Well, @MrChips... it seems that the time for me to contribute something of real value to this forum has finally arrived... What I am to give away has cost me years of research and practice and endless headaches... But I've been helped so much by the members of this forum that I'd be nothing short of an egotistical fool if I were to keep this to myself only.

For a long time I tried to develop a quick and reliable way of producing PCBs in my own workshop with mixed results. I tried toner transfer, photo-etching, screen printing, direct printing, etc... and I failed miserably on some of them, and had mixed results with others... UNTIL...

I discovered the holy grail of toner transfer using a paper especially designed for ceramic and metallic surfaces.... I'm not exaggerating.

The paper is called M3 [M-Cubed] Transfer Paper, and it's sold by QLT. They are the only suppliers that sell this miracle paper.
It is not cheap, it costs $60.00 dlls for a pack of 100 A-size (8.5" x 11") sheets, but the price is not off the roof either, especially when you consider its capabilities.
Its performance is, at least for me, nothing short of miraculous. The paper peels off with the greatest ease, does not leave residue on the surface, and does not need to be submerged in a water tray for it to be removed.... Just print, press and peel.... and you're ready for etching.

But you need to know how to use it... and it helps tremendously to have your own heat press, instead of trying to use an ordinary iron. Trust me, the investment is very much worth the results.

So without much further ado, here's the process in full detail.

1. After designing your circuit, print it using a laser printer (not inkjet, nor laserjet, nor any other type of printer that does not use toner) that does not heat the paper when it prints it. Most old laser printers have a built-in hollow roller tube that has a heating element inside whose purpose is to heat the sheet as it passes through, so as to fuse the toner to the sheet.... You cannot use this kind of printer for this application or the paper will fuse to the roller and make a mess inside your printer. What you need to get is a printer that uses laser-diode technology that will only heat the spot on the paper where the toner is, and not the entire sheet. I own an Okidata C3400n Color LED Printer that does the job nicely. Unfortunately this specific printer has been discontinued, but there are many newer printers out there that will work just fine for this... just make sure that your printer uses laser LED technology.
2. Set the printer for maximum resolution and heavy paper, And use the color red for printing all of your circuit's traces. This color of toner is the one that I've found to be the most etch resistant. You may try using other colors (including black) at your own risk.
3. Preheat the heat press to 350°F
4. Clean the FR4 using a degreasing solution, and dry thoroughly. I like to use Windex for that.
5. Make sure that the FR4 is completely clean and that it doesn't have any fibers or dust particles, or tiny pieces of paper towel, if that is what you used to clean it with. If there are fibers or dust on the FR4, then try blowing them off, but do not touch it with your fingers, if you can avoid it.
   
6. In the heat press, place a 1/8" thick sheet of silicon red rubber on top of the FR4 copper clad laminate (I've found that a size of 6" x 4" 2mm thick works best) and preheat the FR4 by pressing it for 75 seconds.
7. After the FR4 has been preheated, lift the silicon rubber and allow it to cool for a couple of minutes. You don't have to remove either the rubber nor the FR4 from the press for this. Just roll the rubber sheet back to expose the FR4 and allow it to cool a little bit, but no longer than two minutes.
8. Cut the 4 x 6" circuit out from the sheet that you just printed. I recommend leaving an extra 1" flap of paper in the final cutout, this will leave you with a rectangle measuring 4" x 7" that contains the circuit. This makes it a lot easier to peel the paper off later on.
9. Place the printed sheet on top of the FR4, and then place the silicon rubber back on top of that.
10. Press for another 75 seconds.
11. Remove the rubber sheet from the press
12. Using something other than your hands (I use a cooking spatula), remove the PCB from the press.
13. Allow the PCB to cool to room temperature. This takes a little over 5 minutes. You can blow some air on it with a ventilator to cool it faster, if you want.
14. Peel the paper off the PCB, and make sure that all traces are perfect and that none of the ink has remained on the paper. If some little bits of ink remained on the paper (which shouldn't have happened, if you followed my instructions carefully) then retouch the traces at the PCB using one of those stinky permanent ink markers.
    
15. Voila! You're done! You're ready to etch your PCB!
16. After etching, do not try to remove the toner from the PCB using a solvent or anything else. Just let the toner be. Do all the drilling that is necessary and soldier all the components on the board, without doing a thing about the toner. The toner will melt away when you apply the solder and all of your components (including SMT caps and resistors) will be soldered with ease. The only exception to this might be SMT chips. For this case I recommend scratching the toner off the PCB at the place where the chip will be installed, and tin those traces with a little solder so you can later solder the chip by gently pressing each pin to its corresponding trace with the soldering tip. This takes some practice to master, but I've done it with chips that have a pin spacing of only 0.025".
17. Also, leaving the toner on the PCB will better protect the traces from corrosion due to moisture.

