Need Laminator recommendations for Toner Transfer

[moffy]

I bought one of those cheap Chinese bench drills about 20 years ago, the starter cap failed recently which I replaced. Normal chuck takes up to 10mm, it is only recently I have realised how good it is. It will take a 0.8mm drill bit quite happily, and when running there is no noticeable wobble. I use it for my pcb's. I have also found that if I leave the toner on after etching, each pad has a hole etched as well, that the increased contrast helps me find the centers so much easier. It is the little details that can make things easier.

[Miti]

Which is the precision you can achieve? Which is the smallest track you get?

Take a look at this thread.

https://www.eevblog.com/forum/beginners/what-is-the-best-way-to-make-printed-circuit-boards-at-home/25/

[moffy]

Miti,
Thanks for the recommendation. I just ran across this from someone advising  me about another project: https://hackaday.io/project/26216-pcb-exposer
Incredible, 1200dpi resolution direct to the PCB exposure film. Home made and remarkably simple and cheap.

[james_s]

Proper toner transfer sheets will cost the same, if not more than photosensitive PCB.  Therefore, why not just use the photo sensitive PCB method?

That would just work no matter what, sometimes cheaper and with no tricky thermal transfer required.   

They're way cheaper when you use them properly. You don't use a whole sheet for every board, you print on plain paper first, then cut a piece of transfer sheet just a little larger than the printed area and tape it at the leading edge over what you just printed. Run the same sheet through the printer again to print on the transfer sheet. It took me several years to go through my first pack of transfer paper, then not long after I got my second pack I pretty much stopped etching my own boards. It's just so cheap to have professionally made boards done.

[moffy]

The boards are cheap, but the delivery cost and time delay are quite a negative if I can get away with single sided boards.

[SilverSolder]

Large boards (e.g. "motherboards" that provide interconnects with other stuff), and special thick/strong PCB, is still expensive to get made by the usual suppliers compared to DIY, since they insist on making 3 or 5 of them which adds up very fast with large boards.

I've tried various DIY methods, they all have pros and cons and can be made to work: laser printer/toner transfer, inkjet transparency/UV mask, vinyl cutter... 

[SilverSolder]

Well, I've bitten the bullet and ordered a laminator:

4 roller, variable temp, up to 1mm thickness(better than most others), hopeful it does 1.5mm and under $100 AUD(just). Will let you know how it goes when I get it.

I have a variation of that same type of laminator except mine has a digital display, can be set up to 350F and the temperature is reasonably accurate (measured with IR thermometer).  It works really well for toner transfer on large boards. 

[moffy]

Nice. I really love this new laminator.

[SilverSolder]

The main limitation I've found with toner transfer is that the printout is accurate/reproducible to within about 0.8mm across a US Letter size page.

The best accuracy is in the top left quadrant of the page on my printer (Laserjet 2300)

The distortions due to roller slippage means the paper rotates, shimmies, loses positional accuracy as it worms its way through the paper path.  The heat also does things to the paper, especially if it is humid.

Basically - toner transfer has its limitations - don't use for large double sided high precision boards, and don't expect lithography level results!

Inkjet printers are significantly more precise, so for anything "big" it is worth reaching for the gel-coated transparencies and going optical...

[james_s]

The horizontal axis of laser printers is normally very precise, I've encountered printers that had lousy accuracy on the vertical axis though. A printer I used to have seemed to noticeably degrade in this regard, formerly I made some fairly large boards, ~8" long without issues but toward the time I replaced it I was having trouble with even 4" boards. I keep my toner transfer supplies around for the occasional prototype or one-off, it's really handy for stuff like SMT adapter boards and quick evaluation boards. It's increasingly rare that I can be bothered to get everything out and use it though.

[SilverSolder]

The problem is the consistency - the error isn't a problem on a single sided board, but when you try to line up a double sided layout, a 0.8mm error (vertical) can become a real issue for a "tight" design.

My inkjets have no problem keeping repeatability to well below 0.5mm -  I guess the heat applied by the fuser just does stuff to paper... 

