[zoltan]

Ben,

I was thinking about this also myself but I was afraid about reflexion when the coating is burn. Have you tried with something else than UV sensitive coat? I mean some sort of pigment or paint? For UV sensitive pre-coated boards I am using the UV box and the print on transparency, but coated board with regular paint and then "engraved" by laser sounds "intrestintg".

Thank you.

Zoltan

[plexer]

My versalaser is an enclosed unit and so far only tried it with the coated boards the quality of the finish was fantastic, the final etched boards were not so good but was just my etching skills.

The laser only burn/cut when they are focused so the reflected laser beam isn't likely to do much harm.

[zoltan]

Thank you.

Sorry, my mistake. I was referring to the damaging of laser due to reflected beam back by the surface of board after the coat is burn.

Zoltan

[C88mod]

Ok, so i refilled with some cheap ink and some good durabrite yellow.
I was not getting repeatable results so i moved the paper sensor
in further and using a notch instead of a hole for sense on the feedcard and got rid of the
alignment problems between printing multiple passes of different layers.

pics:
1. before moving sensor inward and using hole for paper position sense.
2. component, copper and edge layers line up perfectly on all three passes through printer.
3. added the solder mask layer. (no flash)
4. same

Hopefully, i will have a 2 sided board made up tonight.

[C88mod]

hmm, printer still has alignment problems, epineh you still out there
and have you made a 2 sided board yet?

[C88mod]

i am thinking of destroying an old hp 610cl and running ER-71 photo resist through it...

[epineh]

Yup, I'm still here, I haven't done much lately, been a little busy, I have run out of blank PCB, I don't have a 2 layer board design yet. My plan is to get a single sided board design to get my router working again, then work on the 2 sided version. I tried printing the UHU servo controller board, but I think I altered the magenta levels too much and ended up with too much ink on the board, also warmed the FECL too much and the ink seemed to wash away, I will try again as soon as I can get some more board.

I am not too worried about the bad result, I can print directly from eagle and etch with good results, just have to work out the correct colour/ink for converting a black PDF layout to magenta/red.

I will hopefully try again in the next couple of days.

Russell.

[Mycroft2152]

I'm sure this is old hat to most of you who have disassembled an Epson printer, but I was amazed at the amount of absorbant wadding tucked into the bottom of the printer when I opened it up to clear the drain tube. The C66 was a well used and non-working one from a school. The amount of dried ink build up was astounding.

Pigment inks may have some advantages, but the compromises needed by engineering are frightening. The whole design just screams "disposible printer"!

We'll just have to see if the new Kodak pigment printers (due out this weekend) are any better.

[wnnelson]

I don't see how you got so many repeatable results for alignment. I might get 2 to line up then the next is off. The solder mask is hit or miss, very frustrating. I have a stop for the carrier so I know it starts exacly in the same place every time. There is a sensor under the print head which I can't seem to figure out what it does. It was suggested that it sensed the edge of the paper but it does not go anywhere near the edge when the paper if feeding.

[epineh]

I have been printing a draft copy directly onto the carrier to check its position, then setting up the board to suit, mind you, I have only been doing small boards, opto banks etc. I did have one that seemed to print way off early on, I cannot explain why it decided to do that, but it is printing consistantly every time now.

The only thing that "should" affect it is the alignment left and right and the little optic sensor, maybe the sensor isn't working properly with your carrier, reflecting or something.

Is the alignment out left/right or up/down?

I am doing some more runs if all goes well today, hopefully refine the process a little more.

Are the feed rollers slipping when the carrier goes through? Even just a little.

[dleroi]

First, thanks to all of you that have posted your work. I came across this thread, tonight, and was riveted to it. I've been making PCBs, using various methods, for 30 years and this is the most revolutionary and promising idea that I've come across. I have 2 inkjet printers that are perfect candidates, but I'm also eyeing a little dye sublimation printer that's sitting across the room from me. Has anyone any knowledge or thoughts on how dye might hold up to etchant?

[C88mod]

I don't see how you got so many repeatable results for alignment. I might get 2 to line up then the next is off. The solder mask is hit or miss, very frustrating. I have a stop for the carrier so I know it starts exacly in the same place every time. There is a sensor under the print head which I can't seem to figure out what it does. It was suggested that it sensed the edge of the paper but it does not go anywhere near the edge when the paper if feeding.

Yeah same here, works fine 2 or so times on test card, then it seems to be off exactly
4 - 8 steps on both x and y at random times.
I am drilling and pinning with thumb tacking the work on the feeder and it still goes off.
and yes the feedcard is tight against the rails and the feedshaft is not slipping.
You might have touched on the fact that sensor
underneeth the printhead (the led and photodiode) are likely causes,
or it could be that the printer cannot find home between x number of prints.

Try using black paper/tape on your feeder card and pin your board to the feedcard, I will do the same
and post here later tonight...

