CNC PCB milling
-
I'm unsure what kind of firmware runs on the Woodpecker board, so I have doubts whether it could exploit the Trinamic features. Anyone here know?
So.... I'm looking at a Smoothieboard as perhaps one possibility, although I'm not sure yet as to whether Chillipepr will talkto it.
https://www.panucatt.com/azteeg_X5_mini_reprap_3d_printer_controller_p/ax5mini.htm
Are you looking at this for a 3D printer or for your CNC? If you are looking at it for a 3D printer, have you looked at the RAMPS 1.4 boards? I like their modular design over a fully integrated board for the simple fact that if a part dies, you are only replacing the one part and not an entire board. This make repairs cheaper if something goes wrong.
-
@NeverDie I have one more question of measurements on your CNC. If you look at the pic, can you tell me the measurements of the red and blue arrows on your machine? I have some idea of how tall I want my bracket for the red arrow area, but you were mentioning that you wish you had more travel in that area and I am curious if I have enough room.
Vera Plus running UI7 with MySensors, Sonoffs and 1-Wire devices
Visit my website for more Bits, Bytes and Ramblings from me: http://dan.bemowski.info/ -
@NeverDie I have one more question of measurements on your CNC. If you look at the pic, can you tell me the measurements of the red and blue arrows on your machine? I have some idea of how tall I want my bracket for the red arrow area, but you were mentioning that you wish you had more travel in that area and I am curious if I have enough room.
-
Perhaps slightly off topic but I wanted to share very interesting youtube channel of FRANK from Germany link text who's been working on a PCB pick and place module along with a CNC tool changer in the video above.
Very inginious, and clever work if you've not come accross his videos and work yet.!
-
Perhaps slightly off topic but I wanted to share very interesting youtube channel of FRANK from Germany link text who's been working on a PCB pick and place module along with a CNC tool changer in the video above.
Very inginious, and clever work if you've not come accross his videos and work yet.!
@rfm69 said in CNC PCB milling:
PCB pick and place module along with a CNC tool changer in the video above.
Imho, it's certainly a fun project.. but I think you can go faster by hand ;)
Only a thousands more expensive machine will be faster for assembling, and they are more precise, have a lot more features (cam driven features, more feeders etc.).Pick And Place is not plug and play at all :) there is lot of calibration, preparation. So it's useful and save time, only if you plan volume production. This is why there is often fees when you ask for a pcba, they spend time for calibrating each new board, parts in database etc.
There are multiple different diy pnp, most of them are slow, not enough precise for <0402 etc..
If it's for 0603 size with some dfn, low volume, you don't need this kind of investments I think.. but if you want to tinker it's fun I imagine.For example, if I remember well, someone said he can assemble an easypcb in 30mn, trhough hole. Good but that's the time I need to get assembled&soldered one of my compact smd multisensors boards, no PNP, no handsoldering..
PNP is just for placing, it won't fix test, soldering shorts etc, that still need some agility :)No need of a solder paster dispenser, I use stencils. I like freedom for placing parts, but you can also build a simple manual pick&place in case you shake when placing (no shaking here). With a good magnifier, easy! (I have a big one like dentists, + another one with a cam). Finally, the reflow oven cook it. When I need a repair/reflow -> flux + hot air or fine solder tip. Well organized, you're faster like that. Not the same, if you want to assemble 50 boards, agreed.
So, I've no real xp on PNP, but I'm not new at assembling smd boards. Saying this because I already digged in forums, diy builds, reviews, as I need to invest in a cnc router, and PNP this year. and I'm lucky I can talk with local professionals.. I always "try" to get a good ratio between the need/time/ROI . I'll build the cnc router, but I'll buy the PNP. there are a few interesting but it's still a few thousands dollars.
-
Perhaps slightly off topic but I wanted to share very interesting youtube channel of FRANK from Germany link text who's been working on a PCB pick and place module along with a CNC tool changer in the video above.
Very inginious, and clever work if you've not come accross his videos and work yet.!
@rfm69 Thanks! I've already learned something potentially very useful: instead of doing his etching with an etching bit, he uses a V spiral endmill 30 degrees. Unfortunately, he didn't say the diameter. :(
Interestingly, he didn't even bother with autoleveling.
How to mill a PCB with Chilipeppr s01e02 – 05:34
— Frank Herrmann -
Perhaps slightly off topic but I wanted to share very interesting youtube channel of FRANK from Germany link text who's been working on a PCB pick and place module along with a CNC tool changer in the video above.
