org to / Development / public repository / downloads and ended up here.
https://bitbucket.org/jpcgt/flatcam/downloads/

The file FlatCAM-Win32-8.5-install.exe looks like the same one I downloaded.

Good luck,
Rick
 

Hydroid

Senior Member
Hello John,
I just got the 404 error also but I think that's where I originally downloaded it. I followed links from Flatcam.org to / Development / public repository / downloads and ended up here.
https://bitbucket.org/jpcgt/flatcam/downloads/
The file FlatCAM-Win32-8.5-install.exe looks like the same one I downloaded.
Perfect - Thank You!! Just downloaded the file. Will have to play with it later tonight. Had enough for now...

Just finished playing around with GCode Sender and thought I was going bonkers. Set up my dial indicator quickly (just on the table, so not very accurate) and checked all three axies by moving the drives by one inch.... X and Z drives moved about an inch. Y moved half an inch... Whaa?? Checked the settings within GRBL and X, Y and Z are all set the same, 800 steps/mm... What the heck, same lead screw, same steppers, how can that be??

Played around for about 45 minutes, looking at settings, checking the shield to see if there were any differences that I could see between X/Z and Y to account for it. Then, it struck me to check the actual steppers... They look identical, but the X and Z have model number 42H047H-0504A-003 whereas the Y is 42H47HM-0504A-18...

I see what you mean by "Whatever they happen to had when they ship"... :)

Well, at least I could fix the problem by changing the Y setting to 1600 steps/mm....

Wouldn't be any fun if it were straightforward and simple :)

Thanks again for all the help!

Regards, John.

BTW, even in my rudimentary checks, I definitely see play (backlash) in the drive. May have to try those anti-backlash nuts that were supplied after all. Will wait until after I try and engrave something to see how it turns out first....
 

Hydroid

Senior Member
Anti-Backlash Installed

Well, got Flatcam up and running and managed to export one of my PCB designs through it to UGS. I see a steep learning curve ahead with Flatcam, but at least it has a manual that explains things. Didn't try milling anything just yet, but rather I taped a pencil to the drill chuck and executed the program to get it to trace the cut lines to a blank piece of paper mounted on a piece of cardboard. It was really neat to see it zipping around making the trace. When it was done, it looked pretty good - and would have made a decent PCB. I did notice some funky little curves at points - where the direction changed - and attribute that to the backlash present in the drive system.

I set up my dial indicator and measured the backlash by driving the axis back a few mm, then forward a few mm, then zeroed out the dial indicator, drove ahead one inch and then drove back one inch. In a perfect world, that should have returned the indicator to its zero point, but it didn't. I got a backlash of 8 thousandths of an inch on the Y axis and 6.5 thou on the X.

So, figuring I had nothing to lose in trying the anti-backlash nuts / springs they included, I took apart the X and Y lead screws and installed them:

Anti-Backlash.jpg

I then repeated the same dial indicator procedure to recheck the backlash. With the second nut / spring installed, I got lower backlash numbers; 5 thou on the Y and 4.5 thou on the X. It's better, but I guess I optimistically expected it to be better... Still, pretty good I thought for the price point of this machine. I was going to remove them, but as the numbers are better, I'll leave them on. I was going to do the Z axis as well, but there isn't much room on that drive and I foresaw the PITA factor being too high....

I think I'm going to go Rick100's route down the road and replace the lead screws. Have to try and find a 8mm screw with a 2mm lead, instead of the supplied 4 - and get some proper backlash nuts. These ones work, but what a PITA to install them. I would think that the combination of better nuts and 2mm lead should improve the numbers even more. Plus, I'd like to install a homing system like yours Rick and I'd need longer lead screws to do that...

Having fun, and that's the main thing :)

Regards, John.
 
Last edited:

Rick100

Senior Member
Hello John,
I think I'm going to go Rick100's route down the road and replace the lead screws. Have to try and find a 8mm screw with a 2mm lead, instead of the supplied 4 - and get some proper backlash nuts. These ones work, but what a PITA to install them. I would think that the combination of better nuts and 2mm lead should improve the numbers even more. Plus, I'd like to install a homing system like yours Rick and I'd need longer lead screws to do that...

