Printing Custom Circuit Boards With a 3D Printer

10,522

159

14

About: A place to talk about the junk I make

If this isn’t your first time seeing a 3D printer, you’ll have probably heard someone say something along the lines of:

1) Buy 3D printer

2) Print another 3D printer

3) Return original 3D printer

4) ????????

5) Profit

Now anyone with a reasonable awareness of the capability of 3D printers and what goes into making one would realise that this is (mostly) a joke.

Now people have proven time and time again that 3D printing the mechanical components of a 3D printer is very much possible, but there's always one thing which still lives in the realm of the joke - electronics.

So I suppose what I have done here is brought this joke one step closer to not being a joke… because my 3D printer can print out circuit boards.

Note: Now if you’re interested in converting your printer into a PCB plotter then this instruction set should be seen more as a guideline of how I managed to make it work for me, so to advise your design choices. Every printer and slicing software are slightly different, so any mounting components will likely require a bit of creative thinking to make it work on your own set up.

Step 1: I Made a Video!

I made a video, just in case reading ain't your thing!


I've never made a video of this scale before but hopefully it does the job.

Otherwise keep scrolling to the written steps!

Step 2: Permanent Markers

I started off by looking for the finest
tipped permanent marker pens I could find. I found these STAEDTLER permanent Lumicolor markers on amazon, for less than £20 at the time, I also found black permanent markers in the related products but I preferred the blue as they had a stated tip width of around 0.4mm.

Permanent markers (Amazon link)

https://www.amazon.co.uk/gp/product/B000J6ER0Y/ref...

Once the pens arrived, I tested them out on a bit of scrap copper board to see if it would resist the ferric chloride I use as etchant. I found that Sharpie, blue Lumicolor and black Lumicolor all resisted the ferric chloride with little issue.

Step 3: Mounting the Pen to the Printer

The next part was to figure out how to mount the pen to the machine.

I have converted my machine to use the common E3D V6 extruder as the stock extruder packed in and was out of stock for around 7 months, so I was already made quite familiar with the available mounting points on the extruder axis.

I focused on using these two screws which fix the extruder assembly onto the force sensor included on my printer’s extruder carriage.

Some important points that I considered (and in fact, influenced the entire design) was how much force that was going to be applied to the pen tip. The mount needed some vertical movement so that the pen tip isn’t just crushed by the force of the bed pressing into it.

As well as how much wiggle room the pen had in its mount. If the pen has a small amount of wobble either angularly or in X or Y axis, then the drawn lines would be less accurate as the design is plotted, limiting the ultimate precision of the tool.

I have managed to solve both of these problems with a miniature linear rail.

I will be honest, I didn’t buy this thing. I managed to get it by discussing my design with some of my engineer friends when one of them pulled this out of a drawer and donated it to me. I don’t even know how much it would cost if someone wanted to buy one.

This linear rail is precisely machined to the point where I can’t detect any wobble of the rail and is smooth enough that the weight of the pen mount will pull it down under its own weight.

There are always alternatives though, a quick search for “Miniature linear rail” on banggood brought this to surface. It’s a bit long but nothing a Dremel can’t sort out. Other than that the dimensions of the rail seem like a reasonable candidate for the design. Cheap too.

Miniature rail (Banggood)

https://www.banggood.com/9MN-Miniature-Guide-Linea...

I then designed the pen mount to hold onto the pen with a snug fit when pressed in, with suitable screw holes in order to mount it to the linear rail and then subsequently to the mounting bracket.

I also quickly mocked up a little bracket to hold the unused extruder assembly. I didn’t want to unplug it as it caused the machine the shut down with sensor errors. Besides, one of the best ways to shorten the life of your electrical connectors is to make and break connections over and over.

Step 4: Tricking Your Printer Into Plotting

By now I had a pen stuck to my machine and I needed to figure out exactly how I was going to trick my machine into drawing pretty pictures for me.

My first thought was relatively simple, design a 3D model and make it one or two layers thick. That way when the machine tries to print the part, it actually makes the pen trace the entire surface area that I had designed. There was a slight issue with this though, as with my standard slicing settings the extruder simply moves over gaps, but with the pen in place, it will leave a line that traces these moves. I dug through my slicer settings and found a Z lift feature used to minimise oozing from leaky extruders.

I set the value to something high enough that the pen tip will be removed from the surface of the part where ‘gaps’

exist in the design.

While I was making my own slicing profile for the plotter I also turned all temperature settings on the extruder and heated bed down to room temperature as heating isn’t necessary in this process.

I changed my nozzle diameter down to 0.3 in the settings to better match my pen tip diameter. I chose a slightly smaller value to ensure that there was overlap in the lines drawn on large infilled areas. It’s possible that this creates an issue where the fresh line of ink redissolves and damages part of the previous line but I haven’t spent a huge amount of time optimising this process and haven’t fully investigated that concern.