Very inginious, and clever work if you've not come accross his videos and work yet.!
-
@rfm69 Thanks! I've already learned something potentially very useful: instead of doing his etching with an etching bit, he uses a V spiral endmill 30 degrees. Unfortunately, he didn't say the diameter. :(
Interestingly, he didn't even bother with autoleveling.
How to mill a PCB with Chilipeppr s01e02 – 05:34
— Frank Herrmann -
@neverdie said in CNC PCB milling:
Interestingly, he didn't even bother with autoleveling
Are you sure? What's happening from 3:25 then?
@yveaux said in CNC PCB milling:
@neverdie said in CNC PCB milling:
Interestingly, he didn't even bother with autoleveling
Are you sure? What's happening from 3:25 then?
You're right. I guess he edited most of it out, or else he did just a partial.
-
Looks as though StepCraft makes a tool changer that you can buy:
Cool Feature! - STEPCRAFT Automatic Tool Changer's Auto Tool Touch Off – 04:29
— Stepcraft USA -
@rfm69 said in CNC PCB milling:
PCB pick and place module along with a CNC tool changer in the video above.
Imho, it's certainly a fun project.. but I think you can go faster by hand ;)
Only a thousands more expensive machine will be faster for assembling, and they are more precise, have a lot more features (cam driven features, more feeders etc.).Pick And Place is not plug and play at all :) there is lot of calibration, preparation. So it's useful and save time, only if you plan volume production. This is why there is often fees when you ask for a pcba, they spend time for calibrating each new board, parts in database etc.
There are multiple different diy pnp, most of them are slow, not enough precise for <0402 etc..
If it's for 0603 size with some dfn, low volume, you don't need this kind of investments I think.. but if you want to tinker it's fun I imagine.For example, if I remember well, someone said he can assemble an easypcb in 30mn, trhough hole. Good but that's the time I need to get assembled&soldered one of my compact smd multisensors boards, no PNP, no handsoldering..
PNP is just for placing, it won't fix test, soldering shorts etc, that still need some agility :)No need of a solder paster dispenser, I use stencils. I like freedom for placing parts, but you can also build a simple manual pick&place in case you shake when placing (no shaking here). With a good magnifier, easy! (I have a big one like dentists, + another one with a cam). Finally, the reflow oven cook it. When I need a repair/reflow -> flux + hot air or fine solder tip. Well organized, you're faster like that. Not the same, if you want to assemble 50 boards, agreed.
So, I've no real xp on PNP, but I'm not new at assembling smd boards. Saying this because I already digged in forums, diy builds, reviews, as I need to invest in a cnc router, and PNP this year. and I'm lucky I can talk with local professionals.. I always "try" to get a good ratio between the need/time/ROI . I'll build the cnc router, but I'll buy the PNP. there are a few interesting but it's still a few thousands dollars.
@scalz Agreed... the P&P is beyond my knowledge level, but what I found fascinating in what and how he was doing the project to make something himself, how he evolved the project was very interesting.
He's got something similar for a tool changer on a 3d printer which is very nicely designed... along with the software and controllers.
-
Looks as though StepCraft makes a tool changer that you can buy:
Cool Feature! - STEPCRAFT Automatic Tool Changer's Auto Tool Touch Off – 04:29
— Stepcraft USA -
@neverdie I saw this stepcraft automatic tool changer, its nice, but the design aproach of Franks is really novel, and way cheapper, simple ingenious. But my skill level is much lower, so I could be wrong :)
-
@rfm69 Does he describe in detail somewhere how to make one? I mean, I get the concept, but he did a lot of work on both the hardware and the software to make it functional.
@neverdie you mean the tool changer, yes he has a blog and git pages where he shares. Its neccessary to speed control with current limiting breaking the spindle, which he's done via cheap ebay motor controller and arduino. Same goes for the spindle locking mechanism powered by a servo. and the CNC software..
Heres the git link text
Heres the blog link text
Its not step by step, perhaps he'll get their but I know he'll be happy if people copy, or chime in :), or add
-
I noticed that Frank is apparently using carbon fiber tubing on his 3D printer. It turns out that carbon fiber tubing is an easy upgrade, if what you have already is aluminum rod:
New 3D Printer - Build Log #16 – 19:14
— Tech2CHowever, its Young's modulus appears to be about 10% less than steel, which I interpret to mean that it is a bit less rigid than steel (cf https://www.christinedemerchant.com/youngmodulus.html).
