EP7 - DIY PCB - Best DIY PCB EVER! (Double Sided SMD)

  Переглядів 12,852

ForOurGood

ForOurGood

Рік тому

So this is where the magic will hopefully happen, I finally make a high density double sided PCB. Will I succeed? Will I fail? Watch the video to find out!
This is the seventh episode in a series where I am challenging myself to make the best possible DIY PCB’s using commonly available low cost CNC technologies.
Below is some detailed documentation about my method.
[SOFTWARE USED]
- KiKad Version 6.0.6 (Circuit & PCB Design)
- FlatCAM Version 8.994 BETA (Generates G-Code, but requires editing)
- Grblcontrol (Candle) Version 1.17 (CNC Control)
[CNC SETUP]
RATTMMOTOR CNC 3018 Pro (with modifications as follows)
- Aluminum 3018 Z-Axis Module + X-Axis upgrade kit
- StepperOnline NEM17 42Ncm Stepper Motors (x3)
- LaserTree Gold 450nm 40W-Pro 12V 20mm Fixed Focus Laser
- Custom end stop switches (x6)
- Custom emergency stop button
- Custom Z & X-Y probe
- Custom cable management
- Custom waste board & PCB Mount + Locating Pins
- 0.7mm & 3mm End mills (Range of sizes on hand is recommended)
- Using ThinkPad E495 to control the CNC
[OTHER EQUIPMENT]
- Elegoo Mercury UV Exposer
- Mini-Bandsaw
- 150 Mesh Silkscreen
- 150mm Embroidery Ring
- 10cm Rubber Spatula
- Fine haired toothbrush
- Ikea glass & plastic containers
- Hand File
[CONSUMABLE ITEMS]
- Ferric chloride etchant solution
- 70mm x 100mm Double Sided Copper Clad FR4 PCB
- Ferric chloride etchant solution
- Mechanic UVH900-WY 10ml UV Resin (White) Fast 10sec curing
- WNB UV-W100 100g UV Resin (Green)
- Conduction CD-LT528 138 degree lead-free solder paste
- Hozan H-42-3717 0.8mm flux core leaded solder
- 0.7mm wire (for vias)
- Isopropyl Alcohol (99.8% or greater)
- Tissue Paper
*********** KiCad Track & Clearance Information ***********
- Grid = 0.3175
- Clearance is factor of 0.5 with minimum of 0.15 (0.1mm laser 50% overlap)
- Track width minus clearance should be factor of Grid (examples follow)
0.3175 - 0.15 = 0.1675
0.6350 - 0.20 = 0.4350
0.6350 - 0.25 = 0.3850
0.6350 - 0.30 = 0.3350
- Fill clearance 0.2mm and min width 0.25. Must be 0.635 from board edge
- Board cutout line 0.1mm thick with 0.635 radius corners
- Push 45 degree tracks so they are snug to each other
- Standalone pads should have a 0.4mm stubby track to sink heat during clearing
*********** FLATCAM FILE PREP ***********
[Double Sided Tool]
- Set Axis Y 2-places
- Mirror operation set W/2,0
- Mirror back layers
- Drill hole 3.6mm
- Set grid 0.5
- Locate 2 holes -2.5,-2.5 from left corners (shift-click to copy value-then +Add)
- Create excellon object
** Mill & laser origin alignment still an issue. I am working on this...
[Copper]
- Isolation Routing
- 0.1mm tool, 18 passes, 50% overlap
- Generate geometry
- Z-travel 0, X-Y feed 500, Z-feed 1000, spindle 4000 (So far this looks good. Possible cooler is even ok, but risky)
- Generate CNC job
- Save CNC Job
- Use Text editor to replace unwanted Z-moves & other
[Silkscreen]
- Isolation routing
- 0.001mm tool, 1 passes, 0% overlap
- Generate geometry
- Paint Tool
- 0.1mm tool, Overlap 50%, Offset 0, Method Standard, Disable connect
- Generate geometry
- Z-travel 0, X-Y feed 500, Z-feed 1000, spindle 2500 (So far this looks good. Possible cooler is even ok)
- Generate CNC job
- Save CNC Job
- Use Text editor to replace unwanted Z-moves & other
- Save modified version with spindle 150 (For setting UV resin) add "_UV_Expose" to filename
(150 works well with fast resin "Mechanic" white, clean with alchohol & final UV expose for 30 secs)
[Mask/Clear Pads]
NOTE: This setting is usable but must be improved.