With the slicing profiles set, I proceeded to test the plotter on some paper, then some scrap copper board I had laying around to iron out the kinks in the plotting.

But to summarise the important bits:

> LIFT Z (set to like 5mm +)

> Temperatures set low (otherwise youll heat your bed and extruder for nothing)

> Change nozzle diameter at your discretion (smaller value, higher plot resolution... until a point)

Step 5: Position Your Plot

The position of your plot is quite important when using your soon to be plotter.

We need to know where the machine is going to start drawing our plot and place our copper clad board in that position. To do this I follow a few simple steps:

1) Make a square that is the same size (or slightly larger) than your plot

2) Choose a location for it in your slicing software, if you have a grid that helps otherwise use fixed x/y coordinates

3) Plot the square onto the print bed (might want to cover it with paper or tape when doing this)

4) Position and secure your copper clad board to the print bed using the plotted square as a guide

5) Position your true plot in the same position as your square guide plot

6) Hope youve done it right and start plotting! (dry runs are advisable if you are unsure on your accuracy)

Step 6: Plot Your PCB!

So the next few things to do is iron out any kinks in the plot setting or your pen mount, check for defects and when you have the time try to adjust your set up to eliminate them.

But honestly, we're more or less done. If your plot comes off of the printer cleanly then you may want to inspect your traces and fill in any holes or cut any shorts left behind in the pen ink by hand.

The great thing is once this process is tuned it is clean, safe and relatively quick. Any mistakes are normally followed by some acetone, wirewool and then resticking to the bed for another crack at it.

Step 7: Etch Your PCB!

Now this is not an etching tutorial, so I am hoping that you know how this part goes.


I grabbed the relevant PPE and my favourite etchant - Ferric Chloride. "Favourite" just because that's all I've used.

Dropped the board in a tub, poured on the etchant and shook it about a bit for around 30mins (I did the etch cold, would have been quicker warm).

The output can be seen in the images above. All in all, not too bad if I do say so myself.

I tried to be as honest as possible in displaying defects in the etch, just to show that I have found a method, but not the method. I'm sure there is a ton of fine tuning to be done to get this process up to the next level.

The ink does resist the etchant very well but does seem to break down where it is thinner.

Personally I like to leave the ink on the PCB after etching as a form of silk screen, and it does shift when you go to solder your boards!

I have noticed that the ink as resist only seems to work on copper and brass (I've only had success here) but did not work on steel. I think it's possibly due to the copper content, but honestly I don't know.

Tell me what you think and if you have any questions let me know. I appreciate that this isn't a Lego grade instruction set.

- KdogGboii

Share

    Recommendations

    • Arduino Contest 2019

      Arduino Contest 2019
    • Tape Contest

      Tape Contest
    • Trash to Treasure

      Trash to Treasure

    14 Discussions

    0
    None
    rexty37

    Question 2 months ago

    Bonjour
    Je ne vois pas les fichiers STL dans votre description
    Pouvez-vous me les procurer
    Merci

    0
    None
    moshes12

    3 months ago

    מדועה ארנזיסטור מתח מישתנא עובד אפוךlm317

    0
    None
    moshes12

    Tip 3 months ago

    מדועה זוק מספרlm117 עובד אפוך

    1
    None
    n4mwd

    3 months ago on Step 7

    Nicely detailed. I would be wondering how many boards you can do with one pen before it either runs out or deforms. I'm not sure what's so bad about the chemicals used to process the negative photoresist film. Sodium carbonate for developing and sodium hydroxide for stripping. Both safe enough to pour down the drain. Just don't stick your finger in it.

    6 replies
    0
    None
    KdogGboiin4mwd

    Reply 3 months ago

    Now that you mention it I probably should have kept track of how much use I've given that pen so far... Definitely something to know if this process is to be used!

    You do have a point about the chemicals not being so bad, in fact the one I still use is worst of them. I'll be honest, I use the 2D plotter process so I don't have to think too much. It is without question that the traditional photo resist process gives a better result when done correctly.

    I'm just not very good at it...

    1
    None
    n4mwdKdogGboii

    Reply 3 months ago

    I like Ferric Chloride too. It seems to produce the best results. I tried the home made stuff with Hydrogen Peroxide and Hydrochloric acid, but I found that it undercut my traces too much. Ferric Chloride seems to do a lot better for what I do.

    I found a gerber to gcode converter that might be of use to you on github: https://github.com/pcb2gcode/pcb2gcode I've never used it, but the source code is provided.

