He’s right about acetone smearing the toner across the board and making it permanently look “dirty”. One way to prevent that from happening is to use a brush like he does.

Another method which I have discovered is to use a lot of acetone and wash it off quickly. I put the board on a flat surface, pour a pool of acetone on top of it, and wipe it away quickly with a paper towel. That way the toner remains suspended in the solution and doesn’t [usually] smear on the board.

Posted at 4:25 pm on Nov 9th, 2009 by andrew

I think you have to use pure acetone instead of nail polish remover. I’ve never had a problem getting my toner-transfer PCBs down to crisp bright copper.

And it’s all about the sponge etch method these days, instead of sloshing the PCB around in a bucket. Simply put on gloves, grab your PCB in one hand, an etchant-soaked sponge in the other, and just scrub off the copper. It does get tedious with a very large board, but the average PCB will be done in a couple minutes. Invented (as far as I know) by my friend Jim Newton of Techshop.

Posted at 4:33 pm on Nov 9th, 2009 by macegr

No doubt experiences vary. Then there is “rubbing alcohol”, and the is rubbing alcohol”, it may or may not include lubricating ingredients, sometimes I think that the hardware store alcohol isn’t the cleanest sometimes, but *i* never have had problems with acetone purchased at the hardware store or paint shop. For small critical adhesive jobs I use Everclear, but that’s too expensive for anything, but the smallest jobs. Probably a good simpler method for a large run of boards for am amateur radio club or school classroom project, but pretty involved for the lone ranger doing their own thing. Mr. Burgess *I* approve of this post.

Posted at 5:34 pm on Nov 9th, 2009 by Doug

Every time I see something like this on etching boards I just can’t help but think there has to be a better way to do this. If you watch the whole video you see it takes from the afternoon till what seems like the middle of the night to etch the things. The transfer process is, for lack of a better word, ridiculous. Multiple copies on different papers on different machines, then running it through the laminator a half a dozen times (and pray the mask doesn’t shift), cooling it, etc. It is almost like there is a competition to develop the most convoluted way possible to transfer an image to the board.

I have never etched a board myself, and things like this just keep me farther away from even attempting it, so maybe I have a fundamental misunderstanding of the process. But couldn’t you do something like put a marker in a CNC machine and have it mask off the lines that way? Then just etch in the acid? I have read about masking with lasers in CNC machines in the past, but my understanding was getting a powerful enough laser was the issue currently.

Obviously getting a CNC would be initially more difficult than going to Kinkos, but on large runs like these guys must be doing, it seems like it will pay for itself.

Posted at 5:52 pm on Nov 9th, 2009 by MS3FGX

From my electronics experience, steel wool definitely works the best of any methods to remove toner from your pcb efficiently, while leaving a nice fresh layer of copper ready to solder to.

Posted at 5:58 pm on Nov 9th, 2009 by patrick laing

I am curious about cutting the boards with the paper cutter. I haven’t had time to watch the other videos or have a look at the site so don’t flame me if it’s there in plain English… Is that standard FR4 or is it something easier to cut? I wouldn’t have though it would cut that easy.

Posted at 6:06 pm on Nov 9th, 2009 by Alan Parekh

Alan: it *doesn’t* cut that easily, if it’s normal thickness. Especially if you have a paper cutter that includes any plastic in the construction. You have to apply a lot of force and it’s just not safe, or good for the paper cutter. Scoring and breaking works, or using a metal shop shear.

Posted at 6:20 pm on Nov 9th, 2009 by macegr

@Alan: it’s FR4 board, yeah. Around 0.4mm thickness, though he does suggest the thinner variety will be easier to cut. Having just lost a weekend to making boards that simply ended up in the trash (was using a scroll saw), I immediately had to give this a try…the resulting edge is surprisingly clean and perfectly straight.

Posted at 6:30 pm on Nov 9th, 2009 by Phil Burgess

Not bad. This is nice, as the working method I developed I had to piece together from all over the place. My method:
1.) Laser on National Geographic.
2.) Clean with Comet and a scratchy pad.
3.) Soak and gently rub away.
4.) Shake up in tupperware with HCL and H2O2.
5.) Rinse and scratchy-pad away the toner.
6.) Tin with desoldering iron. (Solder fountain pen)

Posted at 6:34 pm on Nov 9th, 2009 by Alex Grant

here’s my method:
http://www.workinprogress.ca/home-printed-circuit-board-fabrication/

pat

Posted at 7:22 pm on Nov 9th, 2009 by psc

@MS3FGX

Yes, it’s true, nobody in their right mind would use the toner transfer method for production. Cutting away excess copper with a CNC machine isn’t very good either because it’s so slow and is extremely expensive to setup and maintain (those cutting bits aren’t cheap).