[james_s]

I don't think it's the heat applied by the fuser, the old printer I had used to be very good and then it degraded. I suspect the issue is that the short axis can be very precise as the laser beam is scanned by a precision polygon mirror with closed loop feedback but the long axis is just a stepper motor in many cases without precise feedback.

[moffy]

Have you tried different papers to see if it is the heated roller that is warping the paper?
I did a run where I taped the yellow toner transfer paper to plain paper and ran it through the printer. There were little creases where the 2 papers had expanded at different rates. Then I just tried running a full yellow sheet through, no creases. I just wonder if a heat resistant paper would do a better job?
By the way at 50c per page for the toner transfer paper I don't mind the cost if I get a good board as a result.

[SilverSolder]

Try this experiment:  print a full (A4 / letter size) big rectangle made with the thinnest pitch line, as big as will fit on the paper (not black filled, just a plain white rectangle)...  Then over-print that page again, with exactly the same image.

If the two align perfectly....   you have perfect paper, perfect pinch rollers, etc.  - please tell me what printer you use, if that happens, so I can get one! 

[SilverSolder]

Have you tried different papers to see if it is the heated roller that is warping the paper?
I did a run where I taped the yellow toner transfer paper to plain paper and ran it through the printer. There were little creases where the 2 papers had expanded at different rates. Then I just tried running a full yellow sheet through, no creases. I just wonder if a heat resistant paper would do a better job?
By the way at 50c per page for the toner transfer paper I don't mind the cost if I get a good board as a result.

I do use toner transfer paper, I agree it is worth the price.

[Refrigerator]

I use an Ilam Touch A3, that i found at the dumpster, together with extra glossy photo paper.
Laminators need their temp to be bumped up a little but i just use some heatgun on the board instead and it works ok.
Use sandpaper to key the copper, scouring pads won't cut it (pun intended). The toner just sticks better this way, 400-600 grit works fine.
After laminating peel the photopaper off dry, it comes off easily, no need for any water nonsense.
With this method i get reliable traces down to 0.2mm.
Pic included is what i can get, the defect in the trace is because my printer is damaged a little.
I had some better pics but i can't find them but you get the idea.
It's an adapter for a QFN-24 package IC.

[Refrigerator]

Found more pics of the process.

[SilverSolder]

I have attached a PDF of a "rectangle test page" to reveal how "good" the paper path of a laser printer is (and how "good" the paper itself is, for that matter).

Basically, print this page twice on the same piece of paper.  Do the two prints overlap perfectly?

[moffy]

Personally, rather than run the same piece of paper twice through the printer, I would run two pieces of paper once and see how they align, same amount of heat and shrinkage etc for both runs. Trouble with the yellow paper is that it is isn't very transparent. Not sure how to solve that.

P.S. Might try some of these: https://www.officeworks.com.au/shop/officeworks/p/nobo-plain-paper-copier-transparency-film-20-pack-acpp100c20#specifications
made from PET  should be heat resistant.

[moffy]

I went out to Officeworks, bought the transparency films and did three consecutive prints. Over the whole page you are talking about an error of the thickness of the border line drawn, maybe 0.25mm?, it is hard to judge because they are so close. One problem with the film is that after the print it attracts every bit of dust around. Don't know how you would clean that. So I definitely think it is the paper that is the issue not the printer. The difference between inkjets and lasers, one heats the paper and the other doesn't, pretty similar mechanisms ( except for printhead vs laser, and heater/static drum). But they both use rollers to move the paper through the printer, so apart from shrinkage I would expect similar  accuracy .

[SilverSolder]

I went out to Officeworks, bought the transparency films and did three consecutive prints. Over the whole page you are talking about an error of the thickness of the border line drawn, maybe 0.25mm?, it is hard to judge because they are so close. One problem with the film is that after the print it attracts every bit of dust around. Don't know how you would clean that. So I definitely think it is the paper that is the issue not the printer. The difference between inkjets and lasers, one heats the paper and the other doesn't, pretty similar mechanisms ( except for printhead vs laser, and heater/static drum). But they both use rollers to move the paper through the printer, so apart from shrinkage I would expect similar  accuracy .