- Isolation routing
- 0.001mm tool, 1 passes, 0% overlap
- Generate geometry
- Paint Tool
- 0.1mm tool, Overlap 0%, Offset 0, Method (Lines), Disable connect
- Generate geometry
- Z-travel 0, X-Y feed 800, Z-feed 1000, spindle 2500 (Leaves some residue for cold pads, but any hotter is no good)
- Generate CNC job
- Save CNC Job
- Use Text editor to replace unwanted Z-moves & other
[Board Cutout] (Board Cutout line is 0.1mm, round corners)
- Cutout Tool
- Tool 0.1, Cut-Z -0.858, MultiDepth 0.058, Margin 0, Gaps (None)
- Generate Geometry (X)
- Select geometry
- Z-travel 2, X-Y feed 500, Z-feed 1000, spindle 10000
- Generate CNC job
- Save CNC Job
- Make sure cut line well with etched copper zone!
[Alignment Drills]
- Drilling Tool
- CutZ -7, TravelZ 7, FeedZ 50, Spindle 10000
- EndMoveZ 25, PreProcessor (GRBL_11_no_M6)
- Generate CNC Job Object
- Save CNC Job
[Pad Holes]
- Drilling Tool
- CutZ -2.5, TravelZ 1.5, Spindle 10000, Dwell 1
- ToolChange 25, EndMove Z 25, PreProcessor (GRBL_11_no_M6)
- Generate CNC Job Object
- Save CNC Job

КОМЕНТАРІ: 104
@zsigmondkara
@zsigmondkara Рік тому
Your channel is one of those hidden gems, that I love to find before they go big. The almighty algorithm may bless your ones and zeroes!
@ForOurGood
@ForOurGood Рік тому
Well isn't it nice to see a comment like that, thanks so much, I really appreciate it, and hope your right! ;-)
@curtisoneill9929
@curtisoneill9929 Рік тому
Amen
@dennychimwasu9432
@dennychimwasu9432 Місяць тому
Astonished by the quality of the PCB this is the content i was looking for❤
@jtclark5274
@jtclark5274 8 місяців тому
I loved watching your process of not only how you made the final product, but the journey and learning along the way.
@snakezdewiggle6084
@snakezdewiggle6084 3 місяці тому
😎👍 Now I've gotta watch the rest of 'em 😉.
@MrJonbot81
@MrJonbot81 Рік тому
I will stimulate the UKposts algorithm for you.
@ForOurGood
@ForOurGood Рік тому
I very much appreciate the stimulation ;-)
@williamash1849
@williamash1849 Рік тому
An amazing amount of commitment, have followed the series so far and must say its pretty awesome watching the process unfold.
@atarikid
@atarikid 18 днів тому
outstanding! fantastic work and such an interessting video! a big inspiration!
@snitkofb
@snitkofb Рік тому
So close! Can't wait to see how you debug the alignment issue.
@ForOurGood
@ForOurGood Рік тому
Oh no.. now I feel the pressure! Thanks for the support! 😊
@xConundrumx
@xConundrumx Рік тому
forget liquid tin, unless you can get the MG stuff (but that is barred from export outside of the US and Canada last time I checked) the alternatives are all junk. It also won't fix the issue as it also will not adhere to the board for the same reason. What I use is a chemical that is out of complete left field. It's a compound I get from a furniture restoration store that is intended to take the top layers of shellac finish (also referred to as french polish) from furniture (the damaged layers) before new shellac is added to restore the shine. I have no idea what it is made of but have found it is amazing for taking of copper oxidation without actually being abrasive to the copper. A definite plus for pcb's where the copper layer is already thin to begin with. It turns badly oxidized copper into bright shiny copper with just a wipe on and clean off using a paper towel.
@ForOurGood
@ForOurGood Рік тому
Sound pretty interesting, do you have a brand or any other information I could use to research it further? Currently I am using Pikal metal polish, applied with a fine toothbrush, and that seems to work well, and solved the issue I was having. But if there is a better or easier way then I am happy to give it a try! Thanks for the comment.
@xConundrumx
@xConundrumx Рік тому
@@ForOurGood Alas no, it's a concoction made by the store owner himself. As such it only has a name label on it but no indication on what is contained within. At best I could provide a link maybe somehow.
@ForOurGood
@ForOurGood Рік тому
@@xConundrumx no worries, I get it now. Thanks anyway!