The most sensible method for small scale production is photo etching. The copper clad board is pre-coated with a layer of resist which is exposed UV light through a transparency containing the PCB artwork. The board is then developed and etched. Results are very good and repeatable.

Posted at 7:33 pm on Nov 9th, 2009 by lights

Will this work for double sided boards?

Posted at 7:59 pm on Nov 9th, 2009 by Arthur

He seems to be using toner transfer for production.

If you’d ask me, the pulsar toner transfer paper is key–the video might have taken across an entire day, but i believe him that his typical total actual time is 1 hour. I use to do this crap with photo paper transfer, and the whole heat transfer and paper removal process took a good hour or two+, and it didn’t work well most of the time anyway. With the toner transfer paper, I only need two passes through the laminator (probably could use just one–just never tried), and about 5 minutes or less for the backing to come off, and one pass with their TRF foil if I’m doing a fine pitch board. I’ve got a good laser printer already, so that saves the trip to kinko’s. Afterwards it just goes into the etch tank, and I can typically get < 1 hour time to produced single sided board before drilling (a lot of my designs require minimal drilling these days since they're mostly SMT). Great stuff for prototyping, and I'd say that there's definitely a sweet spot in production quantity and target cost where doing this is worth it (I think I've 2-3 other online shops using the same pulsar transfer system for the boards they sell)

Posted at 8:32 pm on Nov 9th, 2009 by threepointone

I’ve made many boards via toner transfer method. I remove the toner with nail polish remover containing acetone. I’ve never had a problem. The boards always come out nice and bright, ready for tinning. They tin perfectly.

Posted at 11:12 pm on Nov 9th, 2009 by Drone

I’ve used ‘Press & Peel’ PCB film (Jaycar part # HG9980) with plenty of good runs. I clean the board (before and after) with running water & steel wool (keep oily fingers off) and press with a household iron set to polyester/cotton. The trick seems to be (apart from cleanliness) to put a sheet of paper between the iron and the film, press firmly until the whole PCB is at the same temperature as the iron and then run the board under cold water to encourage separation. The ONLY time I have had boards fail has been when the Iron was too hot. After peeling the film off, simply etch (Ammonium persulphate crystals in hot water) and then scrub the toner off with steel wool. I never saw the need for Acetone – my way turns out a board from blank to ready to drill in less than 20 minutes.

Posted at 12:00 am on Nov 10th, 2009 by slincolne

I have seen the heating and transfer done a few different ways and always wondered why doesn’t someone just take a few heating coils from some coffee makers, a few sheets of thin steal, some hinges and make the equivalent of a PCB waffle iron?

Heat your top and your bottom and depending on how you weight it and how you do your hinges you can even get even pressure reducing damage on your edges and making this process much easier.

It’s no more difficult to build than an agitator.

Well now I’m going to be down one coffee pot in the morning.

Posted at 12:04 am on Nov 10th, 2009 by Tristra

These days I just order boards for $8 a pop from a guy that does it on big old machine in spare shop time (phototransfer, no soldermask, no metalized holes, just plain board)

Posted at 12:49 am on Nov 10th, 2009 by rasz

nice presentation, but the camerawork…ugh. I got a headache, but came away with lots of good tips.

Posted at 5:52 am on Nov 10th, 2009 by robomonkey

GRRR Every time a toner transfer link is passed around the internets, god kills a kitten >ˆ๏̮๑ˆ<

Why people insist on using this unreliable method that requires esoteric components like "National Geographic paper" when you can use simple, accessible and very consistent photo-etching process? The spread of toner transfer "method" is responsible for the fact that the majority of people believe that making PCB's is a dark art that's impossible to master.

Posted at 8:55 am on Nov 10th, 2009 by svofski

This looks really approachable when you actually see a hands on video, though I’m not so sure about using a paper cutter..those are sort of expensive for this kind of abuse.