For the printer in the video that I linked, I can see how reduced weight may be worth the tradeoff.
However, since weight isn't really much of a factor in the 2418 CNC, I think it makes sense to stick with hardened steel rods.
Am I missing something? Opinions?
-
@neverdie once the assembly is done, you should adjust the stepper drivers' current limiting as well.
it is pololu a4988, you can find the corresponding details here:
https://www.pololu.com/product/1182then, it is always good idea to have the basic settings exported from the board, just in case... you can do it by "$$" command sent from the g code sender gui or directly from the serial terminal.
currently I use the following settings, the machine might be able to create nice results with higher feed rates, but I did not have enough time to test it and I sticked to the current working config.
isolation routing with 2001 bits:
- z cut: -0.05mm
- feed rate: 200
you can calculate the V carving bit's tool width for the given milling depth with the following formula:
tan(bit angle/2) * milling depth * 2 + bit's end width
for excel formula the bit angle should be provided in radians, so it should look like this
tan(radians(bit angle/2)) * milling depth * 2 + bit's end widthedge cut or hole milling with the 0.8mm endmill:
- feed rate: 170
- z cut: -1.7mm
- multi depth, depth/pass: 0.2mm
drilling:
- feed rate: 130
- z cut: -1.8
the spindle should be 1000 everywhere.
most probably your board will not have a bootloader, so it will not be possible to update the firmware via usb serial connection (with avrdude), but it is worth to try it. for me it did not work, so I traced back the MCU pins to the pin rows and used ISP to upgrade the firmware to grbl v1.1f (the board will come with 0.9j if I remember correctly). do not forget to export the gerber settings before you upgrade the firmware, as it will loose those, and you have to re-assign the given values again, after the update.
the ISP pinout (from the pin row's top left corner):
Reset -> pin 2
SCK -> pin 3
MISO -> pin 12
MOSI -> pin 135v -> pin1
gnd-> bottom row(!) e.g. pin 1
-
I'm noticing that the tips on 0.1mm etching bits last well enough if the PCB substrate if bakelite, but not very long at all if it's FR4. 0.2mm bit tips on FR4 seems to last much better.
-
I noticed that Frank is apparently using carbon fiber tubing on his 3D printer. It turns out that carbon fiber tubing is an easy upgrade, if what you have already is aluminum rod:
New 3D Printer - Build Log #16 – 19:14
— Tech2CHowever, its Young's modulus appears to be about 10% less than steel, which I interpret to mean that it is a bit less rigid than steel (cf https://www.christinedemerchant.com/youngmodulus.html).
For the printer in the video that I linked, I can see how reduced weight may be worth the tradeoff.
However, since weight isn't really much of a factor in the 2418 CNC, I think it makes sense to stick with hardened steel rods.
Am I missing something? Opinions?
@neverdie I'm gathering parts to make the same 3d printer with the carbon rods. I understood the weight reduction on the moving parts with carbon rods was the reason to go for them. And with the light 3d printer head that model can print much faster because of the stability and lightness. Which perhaps dosn't apply for the 2418 cnc ?
-
I noticed that Frank is apparently using carbon fiber tubing on his 3D printer. It turns out that carbon fiber tubing is an easy upgrade, if what you have already is aluminum rod:
New 3D Printer - Build Log #16 – 19:14
— Tech2CHowever, its Young's modulus appears to be about 10% less than steel, which I interpret to mean that it is a bit less rigid than steel (cf https://www.christinedemerchant.com/youngmodulus.html).
For the printer in the video that I linked, I can see how reduced weight may be worth the tradeoff.
However, since weight isn't really much of a factor in the 2418 CNC, I think it makes sense to stick with hardened steel rods.
Am I missing something? Opinions?
@neverdie Wall thickness as well as the material characteristics determine tube stiffness, comparison on the modulus alone is misleading.
As well as carrying the load, the section must also support it's own weight, one aspect where carbon fibre will out-perform steel. -
What's the best way to check the RPM on my spindle? From looking at similar spindles on-line, I think it might be as low as 6000RPM at 24v.
So.... if I upgraded to a spindle that could go, say, 12000RPM, could I simply increase my feedrate by 2x and be done twice as fast?