• One last thing. The transfer paper has a shelf life of about two years. After this time, it becomes harder and harder to peel off, and it might even begin to jam in the printer, which is a real nuisance since you have to open the thing and carefully remove and clean the mess that has stuck inside.

In a later post I plan to show how I built my etching tank, especially designed to process the PCBs quickly, and using the least possible amount of Ferric Chloride (Iron [III] Chloride, FeCl3).

I had bookmarked this post for a looooong time for me to read it when I would have more peace of mind.
I'll be the first to admit that the results are nothing but astonishing but (everyone has a but to provide) I'm not completely sold, based on the cost of the procedure.
For the cost of the special paper ($65+shipping) and the cost of the heat-press, let alone the case where you have an old, unsuitable laser printer, I think the photo-sensitive boards are a better bet.

Granted, the boards themselves are more expensive to boot (about double the price: grobotronics.com/prototyping-cooper-board-100x75mm.html vs http://grobotronics.com/100x75mm.html), but you can print a transparency even at a photocopy machine and you can find a UV chamber under $100, much less if you build one yourself.

But that doesn't let me admire in awe when someone gets the procedure that has tormented many evenings of mine, digging paper residue off the PCB with a toothbruch, just perfect. Congrats!

 

I had bookmarked this post for a looooong time for me to read it when I would have more peace of mind.
I'll be the first to admit that the results are nothing but astonishing but (everyone has a but to provide) I'm not completely sold, based on the cost of the procedure.
For the cost of the special paper ($65+shipping) and the cost of the heat-press, let alone the case where you have an old, unsuitable laser printer, I think the photo-sensitive boards are a better bet.

Granted, the boards themselves are more expensive to boot (about double the price: grobotronics.com/prototyping-cooper-board-100x75mm.html vs http://grobotronics.com/100x75mm.html), but you can print a transparency even at a photocopy machine and you can find a UV chamber under $100, much less if you build one yourself.

But that doesn't let me admire in awe when someone gets the procedure that has tormented many evenings of mine, digging paper residue off the PCB with a toothbruch, just perfect. Congrats!

Stay tuned. I plan to document the whole procedure with pictures, and also show off my etching tank, this week.

 

This is my etching and agitating tank.

I'm looking forward to some pictures.

 

I'm not completely sold, based on the cost of the procedure. For the cost of the special paper ($65+shipping) and the cost of the heat-press, let alone the case where you have an old, unsuitable laser printer, I think the photo-sensitive boards are a better bet.

You don't have to use the press and peel sheets. A method using wax cooking paper has been mentioned on this forum. Some have had success using magazine paper. I use inkjet transparencies, which cost me much less than $1/sheet, and get ~100% toner transfer and don't have to contend with paper fibers.

Some use a laminator, possibly modified, for transferring. The original method was to use a clothes iron. I prefer the iron because I can see through my transfer medium and can apply heat where needed. I also have the option of lifting a corner of the transparency to see how well the transfer is going.

You can't really form a valid opinion on the various methods until you've tried some and found out first hand what works best for you. I tried the photosensitive method while I was in school (we spun on the coating). When I no longer had access to the equipment, I reverted to hand drawing with a Sharpie marker. I did that until I learned about the Toner Transfer method. I did try using rub on transfers and tape for awhile, but that was more tedious than hand drawing (this was long before the advent of tight pitch components). Some people get good results with paper; I've never had a good transfer with any paper medium I tried. That doesn't mean it doesn't work; I just found a medium that worked better for me and stopped looking...

I'd really prefer to mill and drill the boards with a CNC router, but it's unlikely that I'll ever have one. Another method I considered is painting the board and using a laser "cutter" to remove paint, but a laser cutter isn't likely in my future either.

I bought my Wife one of those Cricut cutters; planning to use it to remove paint on copper clad. Unfortunately, the manufacturer shutdown the 2 companies that were selling software to make it a general purpose cutter. Other cutters are still open enough to do what I wanted, but I've already spent several hundred on the Cricut...

Research is being done on printable circuits. At least that would address all of the steps up to drilling...

Commercial boards are more reasonably priced for low quantity. oshpark.com does 3 copies of a 2 sided board for $5/square inch with a couple week turnaround time.

 

You don't have to use the press and peel sheets. A method using wax cooking paper has been mentioned on this forum. Some have had success using magazine paper. I use inkjet transparencies, which cost me much less than $1/sheet, and get ~100% toner transfer and don't have to contend with paper fibers.
[...]

Commercial boards are more reasonably priced for low quantity. oshpark.com does 3 copies of a 2 sided board for $5/square inch with a couple week turnaround time.

So you say that you use inkjet transparencies on a laser printer and then you iron said transparency on top of the board? Did I get that right? Are you using a protective medium between the plastic transparency film and the hot metal plate?