    I use negative photoresist film exclusively because its consistently high quality and predictable. The trick to applying it is to clean the board really good and apply the film underwater. Then run it through a laminator to make it stick.

    The guy in the video below has inspired me to try to design a laser photoresist exposer. I wouldn't be using steppers like he did, but I do think he has the right idea to use a raster scan instead of using gcode. Gcode is good for a plotter like yours, but a laser exposer would leave little cracks if it tried to draw this with gcode. Another issue with his comments is that he seems overly concerned about ambiant light ruining his print. However, in my experiments, ordinary incandescent lighting will not expose the film even when put right next to the bulb. It exposes only with UV.

    0
    None
    KdogGboiin4mwd

    Reply 3 months ago

    Oh cool! Thanks for dropping the link to the github, hopefully I can finally stop fighting with sketchup... which I can't stress enough how awful it makes modelling PCBs!

    The laser exposure method really does look like a good few steps up from the marker pen technique I use. I may consider moving in that direction in the future.

    1
    None
    n4mwdKdogGboii

    Reply 3 months ago

    I hope the link is helpful, but if not, it wouldn't be that hard to write a program to convert gerbers into gcode the way you want it. Gerber and gcode are similar, but not exact.

    I can think of complexities that I don't know if the software handles properly. For example, in Gerber, the command would say "Draw a fat line from point A to point B." In gcode, it would have to draw the outline of a fat line and then fill it in.

    I totally agree with you that its better to make a prototype at home rather than to order it from china, wait 3 weeks, and then find out you made a mistake. I recently had to go through three different home made prototypes on a board before it was ready to send to china. I needed 25 of them so it would have been very painful to find out I did it wrong after a three week wait and a good size hole in my wallet.

    Its a bit harder to make double sided pcb's which is what my board was. You mentioned that you tried the routing method of making pcb's with the 3d printer. Probably wouldn't be that hard to modify the router hardware to make a PCB drill out of what you already did. That way, when you plot with the pen, you can adjust it until the alignment with the holes is perfect.

    0
    None
    KdogGboiin4mwd

    Reply 3 months ago

    Slight correction on that- I plan to modify my machine to route using a dremel but havent quite made it that far yet! (Work and university steal a lot of my time)

    But yes I have thought about double sided boards and I agree that keying holes would be the way to do it. I would probably like a build plate with precision located threaded holes as well for said location jigging.

    I definitely intend to do double sided PCBs on this set up so if I have success I may make an instructable for that as well in the future.

    0
    None
    n4mwdKdogGboii

    Reply 3 months ago

    I hear you about the projects stealing time. I've been working on building my Hypercube Evolution for over 6 months now and I'm 99% complete. Seems like everything else gets in the way of working on it. I'm waiting on replacement thermistors as the first batch from china turned out to be fakes.

    Once I get it working, I'll be able to experiment with various ways of making PCB's. Thanks for taking the time to write this instructable.

    0
    None
    alcurb

    3 months ago

    Very nice project. I'm going to give it a go and adapt it to my printer, but I'm going to try not to remove the print head. Might be too tricky.

    I have used the sharpie method to draw resist on a board by hand and found that sometimes etchant leaches under the resist where some simple lines were drawn. I was wondering: Have you tried drawing with a pen at low speed with tighter fill patterns for laying down thicker resist thus avoiding etchant leaching? I don't mean laying resist on top of resist which only serves to dissolve the first resist layer, but rather laying down a tight pattern on the first go, e.g., drawing a line as a narrow box and filling it in rather than drawing a line with one or two line-drawing passes. The ideal might be wrong thinking, but it plays out very well in my head. ;-)

    I have also found, with the manual drawing method, that a clean copper is a happy copper lad...uhm...clad. ;-) The cleaner copper surface will accept the resist better and also avoid etchant leaching to some extent. Once the copper clad is clean, don't touch it with ungloved hands to avoid depositing grease on the copper undoing your efforts.

    1
    None
    PhilTilson

    3 months ago

    I can't believe I have not thought of this before! I have tried making PCBs by fitting a laser to the 3D printer and using it to expose photo-resist. It works quite well, but you need to get the speeds right. I have also mounted a Dremel-type drill and tried milling the copper, but it's difficult to get the repeatability right (due to backlash, I suspect) and bed-levelling is a pain. I must try this method, though I have come across the 'leaching' problem before when using permanent marker to infill large areas. And you're right - going over it again actually makes it worse as it seems to dissolve what's already there!

    1 reply
    0
    None
    KdogGboiiPhilTilson

    Reply 3 months ago

    Its funny that you mention the dremel conversion on your machine because thats exactly why I purchased my printer - The Renkforce RF1000. It was meant to be a 3D printer with laser cutter and milling modules that you could swap the extruder out with. What kind of materials can you cut with it?