Electro Engraving Problems

[vtsteam]

I've been trying to electro chemically engrave a small metal plate or my new lathe. Not much luck yet. I first tried doing a resist transfer using a laser printer image on paper.



To make the transfer sheet, following several how-tos posted on the net. For a transfer medium I tried printing on three different suggested papers I could find around the house: glossy magazine stock, the back of a sheet of labels, and some coated printer paper. The test piece was aluminum, which I sanded with fine sandpaper, then hit with Scotch Brite, and cleaned with alcohol.

None of the above papers transferred completely. The glossy magazine stock may have worked the best. For that trial  whitespace was removed by soaking the paper in water, and rubbing it off.

The other two were supposed to transfer just by peeling off without water. The label backing did not print well through the laser -- it was a little pale and blurry, and left an after image further down the page I didn't even try with that one.

The coated paper did a little better and it did transfer somewhat, but not well enough.

I don't know if the heat was high enough, or the pressure wasn't sufficient, but I was using a model airplane covering iron set to the highest setting, and spent a lot of time pressing very hard -- it just didn't print well.

I'm suspecting that maybe the laser toner itself is at fault. It's an aftermarket cartridge, so that may explain the difference I'm getting compared to others I've seen in videos. The normal printout on paper is somewhat light as well, so that may well be the culprit.

[vtsteam]

So then following a second suggested method in a video, I pulled out of storage a small cheap 2.5 watt CNC laser engraver I'd impulse purchased a couple years ago for model airplane work. The YouTube suggested method was to paint the alumunum test piece with black spray lacquer, and then burn the design through with the laser. Afterwards you etch and then remove the paint with lacquer thinner.

Well, because the size of this plate was only 4.25 inches wide, the lettering is quite small. I had saved the working png file at 300 dpi to preserve detail. After doing a couple test burns of a few letters on the prepared stock, I found that I needed full 100% power, but shortest burn time for best results. So, settings known, I started a full burn of the image at 7:00 PM, and figured it might take over an hour. I was using the software bundled with the engraver, directly with the png file.

Nope. It took 5 hours -- I was just about to abort the burn so I could get some sleep when it finished at midnight. 

[vtsteam]

The burnt image wasn't perfect but legible, and this was just to be a test piece anyway. Next step, etch. There are quite a few methods and recipes on the net for etching, but the one that appealed to me as probably being as effective as any, and using available non-toxic household chemistry was one that mixed distilled vinegar with table salt and the used electrolysis to effect the etch.

So I hunted up the ingredients and found a vintage variable power supply (a 1963 kit nicad charger) and a tall glass vase and some wires and tape. Taping positive to the workpiece, and negative to a scrap of aluminum, I was in business. It bubbled, and  looked suitably mad scientist enough to raise a couple of eyebrows in the household before dinner...... well maybe a couple of jaded shrugs. Luckily I owned the vase.

[vtsteam]

Most of the online accounts of this type seemed to get a deep etch at about the half hour mark. I had periodically switched off the power supply to pull out the workpiece and check progress, also to help agitate and clear away bubbles. But after 30 minutes it didn't look like much had been removed, if any. The cathode plate had been bubbling vigorously, but the workpiece's etched lines were generating gas at a more sedate rate.

I decided to go longer. I had the charger cranked up to its highest setting. Though without a meter, I didn't really know how much current I was pushing. Or the voltage. I just assumed everything was working because of the bubbles.

After an hour, I thought I could see definite etching, so I decided to call it done before some of the delicate spaces inside the small letters got undercut.

The result after cleaning off with lacquer thinner? Barely etched -- some areas not legible. Afterwards I did check the voltage after reconnecting everything, and it was 1.25 volts, far below what most recommend (ranging from 3.5 to 12 V). I probably should have used my 12V car battery charger. Oh well. 

So, determined not to spend another five hours babysitting a CNC router, I decided that raster (pixel) graphics were probably not the best way to go, and I'd look into converting my image into vector graphics and ultimately a g-code file.