@agustinbermudez7027
@agustinbermudez7027 Місяць тому
An amazing work.👏👏👏
@rioferzz
@rioferzz 29 днів тому
This is what I'm looking for
@ForOurGood
@ForOurGood 29 днів тому
Happy to hear that! Thanks for watching 😀
@matheokoning4456
@matheokoning4456 Рік тому
Again a nice and good video. I'm already looking forward to the next one in which you have found the perfect way to make a pcb. That's a reason for me to buy a laser
@ForOurGood
@ForOurGood Рік тому
Step by step, I will keep improving the process, they also say practice makes perfect! Thankfully it seems I am on the right track, with the biggest parts of the puzzle now in place. Thanks for the comment, and good luck with your new laser!
@EUPThatsMe
@EUPThatsMe 8 місяців тому
I've been trying out all sorts of PCB methods and so far they are all too wet/liquid or dusty. This is the best method I've seen so far esp considering the result.
@ForOurGood
@ForOurGood 8 місяців тому
Give it a try and let me know how you go!
@AndyJenkins999
@AndyJenkins999 Рік тому
Massively impressed with your dedication and boy has it paid off. That pcb looks incredible. Well done. Might have to dig out my old 3018 now and have a go it’s very inspiring.
@ForOurGood
@ForOurGood Рік тому
Actually I am really enjoying using this process, it is getting very consistant results and only improves as I learn about and further refine the process. The result is of course important, but if the process is such a hassle that you avoid it, then what is the point right? Glad your inspired, if you break out the 3018 then let me know how you go!
@AndyJenkins999
@AndyJenkins999 Рік тому
@@ForOurGood Thanks very much, yes my 3018 is quite an (very) old one, single 2020 extrusion base frame and uprights with 3d printed spindle and laser carrier. I had modified it years ago to make it stiffer but looking at it now I am not happy the frame is stiff enough so will be working on that and in addition the 10 mm rails for the x axis and carrier for the spindle do flex a bit so I am thinking of fitting some linear rails to stiffen it all up.. I did upgrade the spindle motor and think its ok but need to check the run out to see if it can be used.. All good fun though and if I can get get halfway close to your results I will be pleased. Thanks for the videos, hope your channel goes massive, its very interesting :)
@ForOurGood
@ForOurGood Рік тому
@@AndyJenkins999 Oh, that sounds like a very early version... But I have some good news for you, mine is really no better. Cosmetically I did a few things to it, and added the end point switches, but otherwise the whole thing flexes an moves around like crazy.. using it as a machining CNC is a bit of a joke really. But when using it just with the laser, there is no cutting load, and in this case, it is surprisingly accurate. As long as you can mount a decent laser on it, I expect you will see results close to what I have been getting. Even with all that said, I also plan to do some basic upgrades on my 3018 soon too.. couple of linear rails etc.
@AndyJenkins999
@AndyJenkins999 Рік тому
@@ForOurGood totally agree thanks that’s encouraging. I did get a pretty good level of stability by bolting it down to a box made to house the power and controls so it should be ok and did get some pretty good results milling small hardwood and some small aft aluminium but, it’s just the gantry is like 2020 uprights supported by corner brackets and some 3d print braces and it’s a bit rubbish so I will make some stronger sides for it to beef it up. The Laser should be ok too it was a later purchase I can’t remember the power but looking at your videos it looks like the boards don’t need massive power anyway so think it will be fine. The main issue is the carriage for the laser and spindle they were 3d print versions which I designed and they worked well for a year or so but have been bolted and swapped over so many times they are looking tired and sloppy. Also made the mistake of using PLA so they warped as they got hot. I will need the spindle for drilling but may also use the machine for other small jobs too so will redesign and reprint in ABS now that I started. I think half the fun is modifying the machine in the first place anyway. 😃
@ruslaninst
@ruslaninst 2 місяці тому
Nice content, worth of watching
@cwflemmer
@cwflemmer Рік тому
Awesome, best diy pcb videos i have come across...Please do more of pcb making videos..
@ForOurGood
@ForOurGood Рік тому
Hey thanks for the nice comment! Still a work in progress, so there will definitely be some new videos coming out soon.
@mattiasfagerlund
@mattiasfagerlund 11 місяців тому
Very impressive results!
@ForOurGood
@ForOurGood 11 місяців тому
Thank you!