Having fun, and that's the main thing :)
It looks like your making good progress. Glad to hear you having fun. Your experience with the machine seems close to mine. I had good luck getting lead screws and nuts from this Ebay supplier:
http://www.ebay.com/itm/8mm-Acme-threaded-Rod-Stainless-steel-Leadscrew-T8-Nut-For-CNC-3D-printer-Reprap-/131817911254
I also got anti-backlash nuts from that supplier but they were to tight on the leadscrews so I've ordered them from Banggood. Some anti-backlash nuts are not drop in replacements since the mounting flange diameter and hole pattern diameter are too large. Banggood sells leadscrews with the anti-backlash nuts included but they appear to be the larger size. They sell the proper size ones without the leadscrew.

I assume you will use a wooden spoil board to mount your pcb for milling and drilling. Here's a useful link for generating bed leveling code:
http://www.cncwebtools.com/Apps/Table_Milling_GCode_Generator/index.html
You can produce the g-code in Flatcam but this is easier.


Good luck,
Rick
 

Hydroid

Senior Member
Hi Rick,

I had been looking at the lead screws and anti-backlash nuts on Banggood. I went there looking for 2mm lead and I noticed they also have 1mm lead. It seems to me that the smaller the lead, the more accurate it should be - all other things being equal. But, in looking at it some more, that also means you need more turns from the steppers for the same distance, faster rotation for the same milling speed.... and a few more things that are beyond my understanding at the moment (critical speed of the lead screw and such). I've been reading quite a bit of stuff about this - a lot to learn!

Case in point, I almost ordered the anti-backlash nuts from Bangood because I'm not really thrilled with the method I installed mine (not a fan of the two longer bolt method of keeping the second nut from turning - a PITA to setup). Luckily, before I ordered them, I read enough to understand that a 8mm lead nut won't work on a 4mm lead screw... D'Oh. As well as the anti backlash nut is for ACME thread and it appears that the machine came with ISO screws....

I had good luck getting lead screws and nuts from this Ebay supplier:
http://www.ebay.com/itm/8mm-Acme-threaded-Rod-Stainless-steel-Leadscrew-T8-Nut-For-CNC-3D-printer-Reprap-/131817911254
I also got anti-backlash nuts from that supplier but they were to tight on the leadscrews so I've ordered them from Banggood...
When you replaced your screws from the stock ISO 4mm lead (assuming here that your machine had the same screws as mine), with ACME 2mm lead, so you think you got your improvement in accuracy from the reduction in lead from 4 to 2mm, or from going to ACME from ISO - or maybe a bit of both?

My plan was to go with 250mm long 2mm lead screws and nuts from Banggood and order a second set of maching nuts too. I have a combination lathe / vertical mill and I was going to machine the nuts to give them the slots and fingers - basically convert them into the anti-backlash system as I already have the springs. But now, in reading stuff about lead screws, it seems that ACME screws may have a performance edge over ISO...

Too much info, not enough understanding yet, so I think I'll just stick with what I have for now...

I'll be interested to hear what you think of the Banggood anti-backlash nuts and what accuracy gains you get from them.


I assume you will use a wooden spoil board to mount your pcb for milling and drilling. Here's a useful link for generating bed leveling code:
http://www.cncwebtools.com/Apps/Tabl...tor/index.html
You can produce the g-code in Flatcam but this is easier.

Thanks for that link - Looks great and I've added it to my ever expanding bookmark list of CNC stuff :)

Regards, John.
 

guydu99

New Member
Thanks also @Rick100 for the link to the bed leveling tool. I'm trying to build a low cost PCB milling machine with old printer/scanner parts and my next step is to adjust the bed level. I've bookmarked this helpful link too.
I would also like to point out CAMOTICS (http://camotics.org/) tool which is a Windows open source SW intended for the simulation of GCODE files execution. I run CAMOTICS on my PC before sending the gcode file to the remote PC that hosts the engraver software.
 
Last edited:

Rick100

Senior Member
I had been looking at the lead screws and anti-backlash nuts on Banggood. I went there looking for 2mm lead and I noticed they also have 1mm lead. It seems to me that the smaller the lead, the more accurate it should be - all other things being equal. But, in looking at it some more, that also means you need more turns from the steppers for the same distance, faster rotation for the same milling speed
That was my thinking with regards to accuracy and speed. The quality of the manufacturing may be the most important factor on accuracy though. My first attempts at diy cnc involved 1/4 20 threaded rods as leadscrews and the low speed was definitely a problem. Even with the 2MM lead, the speed would probably be considered slow.