PCB services sure have gotten cheap, but you still have an overhead time of 1-2 weeks where I stay. That's the real show-stopper.

 

So you say that you use inkjet transparencies on a laser printer and then you iron said transparency on top of the board? Did I get that right? Are you using a protective medium between the plastic transparency film and the hot metal plate?

Yes. There was a chance that inkjet transparencies wouldn't tolerate the fusing temps in the printer, but I did it anyway. I've tried half a dozen different brands and didn't have printing problems with any of them. One brand (CompUSA) produced a very dark print, but it wouldn't release well; it would probably be good for exposing photosensitive boards.

I use a paper towel between the transparency and iron.

 

Update.

My good 'ol Okidata C3400n has gone to a better life, and will long be remembered for its dependability and performance that went beyond the call of duty...

That event wreaked havoc in my personal and professional life, since I depended on it 100% to manufacture my PCBs in-house. So I was reluctantly forced into embarking on a quest to find its successor. And after much, Much, MUCH searching and careful consideration, I finally settled on the Brother HL3140CW. And after a lot of testing and calibration, the printer has been performing flawlessly ever since. Its only drawback is that I was forced into using pure Cyan as the printing color, because black or red would not adhere properly to the transfer paper, and would produce very undesirable run-outs and smudges here and there. Cyan as a color does not bother me in the least, but it does not "plastify" itself onto the copper-clad FR4 as hardly as the previous toner color did (red). It prints out a little "chalky" instead, and a bit too easy to scratch off. That means that one has to be very careful after performing the toner transfer process, and before dunking the thing into the FeCl3 tank.

Other than that, I'm quite pleased with it because it prints far faster than the Okidata, and consumables and toner cartridges are much easier to find. And also because it's a very affordable device that sells for less than $200.00 usd.

​

 

Cyan as a color does not bother me in the least, but it does not "plastify" itself onto the copper-clad FR4 as hardly as the previous toner color did (red). Instead, it rather comes out a little "chalky", and a bit too easy to scratch off.

I'm still using my trusty Lexmark Optra E+ BW laser printer. I recently had some transfers that weren't very good and tried to remove the toner with the green scrubber I use to prepare boards. It had very little effect on the toner and I ended up using acetone.

That printer used to be the workhorse before we got a Phaser 6250DP. Now it's only used for address labels, drafts, and toner transfer.

 

Update. The special (and rather expensive) transfer paper is no longer needed for my transfer technique to work. Check this post.

 

I used to do a lot of home made boards myself, and could manage down to10mil tracks with a double sided board. Using UV exposure and 2 laser printed transparencies. It can be hard getting a good solid black on the printer however.
The toner transfer method worked OK, but it was all to easy to have areas and spots where the transfer did not work - I actually used this technique for making labels for anodised aluminium plates on projects instead.

One of the problems I found with the printing was that the printer I used actually skewed the board slightly so effectively the bottom right corner was a bit stretched (and larger) than the top left. (Only a real problem for 2 layer boards).
I had wondered if this was a problem with the software, and back in the day the good people at Cadsoft who made Eagle actually did a test for me, and found this was a common printer problem and varied from printer to printer.

For big work (something I use in the lab at work in fact - Though I prefer to call it "lithography" to make it sound more expensive and high-tech!) You can use one of the silhouette cutters and make a vinyl sticker "transfer" to stick directly to the PCB. You can't get better than details of the order of 1mm or so with (the cheap ones of) these however so drill holes for pins will not work, but the vinyl is very tough, and also makes a very good optically opaque template for UV exposure, for solder masks etc.

I gave up making my own boards nowadays however as:
-the etching chemicals were always getting depleted - and you have to be careful how you dispose of them.
- Its very hard to avoid "via pins" and the only DIY option is wire links or "rivets" which are cumbersome and cannot be easily place under ICs.
- Surface mounting devices are now really the norm, and good tinning on a board for easy soldering of these really need additional tin plating - more processing chemicals
- A good solder mask is really needed for fine work to prevent solder bridging.
- and the final nail.. you can buy 10pcs of 10x10cm PCBs from somewhere like PCBway for just $5! with 6mil tracks and 0.3mm holes

 

Best printers ever made for PCB printouts aim was HP-5th and 4th series Laser Jets. The resolution beyound 600 dps means the colour is thick enough and not transparent like 2400 dps printers have habit to have. The filling ink by the ink-tin is fascinating by default and make everyone open the heart. So, HP-5 or HP-4 and nothing else. HP-6 is still not bad, but much worser. And Canons are bullsh* for people not understanding what speaking. With my Canon Image Runer 2520 I am not able to make ANY quality printout for nor an ironing method, nor Positivus-20 method. The picture contains simply too small amount of ink, whilst on century-old HP-6C at least the quality is rather OK however not so good as at home where I own the 5-th model.