@lagartocarburante
@lagartocarburante 7 місяців тому
Fantastic UKposts channel, it deserves way more subscribers.
@ForOurGood
@ForOurGood 7 місяців тому
Much appreciated! 🙏
@user-ge1qc5hy8o
@user-ge1qc5hy8o 6 місяців тому
Going to start at the beginning, I like your detail videos.
@ForOurGood
@ForOurGood 6 місяців тому
Thanks for the comment, and I am happy to hear you enjoyed the video!
@user-ge1qc5hy8o
@user-ge1qc5hy8o 5 місяців тому
I don’t recall you laser wattage, please share.@@ForOurGood
@ForOurGood
@ForOurGood 5 місяців тому
@@user-ge1qc5hy8o Please check the detailed information provided in the description of this video, this includes the information about the laser. Thanks for the question!
@Gkid8
@Gkid8 5 місяців тому
Love this, I want to do this myself
@sparky7071
@sparky7071 Рік тому
Very nice! Subscribed!
@twotool
@twotool Рік тому
Loved your video 💛💛💛
@ForOurGood
@ForOurGood Рік тому
Thanks 🙏
@user-xb5zu6zu7j
@user-xb5zu6zu7j 4 місяці тому
Cool stuff!
@ForOurGood
@ForOurGood 4 місяці тому
Thanks!
@enkidu9989
@enkidu9989 Рік тому
nice work.
@ForOurGood
@ForOurGood Рік тому
Glad you liked it! Some of the videos after this one starts to make more practical PCB's using this method. Thanks for taking the time to comment ;-)
@carlanwray8718
@carlanwray8718 Рік тому
This is epic! Thanks!
@ForOurGood
@ForOurGood Рік тому
Glad you liked it! Thanks for the comment ;-)
@carlanwray8718
@carlanwray8718 Рік тому
@@ForOurGood I think I'm going to try to replicate your setup. I wonder if some kind of polishing pad and a cleaning fluid would solve the copper tinning problem. And I'm referring to doing it with the CNC mill. Basically, flood the board with something and then run a program to hit all of the pads with the cleaning pad.
@ForOurGood
@ForOurGood Рік тому
@@carlanwray8718 That sounds like an interesting idea, I will definitely keep it in mind. For now, using the metal polish and fine tooth brush is giving good results and is pretty easy to do. The areas I am focused on improving, is laser clearing the pads, and laser exposing UV resin (see EP8). Both of these are very close, with a clear path forward, but it might be a couple of episodes before I can declare I have fully solved that part of the puzzle. I am very interesting to hear about results from other people trying to replicate this process, so be sure to let me know how you go!
@carlanwray8718
@carlanwray8718 Рік тому
@@ForOurGood I will keep you in the loop. It was really fun watching you go through the process because I had worked all of it out in my head and just wanted to see if someone else had came up with the same thing and you did pretty much exactly what I had in mind. I hadn't thought of the screen printing method for spreading the solder mask but having experience with that I probably would have sooner or later. I thought that was brilliant. For the silk-screen, I don't think I will ever attempt to paint the lettering. I think I'll probably just figure out what laser setting works to make it so it's visible with the right light or a magnifying glass and not worry about it. It's not like you need to reference it all the time.
@jstro-hobbytech
@jstro-hobbytech Рік тому
I think it's incredible. This has been a dream of mine. I ordered a cnc and a laser plus a bunch of bits
@ForOurGood
@ForOurGood Рік тому
Good luck! 🍀
@freemancrowbar
@freemancrowbar 8 місяців тому
@23:26 Just dump the board into the etching solution for half a minute, it will clean the oxidation just fine.
@ABehrooz
@ABehrooz 7 місяців тому
I haven't tried this myself but google says table salt and sour lemon juice might do for removing the copper oxide. I have the same setup with Kicad, candle and CNC3018 and found myself here trying to make PCBs by painting over the PCB and removing a layer. I just don't have the laser and I'm researching ways to do it with V or end mill bits. I'm gonna have to see the rest of your videos to see what you did for height mapping. Thanks for sharing this.
@ForOurGood
@ForOurGood 7 місяців тому
Thanks for the comment. I think I saw someone else make a comment suggesting something similar, but I haven't tried it myself yet. Right now I am using generic metal polish and a super fine hair toothbrush to polish the surface, and it seems to work well (check the later videos). Even so, I will probably test a few of the chemical methods suggested in the comments at some point. If you can, I really highly recommend purchasing a laser and mounting it to your CNC3018, it's totally worth it.