Case in point, I almost ordered the anti-backlash nuts from Bangood because I'm not really thrilled with the method I installed mine (not a fan of the two longer bolt method of keeping the second nut from turning - a PITA to setup). Luckily, before I ordered them, I read enough to understand that a 8mm lead nut won't work on a 4mm lead screw... D'Oh. As well as the anti backlash nut is for ACME thread and it appears that the machine came with ISO screws....
I have to admit I don't know an what an ISO leadscrew looks like. I've been Googling and came up with these pages:
http://www.meadinfo.org/2009/06/metric-acme-lead-screw-trapezoidal.html
http://www.cncroutersource.com/acme-thread.html
https://en.wikipedia.org/wiki/Leadscrew
https://en.wikipedia.org/wiki/Knuckle_thread

Now I'm really confused. Anyway, the threads on the leadsrews that came with my kit look the same as the ones I bought, to my eyes. %-)
What do yours look like?

When you replaced your screws from the stock ISO 4mm lead (assuming here that your machine had the same screws as mine), with ACME 2mm lead, so you think you got your improvement in accuracy from the reduction in lead from 4 to 2mm, or from going to ACME from ISO - or maybe a bit of both?
The ones that came with my kit had a 4MM lead also. The nuts fit very loosely in every direction so I think they were just poorly made. When buying these cheap parts, it's hard to know if you made a good decision or just got lucky.

My plan was to go with 250mm long 2mm lead screws and nuts from Banggood and order a second set of maching nuts too. I have a combination lathe / vertical mill and I was going to machine the nuts to give them the slots and fingers - basically convert them into the anti-backlash system as I already have the springs. But now, in reading stuff about lead screws, it seems that ACME screws may have a performance edge over ISO...

Too much info, not enough understanding yet, so I think I'll just stick with what I have for now...

I'll be interested to hear what you think of the Banggood anti-backlash nuts and what accuracy gains you get from them.
I have a Taig lathe and thought about doing the same thing but I would have to find some springs and set up the milling attachment. I'll let you know how the Banggood parts work. I get overwhelmed by all the info also. After you make your first board, you'll know what you need to fix, if anything.

Good luck,
Rick
 

Rick100

Senior Member
Hello guydu99,

Thanks also @Rick100 for the link to the bed leveling tool. I'm trying to build a low cost PCB milling machine with old printer/scanner parts and my next step is to adjust the bed level. I've bookmarked this helpful link too.
I would also like to point out CAMOTICS (http://camotics.org/) tool which is a Windows open source SW intended for the simulation of GCODE files execution. I run CAMOTICS on my PC before sending the gcode file to the remote PC that hosts the engraver software.
Thanks for the link. I'll add it to my growing collection of CNC software to try out. I made a couple of half-hearted efforts at building a cnc machine from printer/scanner parts but never got it completed. I still have several 8MM rods, belts, and steppers I collected from them. Have you got your machine running yet?


Good luck,
Rick
 

Hydroid

Senior Member
Threads look the same...

That was my thinking with regards to accuracy and speed. The quality of the manufacturing may be the most important factor on accuracy though. My first attempts at diy cnc involved 1/4 20 threaded rods as leadscrews and the low speed was definitely a problem. Even with the 2MM lead, the speed would probably be considered slow.
I'm thinking that when I upgrade, 2mm lead will be the choice. Today I tried upping the step count on my Y axis from 1600 to 6400 to simulate having 1mm lead screw (instead of the stock 4) and the stepper had to spin so fast (to maintain the current stock cutting speed) that it appeared to be jamming. It was actually stopping and starting sporadically and making a grinding noise. I checked the screw and it definitely wasn't binding. My Google efforts came across an explanation to do with high stepper speeds = lower torque due to reduced timing between steps due to the higher speed and the guy equated it to being similar to pole slipping on a generator - something I'm familiar with as I worked in the power generation industry for 32 years.... Anyway, I reduced the step count to 3200, to simulate a 2mm lead, and all was fine. I even used my hand to hold back the table to simulate cutting load and it was still OK.