[C88mod]

First, thanks to all of you that have posted your work. I came across this thread, tonight, and was riveted to it. I've been making PCBs, using various methods, for 30 years and this is the most revolutionary and promising idea that I've come across. I have 2 inkjet printers that are perfect candidates, but I'm also eyeing a little dye sublimation printer that's sitting across the room from me. Has anyone any knowledge or thoughts on how dye might hold up to etchant?

Hi, and Welcome to the group.
Hmmm, dye might not work so well as there is no pigment to resist the fecl3 etchant,
but i could be wrong, maybe if you took some of that dye sub stuff
and dripped/smeared it on a test pcb and cooked and etched the board to see?

[wnnelson]

Hmmmm, slipping carrier on the drive shaft. Now that is something that I did not see as a problem as I have all the pressure roller still there but they are higher for the PCB. So this could be part of the problem. Not to sure about reflection as it should have white paper between the sensors to block the diode so I don't think that would be it, but I will try somehting else on that edge. More testing, testing testing.
I also do not think the dye inks will work. There are resin binders in the pigments which I think are giving us the real resist when they are fused to the board on heating. You can run pigment through the dye heads as some have reported to do on a Epson R220. I have a R200 which I may try to mess up too.

[epineh]

Well I have to be honest and admit I haven't been doing a solder mask, my need for boards at the moment is for experimental purposes, I don't expect my boards to live a long life just yet, but I thought I better try a mask (peer pressure!!!)

I did a run of small opto boards (4 of) and I thought I would try a mask on one so I could at least comment on the process.

The one I did was ok, but as you can see is a little off to one side, I blame the heat of the board for me not pushing it into the corner properly, thats my excuse and I am sticking to it!!! I am also thinking of a more solid edge on the card carrier would help repeatability of locating the board, maybe 1.6mm thick strip of ally or something.

Also I found that cleaning the board with metho first, then wiping with an acetone soaked rag made a big difference to the ink sticking properly. I might buy the little wipes we use for cleaning fibre optic cable before splicing for a test.

Top left is the board as it came out of the etch, top right is the "masked" version, bottom left is just cleaned (with acetone again) after etching, bottom right is a test version that was not cleaned with acetone before printing, definately a difference between that one and the other three.

The mask is a little off, but I will improve, also have to work out the settings in Eagle to print a proper mask, I just edited the view and left the tracks only, changed the colour to green, heated the board again then printed.

Russell.

[C88mod]

I also do not think the dye inks will work. There are resin binders in the pigments which I think are giving us the real resist when they are fused to the board on heating. You can run pigment through the dye heads as some have reported to do on a Epson R220. I have a R200 which I may try to mess up too.

That would be cool if we could add this resin to ordinary cheap ink as to etch with any color we please, wonder what exactly it is and if it is attainable on the cheap?

[kreutz]

Hello;

I am following this thread with a lot of expectations. You will be the pioneers of the new era in PCB prototyping. Since you have been thinking about this solution for a long time, I have a question for you.

I own an Epson 7600, 24 inches wide, straight paper path feed from the top back in an angle, adjustable paper thickness, I use durabrite inks (big VHS size cartridges).The print-head automatically finds the paper edge at startup, I think it could help with positioning accuracy.

Could I use this printer without modification for PCB prototyping? I already have a temperature controlled (micro-controller based with programmable profile) infrared toaster oven I made, and use, for reflow soldering.

[wnnelson]

Kreutz, when you say "straight paper path feed from the top back in an angle" does this mean the feed tray is tilted down into the printer? I don't see that working. Unless you can get flexible PCB stock I would say flat all the way back to front. Do you want to modify it at all? OEM Durabrite inks are the ones we are mostly using, some are using compatible inks with great results. The oven would be the best for curing the ink and possibly preheating the board. I just looked up the printer on Epson but I still can not see the paper path well. A $3000 dollar printer seems a little high to hack so I can see where you want it to work as is.

[C88mod]

It might be because the carrier/tray is cardboard.
I am not worried though, when i can print all my layers side by side on a single sided sheet which will eliminate any alignment problems there, at the cost of sandwiching
the .016" boards together with silicone sealer, then drilling.
I don't know if this printer is capable of putting the ink dead on the same spot on multiple passes.
pics:
first pass magenta (looks like **** with cross contaminated cheap magenta ink.)
second pass cyan (forgot to snap :P
third pass yellow
fourth pass black
closeup of middle of print
the yellow was off a whole bunch on the y axis, but the carrier got loaded the same way
on all 4 passes.

[wnnelson]

What software are you printing from? I use Eagle and I just noticed that if I change anything on the layout it may not print the same even though the dimension outline stays the same. I cut out a hole in thin cardboard to make sure the board goes back in exactly the same spot but when I print the solder mask it is off. Sometimes it is off by a lot .250" that is really wierd. When I print something in the regular printer it is never that far off. I'm getting cranky over this.
Epineh, I see you are covering only your tracks with the solder mask, I am covering the whole board like a commercil board. Maybe I'll try tracks only too, this will keep the print the same and may eliminate the alignment problems.