@ABehrooz
@ABehrooz 6 місяців тому
@@ForOurGood I got a 2.5W laser. and about a dozen different acids, reagents, etchants and also nickel and tin plating solutions to avoid the copper oxide problem. Still looking for the mesh thingy. I'll share my results as soon as I get my hands on one. If there is anything You want me to test It'd be my pleasure to do it.
@ForOurGood
@ForOurGood 6 місяців тому
@@ABehrooz Honestly, I am most interested to see how well that lower power laser will work. The silk screening stuff should be no problem to get your hands on, only the silk screen material is somewhat special, but still commonly available. For the ring and spatula I am pretty sure you will find something. Good luck!
@ABehrooz
@ABehrooz 6 місяців тому
@@ForOurGood It cuts the UV paint quite nicely. but struggles if its a little bit too thick (I'm currently applying it with a credit card) which is why I'm going for the mesh solution(Which I too saw on wegstr videos). It is quite funny that I rode a taxi for 10 minutes and came back home with 98% sulfuric acid, 30% H2O2 and other things that would certainly put me on a list in a lot of places but can't find a mesh and silicone spatula.🤣
@jstro-hobbytech
@jstro-hobbytech Рік тому
I just realized. Low melt solder is the problem. I use it for delicate repairs and you have to remove it before proceeding
@ForOurGood
@ForOurGood Рік тому
You are correct. I now also believe the low melt solder was the bulk of the problem (talk about that in a more recent episode). Some of the components appear pre-tinned with higher melt solder, so the low melt solder melts first causing them to float. Even so, some types of components may still shift around, so in this case using thermally activated glue for such components fixes that.
@7yyf
@7yyf Рік тому
Most excellent story-telling! It has all the drama of a good novel :) I don't suppose that you'd be willing to share the KiCad board settings/design rules you used for this project, would you? I'd like to import them and start experimenting myself.
@ForOurGood
@ForOurGood Рік тому
Glad your enjoying watching me fumble my way though this, actually I am still improving the process, but hopefully I should lock most of it soon enough. Generally, I am in a very optimistic mood given the progress to date, though some hurdles are harder to clear than others. I am a bit of a newby with KiCad, so don’t think I am not doing anything special, just setting the grid size and track sizes. I also don’t really have a way to send you a file at this time. Maybe in the future I might setup some public files, but that will be a little further down the road. For now I am adding the settings I used in allot of the videos descriptions (not all), so hopefully there is something useful in there, while keeping in mind this is still a work in progress. Thanks for taking the time to comment.
@matheokoning4456
@matheokoning4456 Рік тому
Too bad you stopped this series. You really had come a long way and you had perhaps the best way to make pcb's. I think I am going to order bare pcb's from jlcp. The advantage is that you then have reasonably good quality pcb's and already provided with a nice soldermask. That saves the mess :-). I think airassist can also help to laser clean the pads afterwards. That will keep the copper a little cooler and hopefully will stay good.
@ForOurGood
@ForOurGood Рік тому
This series is not stopped, it is just on hiatus while I'm grinding away on a different (somewhat urgent) project. These things always take longer than you would like, but it looks like it should be done soon. So hopefully it won't be too long before I get back to this series. Thanks for the comment!
@zorglupsmailing1977
@zorglupsmailing1977 8 місяців тому
I also saw on another video that airassist (blowing air onto the laser spot during blasting) was improving substantially the cleaning of the pad by blowing away the burnt resin while it is in smoke form. This is probably worth a try. Thank you for the AMAZING série !!!! Cannot wait for the next one. Hopefully you came to a point were you could do a wrap up video (or article) summarising all success ?
@timogross8191
@timogross8191 Рік тому
If you have an oxide problem I would give citric acid a try for cleaning the pads. Do not know if it works, but it is a reasonable experiment.
@ForOurGood
@ForOurGood Рік тому
Citrix acid, I might just give that a go, thanks for the suggestion! For now I have been having some good success with a toothbrush and metal polish. Still, it would be interesting if a more passive process was available such as soaking in citric acid. Thanks again!
@rudy5360
@rudy5360 Рік тому
Amazing home made pcb! I found your video while looking at smd pcb design, and progressing from through hole. Is the kicad project available to share?