I have to admit I don't know an what an ISO leadscrew looks like. I've been Googling and came up with these pages:
http://www.meadinfo.org/2009/06/metric-acme-lead-screw-trapezoidal.html
http://www.cncroutersource.com/acme-thread.html
https://en.wikipedia.org/wiki/Leadscrew
https://en.wikipedia.org/wiki/Knuckle_thread

Now I'm really confused. Anyway, the threads on the leadsrews that came with my kit look the same as the ones I bought, to my eyes. %-)
What do yours look like?
Same here, don't know too much about them either. I have been looking at this site:

http://www.cncroutersource.com/acme-thread.html

and I can't even see what the difference is between what they call a stub-acme vs an acme. As for ACME vs ISO, it appears the difference is one degree on the thread angle. My eyes have NO chance of seeing that difference. My lead screws look just like the ones on the machines photo on the gearbest website - and to me, they also look the same as the 2mm lead ones on the eBay link you sent me....



The ones that came with my kit had a 4MM lead also. The nuts fit very loosely in every direction so I think they were just poorly made. When buying these cheap parts, it's hard to know if you made a good decision or just got lucky.
Mine were a bit loose too. Yes, it's a crapshoot when ordering from CN. It usually takes FOREVER to arrive and it's hit or miss on quality. Another case in point: I received some mini TV remotes that I ordered for a clock project I'm working on (standard IR remote to change settings etc). I ordered two. One was DOA and the other had a dead battery. So I took the good battery from the dead remote and essentially made one good remote from two... Not even worth returning because they were so cheap (under $4 total)....



I have a Taig lathe and thought about doing the same thing but I would have to find some springs and set up the milling attachment. I'll let you know how the Banggood parts work. I get overwhelmed by all the info also. After you make your first board, you'll know what you need to fix, if anything.
I have to wait to go any further on my machine. I made up a board to mount on the machines table, but I can't mill it flat just now. The machine came with three bits, but they're all the same - pointed 'V' @ about 15 degrees... I ordered some 3/16" end mills from Banggood last night. When they arrive, I'll be able to continue...

Regards, John.
 

guydu99

New Member
Hello guydu99,



Thanks for the link. I'll add it to my growing collection of CNC software to try out. I made a couple of half-hearted efforts at building a cnc machine from printer/scanner parts but never got it completed. I still have several 8MM rods, belts, and steppers I collected from them. Have you got your machine running yet?


Good luck,
Rick
Hi Rick100,
It runs with a pen. X and Y axis are OK. I have eliminated the X/Y backlash with the SW (LINUXCNC).
I use 2 flatbed scanners for the Y and X axis.
The Y axis is made with an Epson Perfection 610 (1.8° bipolar stepper motor) base.
I have cut the base of an Epson Perfection 1250 for X axis (1.8° unipolar stepper motor) and put it on the Epson 610 base.
I think that the Z axis will be the most difficult mechanical part to do.
It's currently made with a DVD reader and a motor (probably 7.5°of step angle) from a printer, I do not remember which one. I've installed the spindle motor but it's not yet wired.
Should this Z axis not work I will replace it with 8mm rods. I already have the parts, including a stepper motor (1.8°) salvaged from an Epson Stylus C64 printer.
One can find the type of motor in the Epson service manuals which are easy to retrieve on the web and It worth looking at such kind of document before disassembling the product.
Some pictures of my engraver:



Regards
 

Attachments

Hydroid

Senior Member
Hi Rick,

While I'm waiting on parts, I'm going to make up a spoil board. I had a look at your U-Tube video and I assume your spoil board is glued to the carrier board which is then attached to the machine's table by wing nuts.

How is the board being milled in the video being held on the spoil board? I don't see any hold-downs. Double sided tape?

Regards, John.
 

Rick100

Senior Member
Hi Rick,

While I'm waiting on parts, I'm going to make up a spoil board. I had a look at your U-Tube video and I assume your spoil board is glued to the carrier board which is then attached to the machine's table by wing nuts.
Yes, it's just a piece of 1X4 glued to a piece of 6MM plywood. My first test was made with scrap piece of particle board without leveling. I just touched off(Z axis zeroed) the bit on the wood and let it route the pattern into the wood.

How is the board being milled in the video being held on the spoil board? I don't see any hold-downs. Double sided tape?
Yes, Scotch double sided tape. I've had problems with the tape sticking to the wood, so I've started covering the 1X4 with clear packing tape. The double sided tape sticks to the pcb material just fine. The main problem I have now is that the pcb material isn't flat. I've heated it with a heat-gun and bent it the opposite way of the bow and that helps a little. I will try the autolevel (probing feature) of ChilliPeppr sometime soon.