 

RE:""-the etching chemicals were always getting depleted"" --> just apply it with sprinkle and hardly warmed, this extends an etching capacity at least four-fold.
RE:- Its very hard to avoid "via pins" and the only DIY option is wire links or "rivets" --> Yes, and 5000 pieces costs only 30 Eur (Italian, of course what else if not Regio Emilia). Yet it is indeed less beautiful like chemically grewn, them are trustful, stable, strong, and takes only 0,1 mm of heighth.
RE:- Surface mounting devices are now really the norm, and good tinning on a board for easy soldering of these really need additional tin plating"" --> Sure no. Tinning is increasing for the decade at least the resistance, especially the high-frequency resistance of patches. As less to use it, as better will work the 5 GHz technoque. Copper must be mirror-shiny and not to loss this it must be lacked. Chineese photo-sensitive lackeur cost only 5 USD per 100 grams. All the system is identical to Positivus, only if it demands a 11 seconds exposition, the mask demands a 30. More over, the need for tinning basically grews out of the fact that nowadays people havent dislearned how to solder corectly. Just compare the instructions given at 50ies and now. One are speaking about resin and copper tip cleaning with sandpaper, other is forgot about any resin at all, and tip will be dead instantly if touch by sandpaper. However the quality of solder contact, beauty, accuracy, etc any observable parameters are far better within old-school method, and for a while I have no ANY disturbance to solder in old-method even vsop.
RE: - A good solder mask is really needed for fine work to prevent solder bridging."" --> For sure, except the test-pcb where massive changes will be made while adjust. But end version the mask is only way to save the copper oxidation (and sulphatation).
RE:- and the final nail.. you can buy 10pcs of 10x10cm PCBs from somewhere like PCBway for just $5"" --> but wait the month for postal service be completed. And then what to make with those cubic meter of pcb material laying in garage between fire-wood? It costs a rather much money. Therefore I would sum it up like most rational be - robust experiment pcb (like dip with elements of smd like 2012/2520) the ironing, the experimental pcb with vsop and 0402 smd the Positivus plus pistons, but when the count of tyrage is at least 10 or more, then China rise-eater services. And when this pcb is made for outer Space missions, then American or Nederlandian services with golded patches on ceramic base. By the way, hope know the FR4 have tan(delta)=0,03 whilst the best out of Rogers Duroids have 0,00003. Hope understand how important it is for RF, MW and hight reactive power circuits. Roughly 600 USD per A3 sized plate.

 

Best printers ever made for PCB printouts aim was HP-5th and 4th series Laser Jets.

I agree when when using the transparency for a photographic mask. Those older printers used a lot of heat to fuse the toner to the plastic. More modern ones use LED's or similar. Also, the toners may have been different.

For photographic masking, I use a color inkjet printer now.

 

inkjet?????!!! HOW?

 

inkjet?????!!! HOW?

Post #50 is talking about a photographic process. Of course, this 2015 thread started with a question about toner transfer, it has wandered. So, I was responding to the most recent tangent. That is, post #50.

In my opinion, when using a photographic method. inkjets are superior even to laser. However, if one uses laser, the older laser printers are better than the newer ones, because you can darken the laser image with a dry erase ("white board") marker. The toner used with newer laser printers seems less durable and is slightly smeared when darkening with such markers.

 

You mean for MASKING. The even those years when tutors was using an A4 films to get escape of smearing the chalkboards everyone knew the ink-printed films gives a less contrast or other wording, they are more transparent as laser printed (read toner powder printed). And if we talk today that toner on modern printers gives too much transparent blacks, how then about ink-printed even worse? So or Your Positivus-20 is somehow more advanced as mine, or MAYBE the evil resides in that I use an ultraviolet lamp for exposition??? Anyway I need a VERY good black to have a good result even to photo-mask method.

If there I need to use a modern machine, the handy way to get near the needed quality is to set the full colour mode (let the black is composed of colour mixture), plus all toner savers off and darkness set to 115-120%.

 

I am sorry, but I do not understand what you are saying. The language barrier is difficult. But just taking a few sentences,

And if we talk today that toner on modern printers gives too much transparent blacks, how then about ink-printed even worse?

What I said is "ink-printed" (i.e., inkjet) is better, not worse.

So or Your Positivus-20 is somehow more advanced as mine, or MAYBE the evil resides in that I use an ultraviolet lamp for exposition???

I have no idea what a "Positivus" is. I can assure you that the problem is not the UV lamps that I use. In fact, I have had no problems with inkjet-printed masks.

As for the rest of your statement, I generally set the inkjet to highest resolution and may increase the ink amount by a little. The inks used by those printers blocks near UV, which is used for exposure, very effectively.