@ForOurGood
@ForOurGood Рік тому
Thanks for your comment and query. I don't actually have a place to share files at this time. The good news is there really is not much to it, so it is probably worth fumbling around and making your own as a way to learn. In some of my video descriptions I do provide details about the settings and equipment I am using. But as I am still working to improve the process everything is still pretty fluid. Wish you the best of luck with your SMD PCB making adventure!
@user-dd1zf7go6y
@user-dd1zf7go6y 8 місяців тому
I have been working on this exact same process with my 3018 CNC. I ran into the exact same issue you did with there being some sort of film on the pad after lasering away the resin coating. Did you ever find a resolution for that? This series is amazing btw, wish I had found it a few weeks ago, would have saved me a bunch of time, haha.
@ForOurGood
@ForOurGood 8 місяців тому
I think I show it in a later video, but I am using some generic metal polish, and scrubbing the surface with a very fine hair toothbrush, and that seems to fully solve the problem. I have also seen suggestions of dumping the board back in the etchant for a few seconds, possibly that will work too, but generally I prefer to deal with the etchant as little as possible, so I am a bit hesistant to add that as a step in the process. Anyway, good luck and thanks for the comment and question. BTW: How did you discover this video, was it a search result or UKposts recommendation (or other)?
@user-dd1zf7go6y
@user-dd1zf7go6y 8 місяців тому
@@ForOurGood After this comment I watched your next episode. Just bought some metal Polish and I'm going to try it today. My exact google search was "silk screen PCB UV paint". I had seen another video where someone did Screen Printing for the UV paint, so I looked to see what else was out there on the topic, and your video was the top search for this. I haven't gone down the route of etching yet, like you said, I'm trying to avoid chemicals where possible. I am still doing Milling out of the PCB. I have a spring loaded bit holder like you made, that I was originally intending to use for the mask removal, that I decided to try for the milling of the PCB. I used it in combination with a height map, and the results are pretty good (using a .1mm 20 deg V-bit).
@ForOurGood
@ForOurGood 8 місяців тому
@@user-dd1zf7go6y Thanks for all the great feedback, it really helps me understand what is going on with viewers like yourself. I am not surprised you came via a search, UKposts is not generally recommending my content, but seemingly I must be at least making it into search results. Good luck with the PCB milling, personally I didn't like it so much, but it seems a lot of people out there get good results with that method too.
@johnsmith-000
@johnsmith-000 Рік тому
Great series. You obviously don't give up easily. If I'm not mistaken, you never tried to use positive photo resist spray, and only use laser for exposure? I don't know if your laser wavelength is within the required parameters of standard photo resist, but I suppose it might be. If this works, laser exposure would eliminate the need to make a film, and would probably give crispier results.
@ForOurGood
@ForOurGood Рік тому
It is a good question, and something that has crossed my mind in the past. I think the answer is pretty much in the next episode, where I switch to an expose only method for doing the text overlay. What I found is there is allot of bleeding around the core beam that causes unwanted exposure. Now I am working on ways to improve that, such as using a blinder and messing around with the settings, but indications are such that I wont be able to get a crisp result like I get when I simply burn the resin. Now I expect that would be the same for an applied photo resist coated boards, of course an experiment would be required to be 100% sure. But right not I don't have much motivation to work with a exposed photo resist process, because the current process is working great, and because I have had many years experience making boards with photo resist, and it sort of sucks in my opinion. Of course you would now be dealing with another different bunch of chemicals, and adding another step to the process (development), and my exprience has shown that the photo resist can be really delicate, often resulting in defects. Anyway, I would really love someone out there to give it a go and see how good a result they get! Thanks for the comment.
@s_bandera
@s_bandera Рік тому
Great series🎉❤
@ForOurGood
@ForOurGood Рік тому
Thankyou! Hopefully I will have some new episodes coming out soon.
@aureliaonunes37
@aureliaonunes37 9 місяців тому
Hello, I'm Aureliano I live in Brazil, I'm buying Twotrees TTS-55 Laser Engraver to make my prototypes, the idea would be not to use that green liquid mask and use dry film (photosensitive paper) do you think I'll have good results?
@ForOurGood
@ForOurGood 9 місяців тому
Hi Aureliano, nice to meet you, I have many viewers from Brazil. From my experience, I don't think you will get a good result by using the dry film and exposing it with the laser. The reason is because most of these lasers have some light halo around the core beam, and I expect it will make the resolution poor and blurred edges. But I have not tried this method myself, so maybe I am wrong, so please experiment with this method and tell me about your results. For the UV resin method, only the core beam is used, where the power is strong and can burn off the resin, that is why I get a very clean result, and the halo does not affect the result. Thanks for your question and comment.