When I do this on toner paper, I don't get more than 0.8mm accuracy on a big page...

If you are doing transparencies, the inkjet is the way to go (around here, anyway) - it seems capable of 0.15mm or even better accuracy.


Then there is the resolution...  I have attached a couple of resolution tests that will tell you how good the printer is at resolving thin lines, should be printed both vertical and horizontal orientation as the resolution might be different.

[moffy]

If you are doing transparencies, the inkjet is the way to go (around here, anyway) - it seems capable of 0.15mm or even better accuracy.

I think you missed the point of what I was doing. I was trying to differentiate between the accuracy of the laser printer vs the paper/transparency. I am not doing transparencies, they were chosen as a stable medium (PET) that allows easy overlay and comparison and would test the laser printer accuracy with minimal distortion due to the media.
The result is that the laser printer on the test page provided, has a maximum error of around 0.25mm over the whole page. The 0.8mm errors you reported, I am pretty sure, unless I have an exceptional machine, are due to paper shrinkage/expansion. The inkjets are better because they don't have the shrinkage/expansion issues of the laser. They might also provide blacker and more consistent blacks, I don't know as I don't have one.
Anyway, it was more a matter of curiosity on my part, but it also points to the outcome, that if transfer papers were more thermally stable, the results of the laser would be better, or use an inkjet which doesn't have the issue.

[SilverSolder]

If you are doing transparencies, the inkjet is the way to go (around here, anyway) - it seems capable of 0.15mm or even better accuracy.

I think you missed the point of what I was doing. I was trying to differentiate between the accuracy of the laser printer vs the paper/transparency. I am not doing transparencies, they were chosen as a stable medium (PET) that allows easy overlay and comparison and would test the laser printer accuracy with minimal distortion due to the media.
The result is that the laser printer on the test page provided, has a maximum error of around 0.25mm over the whole page. The 0.8mm errors you reported, I am pretty sure, unless I have an exceptional machine, are due to paper shrinkage/expansion. The inkjets are better because they don't have the shrinkage/expansion issues of the laser. They might also provide blacker and more consistent blacks, I don't know as I don't have one.
Anyway, it was more a matter of curiosity on my part, but it also points to the outcome, that if transfer papers were more thermally stable, the results of the laser would be better, or use an inkjet which doesn't have the issue.

Ah, I see.  -  I use transparencies sometimes so it didn't seem unnatural to me! 

Inkjet, with the correct type of ink and the correct type of transparency, is unbeatable for photo etching.   First you need the special Silkscreen Inkjet Transparency material as used by the T-shirt industry:  this film has a light milky white appearance, and the microporous waterproof technology coated paper can absorb a ton of ink (giving great contrast) and dries immediately. Second, you need dye ink (not pigment), it gets absorbed into the film and gives a darker result.

This is what I do when the boards get big, and precision matters.

For everything else, toner transfer is "good enough for Australia"! 

[moffy]

Thanks for the info SilverSolder, I don't know much about inkjets as I have tended to steer away from them because of ink costs. It sounds like you have excellent accuracy, you might want to take a look at this direct to pcb film printer which I posted previously:  https://hackaday.io/project/26216-pcb-exposer
Probably $100 worth of parts for exceptional accuracy.

[SilverSolder]

Thanks for the info SilverSolder, I don't know much about inkjets as I have tended to steer away from them because of ink costs. It sounds like you have excellent accuracy, you might want to take a look at this direct to pcb film printer which I posted previously:  https://hackaday.io/project/26216-pcb-exposer
Probably $100 worth of parts for exceptional accuracy.

I did look at that when you posted it earlier, definitely a cool idea.

I have both laser and inkjet printers, each is good at different jobs...  I don't have a color laser yet... 

The cost of ink is not too bad, if you refill the cartridges.