Posted at 9:05 am on Nov 10th, 2009 by Skitchin

I’ve made many boards using both photo resist and toner transfer (using high gloss photo paper). I like the toner transfer method better, because it’s faster, uses cheaper boards, fewer chemicals, and less equipment. The toner is also far more resistant to the etchant than the photoresist stuff, so you can scrub at it and leave it in the etchant longer if you need to.

It’s true that you can achieve higher resolution (ie. for fine-pitch SMD) with photoresist, so I haven’t given away my equipment yet. But for general purpose, I’ve found that toner transfer is faster and has a higher yield.

Posted at 9:10 am on Nov 10th, 2009 by jeff-o

@Phill Burgess: I’ve had good results cutting PCB with a scroll saw. The normal blades (metal cutting or wood) DO NOT work. I bought some of the diamond-coated wire blades off Ebay, clamped a piece of wood to the saw table as a guide, and you keep the blade and PCB wet (w/water), then just slowly push the board through the blade, using the wood guide to make a straight line. The blades were less than $5 for a pack of 4, and I’m still on the first one. The blades cut in any direction, but this can be more of a minus than a plus sometimes.

It helps that my scroll saw is all aluminum frame so the water doesn’t bother it much. I just wipe up the PCB dust while it’s still wet.

Posted at 9:58 am on Nov 10th, 2009 by Eric

These videos would’ve been a lot better if they were one video and about 70% shorter. I’m not a member of the ADD generation but this really started to get on my nerves. I don’t need to see people waiting for a copier spitting out papers or watch a copper clad sloowly go through a laminator 5 times. If anything these videos actually made the whole process seem a lot more complicated than it is.

Also you don’t necessarily need the expensive transfer paper. I’ve had great success with paper from different catalogs, one in particular doesn’t even need to be soaked, it just peels right off leaving the clad nice and clean ready for etching.

As for the use of acetone. I’ve been using it for years and never ever had a problem. Use chemically clean acetone and don’t be cheap with it and you wont have a problem. It’s also a hell of a lot faster than scrubbing the heck out of the board that this guy does.

I know I probably come across like a total ass in this post but I’m too tired to be polite tonight ;)

Posted at 10:37 am on Nov 10th, 2009 by sneakypoo

Slightly off topic but I’ve got an old EPROM eraser kicking around, apart from the health risks of exposure to short wavelength UV, would this light source work for photoresist PCB etching?

My eraser is large enough to accept a decent size PCB and has a timer built into the light-tight enclosure.

Posted at 12:01 pm on Nov 10th, 2009 by Lobo

Damn – wrong tab (that’ll teach me for reading multiple hackaday articles at the same time)… reposting @ pcb-light-box-in-a-scanner-shell.

Posted at 12:35 pm on Nov 10th, 2009 by Lobo

Nothing scrubs and degreases better than regular toothpaste btw.

Posted at 1:52 pm on Nov 10th, 2009 by svofski

After trying the toner tranfer method myself and having to struggle through it the first time, things like this tutorial are a good wrap up of tips.

I noticed a lot of things in that video that other people forget to mention, and they are the details that matter most times, they can make or break your boards.

But for cutting PCB’s, would you really want to buy a full-size metal paper cutter or just use tin snips? I’ve never had a problem cutting my boards (some fairly large) with metal shears and I have a perfect edge as long as I’m careful.

Posted at 8:14 am on Nov 11th, 2009 by Formori

Also, a band saw cuts PCBs much better than a scroll saw, and you can cut curves with it should you need to.

Posted at 10:06 am on Nov 11th, 2009 by jeff-o

I always had this question. What are the ways to protect the copper traces from oxidizing ? Coat it with varnish ? Are there any other easy/better methods ? What do you use ?

Posted at 12:29 pm on Nov 11th, 2009 by Copper protection

@Copper protection: You could tin the traces with your soldering iron or use a chemical tinning solution (I forget the name). There are also spray lacquers that will protect the board while still allowing you to solder right through it.
I’m sure there are other methods as well but those are the ones I usually hear about. Personally I’m too lazy to do any of it (my projects don’t survive long enough to get destroyed by oxidizing anyway).

Posted at 3:18 pm on Nov 11th, 2009 by sneakypoo

@Copper protection: just tin it with your soldering iron, it’s very easy to do. It will also thicken the conductive layer and probably cover some tiny crackles in the traces that could be a problem otherwise.

Posted at 5:04 pm on Nov 11th, 2009 by svofski