If you come up with any better methods, I would like to hear them. I'm just using the materials I have on hand.

Good luck,
Rick
 

Rick100

Senior Member
Hello guydu99,

Some pictures of my engraver:
Looks like your getting a lot out your old printer/scanner parts. It would be nice if GRBL had backlash compensation like LINUXCNC. I am looking forward to finished pcb results.

Good luck,
Rick
 

mikeyBoo

Senior Member
CNC Stuff

hi Rick, I attached a pdf showing my setup. Might give you some ideas. Unless I'm making boards larger than 2.25in wide, I usually just use the vise.
Hope your project works out the way you want. Have fun!
 

Attachments

Rick100

Senior Member
Hello mikey,

hi Rick, I attached a pdf showing my setup. Might give you some ideas. Unless I'm making boards larger than 2.25in wide, I usually just use the vise.
Hope your project works out the way you want. Have fun!
Very interesting methods and machine. I used a small machinist vice when I drilled my boards on a Dremel drill press. Is that a 3020/3040 mill or something else? I wonder how many people on the forum have cnc machines?

Thanks,
Rick
 

Jeremy Harris

Senior Member
I wonder how many people on the forum have cnc machines?
I have one of the small Chinese CNC desktop routers, a Sable 2015, but I've been too busy with other stuff to get it set up and working. I have the stepper drivers, power supply, spindle etc, just need to find a few hours to concentrate on getting it all connected up and working, and learning how to drive it.
 

mikeyBoo

Senior Member
more of my CNC ramblings

Hello mikey,



Very interesting methods and machine. I used a small machinist vice when I drilled my boards on a Dremel drill press. Is that a 3020/3040 mill or something else? I wonder how many people on the forum have cnc machines?

Thanks,
Rick
Yep, it’s called a CNC 3040. The electronics & software are early Flashcut running on an old Sony Win98 laptop. The advantage to using a PCB layout app for milling & drilling is that your circuit board artwork & the enclosure it fits in are the same file.
I use unused inner layers for a drill layer, mill layer 1, 2 etc. All are referenced to the same lower-left datum so it’s easy to see how it all fits together. Each layer is in the Z (depth) dimension, so you can overlay the layers for the bottom & top enclosure cutouts. If there is an enclosure I want to reuse, I just make a “part” of the milling layers & it’s ready to go for the next project.
When making milling layers, it’s very convenient to have relative coordinates (i.e. ability to make any point on a drawing = xy 0 0). It’s also important to be able to move mechanical “parts” with the cursor keys to snap to a grid (I usually use a 0.025 & 0.005 snap for both circuit board & enclosure so everything “locks” together). Unfortunately, DipTrace has no relative grid or snap move (albeit it makes nice boards) which is why I use the ancient Tango PCB most often. Kicad has both features, but it’s still a little quirkly.
It’s baffling to me that the old Tango blows the doors off any “modern” PCB software (except the pricey Altium) for getting things done quickly. I guess sometimes newer ain’t better.
I use Coreldraw to print out circuit-board artwork (because I need scaling corrections for my laser printer) from postscript (best) or .pdf files from the PCB layout software. Diptrace has clunky pdf output, but it can be made to work.
If I were starting from scratch, I’d buy an old full version 6 copy of Coreldraw along with the DXFTool extension
http://www.coreldrawtools.com/standard/index.html
Geez… didn’t mean to write a book (there’s a lot to this CNC stuff).
Don’t know how many folks are interested in CNC on this site, so forgive me if I’m running on.

Also, it would be possible to build a cheap drill machine using a Picaxe controller, but I don’t have a need for it. The problem with going cheap (i.e. not using ballscrews) is backlash, The only way to prevent backlash with Acme screws is to never reverse directions. This could be done by mounting a stepper directly to a circular or square plate that steps in small increments in a circle, so the first coord is a step on the circle. The 2nd coord is a stepper that moves from the outside of the circle inwards (picture a CD-ROM drive). The spindle is stationary & only moves up/down. The coords could be “picked up” from an imported drill file on a PC, then sent to the Picaxe. This would be an inexpensive piece of machinery, but there would be no way to justify the design time unless it was made for resale (an opportunity for some bright young engineer).
Anyhow, it’s nice out, so I’m gone fishing.
 
Top