@aureliaonunes37
@aureliaonunes37 9 місяців тому
​@@ForOurGoodThanks for the feedback, as soon as the equipment arrives I will start the tests and let you know if it works. Hug!
@andreal1061
@andreal1061 3 місяці тому
Great video! Do you have a plan to integrate poper vias in your process?
@ForOurGood
@ForOurGood 3 місяці тому
So there are 2 ways I have seen people doing vias. The first is applying a conductive surface inside the holes and then electroplating, the second is a rivet style. To be honest, I am not so interested in either of these processes. For now I am thinking of just trying to use a much smaller pad and wire. For most cases this is good enough, but does still require the occasional design compromise. I also continue to think about a new way to do a real via process without the hassle, but it is not looking promising yet. Thanks for the comment!
@andreal1061
@andreal1061 3 місяці тому
I have seen an interesting technique where you only drill through the top copper layer (you can use conductive testing through the drill bit and the bottom copper to stop the CNC from drilling the bottom layer too). This creates Wells instead of holes. The Wells can be filled with solder paste maybe and then heating that up could create the vias.
@ForOurGood
@ForOurGood 3 місяці тому
@@andreal1061 oh that is an interesting idea, will definitely keep it in mind, thanks! But no good for connector pins or similar through hole no solder access components. Anyway, you got me thinking about the problem again, so hopefully something will pop out of my brain. Just at the last stages of a major rebuild of my CNC, so hopefully I will be back at it soon.
@rioferzz
@rioferzz 22 дні тому
what is the best type of engraving tools to make pcb routing?, what degree? (15, 30, or 60°?) And what type for making masking text?
@ForOurGood
@ForOurGood 22 дні тому
Hi there! Actually I gave up on trying to mill PCB's after my first try was a disaster, so I don't really have any good advice. What I used was 0.1mm@20deg and 0.2mm@30deg. Currently I am using the laser and etching method I demonstrate in this series, and that has been working very well. Thanks for your question
@jstro-hobbytech
@jstro-hobbytech Рік тому
Some liquid tin to finish it off would be beautiful but it's 200 for a half liter
@ForOurGood
@ForOurGood Рік тому
Metal polish is working fine for me now. Ultimately all the pads get solder on them anyway, so pre-tinning is not so important I guess..
@ShopperPlug
@ShopperPlug Рік тому
Wow. This is great. Didn’t expect green solder mask can actually provide .1mill clean sharp traces by literally burning it off the solder mask. Great research and experimentation. I wonder how UV green solder mask compares to UV black solder mask. I know many people with laser engravers using the red laser uses black ink and actually burns off the ink and then etch off the copper and get decent results. But never at .1mil.
@ForOurGood
@ForOurGood Рік тому
Actually this process surprised me too, a bit of a lucky discovery. The key to the process is that the green solder mask is very robust, I have used photo resist in the past and I always found it very fragile in comparison. Another nice upside is you don't need to clean off the resist layer which could risk damaging the tiny traces, just put one more layer of solder mask and you get one nice looking board! BTW, I am already planning to test white and red UV resin mask soon enough, and maybe using dual colors could give an interesting result too. I'm not sure how black would work, I suspect it would be ok, but maybe having the resin and burnt areas both being black is maybe not a good idea from a process point of view. Thanks for the comment!
@ShopperPlug
@ShopperPlug Рік тому
@@ForOurGood The results you got are really fab grade level. If this is consistent, then it means it can also be used by fab industries too, they can perfect it immensely using dialed and tuned green solder mask concoction and do not need expensive UV lasers and optics. I haven't seen anyone using green solder mask that can get super clean and sharp traces in this PCB industry, maybe you should patent it and start your own business on this, make your own patented green solder mask bottles and prove everyone that it can completely replace photo resit. It also benefits PCB hobbyist not needing specific special UV lasers and UV photo resit ink/film. Photo resist is indeed super fragile and requires many preparations and precision UV lasers and precision CNC gantry are needed. What kind of laser wavelength and setup are you using? It seems you are using high wattage 450nm laser. What kind of beam quality are you getting? Is your laser beam focus point
@ForOurGood
@ForOurGood Рік тому
@@ShopperPlug The laser is a LaserTree 450nm 40W fixed focus laser from AliExpress (apparently more like 10W in reality). At the current resin thickness it seems burning a width of 0.05mm is possible, but typically I am burning at around 0.1mm because that take less time and fits with the accuracy level of the CNC.. The power level and feed speed affects the width of the burn. Another fly in the ointment, is the laser spot is about 15% wider than it is tall, and the G-code generator has no option to take that into account. As we are burning the resist, or the negative, it is possible to get features down to 0.04 mm, but the copper board is 0.035mm thick, so going less than than seems dangerous. In any case the XY gantry has play of around 0.03mm or so, so right now my minimum feature size is 0.1 to 0.15mm. I have ordered some parts to upgrade the precision of the CNC to some degree, but as the objective is to keep the cost low, it is still likely not going to be super precise. As for other things you touched on to do, what you see in these videos is pretty much happening right now, and with producing the videos taking a huge amount of time, I don't have a much spare time, so things take a lot longer to progress than you might imagine (but I will get there eventually!) Thanks for the comment!
@ShopperPlug
@ShopperPlug Рік тому
@@ForOurGood Thanks for the detailed reply.
@ClockDesignCo
@ClockDesignCo Рік тому
Defo old Tony vibes
@ForOurGood
@ForOurGood Рік тому
Got a long way to go to measure up to the great TOT! Thanks for the comment.
@matheokoning4456
@matheokoning4456 Рік тому
Laser almost in and going to try my first attempts via lightburn because I don't have a cnc (yet). The description says you do 18 passes before the etch happens. is this correct? When removing the pads, there is only 1 pass.
@ForOurGood
@ForOurGood Рік тому
"Passes" is this is the name of the setting within Flatcam. This name is confusing and wrong. But that is what it is called! What the setting actually does is change the number of paths that surrounds (isolates) a feature.. These "passes" don't actually overlap of course...Make sure you wear eye protection, these lasers can be very dangerous if you are not careful.
@matheokoning4456
@matheokoning4456 Рік тому
@@ForOurGood Thank you. so if I understand correctly then the laser does not go over the same part 18 times or does it?
@ForOurGood
@ForOurGood Рік тому
@@matheokoning4456 You are correct, the laser only needs to burn over the same path once for resist cleaning and pad clearing.
@matheokoning4456
@matheokoning4456 Рік тому
@@ForOurGood Okay then I get it, thanks again. I saw an overscan option in lightburn. Maybe that is also a solution for you, for the pads that get too hot. Depending on the speed and the overscan percentage, the pad will have time to cool down very briefly after each pass.
@ForOurGood
@ForOurGood Рік тому
@@matheokoning4456 Good luck with your experiments. It is tricky to figure out a system that works. I have not used Lightburn, so will be interesting to see how you go!
PCBs in Two Hours? Zack Sucks at CNC
17:45
Zack Freedman
Переглядів 269 тис.
DIY Double sided PCB with Vias | Best Method
28:23
InvertLogic
Переглядів 71 тис.
EP8 - DIY PCB - Corner Probe for CNC3018 (Contactless)
28:18
ForOurGood
Переглядів 2,6 тис.
Make a PCB with 3D printer, UV led and photosensitive film
14:25
Wil Straver
Переглядів 188 тис.
PCB making and PCB prototyping - 0.05 mm traces/spaces
11:05
WEGSTR CNC
Переглядів 66 тис.
EP7 - CNC3018 Rebuild - DIY CNC Controller
29:45
ForOurGood
Переглядів 680
Makera Carvera Desktop CNC: Great for Beginners ?
26:58
247printing
Переглядів 142 тис.
DIY PCB: Professional Results with CNC + Laser
20:23
El Taller de Álvaro
Переглядів 160 тис.
Applying and exposing dry film photoresist and building a PCB.
36:21
bigclivedotcom
Переглядів 103 тис.
FlatCam 2 sided PCB milling on a CNC
15:46
DIY TECH BROS
Переглядів 73 тис.
"A Machine PERFECT for Making PCBs at Home" Is what they said!
12:29
GreatScott!
Переглядів 481 тис.
Android top🔥
0:12
ARGEN
Переглядів 1,1 млн
Рекламная уловка Apple 😏
0:59
Яблык
Переглядів 799 тис.
Теперь это его телефон
0:21
Хорошие Новости
Переглядів 927 тис.
M4 iPad Pro Impressions: Well This is Awkward
12:51
Marques Brownlee
Переглядів 5 млн