Well 5 (coincidentally) hours of research later I had converted the image to an SVG file via Inkscape, then downloaded and installed a gcode generating extension for the same program, then found a detailed set of instructions how to make the file conversion suitable for a laser cutter (modifying the plasma cutter tool setup) and finally finished the job! Not. The generated G-code would never have worked, I could see quite plainly in the tool paths. The problem was the tiny size of the letters. I'm sure it would have worked on bigger graphic images.

So finally, I'm back to square one. I think the best way to get this etched for this particular image is likely to be through the laser printer paper transfer method, not the cnc laser engraver route.

I will need to get a denser print with maybe Dell (or Xerox) toner on proper transfer paper and that means sending away for stuff. Unless of course anyone else has a workable alternative suggestion...

[AdeV]

Hm, you've got me thinking now...

I bought a cheap Samsung M2026W laser printer for my experiments with PCB etching. Like you, I went through all sorts of pain trying to find a decent printable medium which would transfer well onto the copper board. I also struggled to transfer it well using an old clothes iron. As mentioned elsewhere, I ended up using a cheap laminator, which worked great for a bit, but then the overload clutch started spinning; so I had to remove that (or, rather, jam it up so it'd end up stripping teeth rather than clutching... it's worked OK so far! The printer I bought from Argos (a sort of sells-everything catalogue shop in the UK), the laminator came from Amazon.

The best stuff I've tried so far is laser printable vinyl, printing actually onto the vinyl surface itself. Pass this through the lamintor several times (to build up sufficient heat in the piece), and so long as I didn't get too impatient, I was able to get 100% toner transfer.

AS for the etching itself.... I was using a mixture of hydrogen peroxide and hydrochloric acid, both of which I appear to have lying around the workshop  for my PCBs, and had some success, although far more error than trial.... for aluminium I'd use sodium hydroxide - drain cleaner - which I also appear to have "a quantity" of around here... Actually, I may even have some distilled vinegar come to think of it.... don't ask.

Anyway.... I'm going to have a go tomorrow I think, just to satisfy my own curiosity... and if it works, I'll put some photos up for you.

If we didn't live on different continents, I'd offer to print your design onto vinyl and post it to you (actually - I will offer to do that if you like, but I've no idea how long the post will take!).

[RussellT]

My experience with circuit boards is that the best way to get toner to transfer is to buy the special paper (Press n Peel is one brand) - I know that's not the Mad Modder way - I've tried various stuff suggested on the web with little success.  I haven't tried Ade's vinyl method.  Even with the special paper I try to keep tracks as wide as I can and I think your letters may be a bit thin.

I've never tried etching aluminium but it sounds as though your laser method might work with copper or brass.

Good luck.

Russell

[vtsteam]

Thanks for the suggestions guys! Yes I think purchasing some vinyl or other special purpose transfer media is probably what I'll need to do to make that work.

I don't think I'll buy a laminator, because I don't see myself needing to do this kind of thing much, and I have so little storage and an extremely small shop. I used to have good luck with printed circuit boards just using the photoresist method and a normal printout on paper, face down and exposed through. Not that I often do any printed circuit work at all these days.

I think one of the biggest problems in these methods for me is the needed fineness of the lettering.

Just listening to the laminator successes makes me think that the real problem is insufficient pressure. Maybe it's insufficient heat, but I kind of doubt it because the laser printer instruction manual mentions that media must be able to withstand 400 degrees (F) so that seems to be the temperature used by the laser to fuse the pigment. It seems to me that the iron is getting up to that..... well I don't know. Maybe it isn't. I should check.

I'm sure that all the etchants and electrolytes mentioned will work, given proper time temperture and agitation. I chose the vinegar and salt because it seemed to work on all metals shown in the following video -- steel copper brass, aluminum and stainless steel. The actual stylus etching method shown didn't make a lot of sense to me (vs bath immersion) except for the stainless cup -- or where large size makes it impractical to immerse. But I think it's a worthwhile video.

[awemawson]

You used to be able to buy spray on photo resist, and indeed pre-coated pcb's with UV sensitive resist already on then - I have some in a drawer but if they are any good now I very much doubt,

I have a complete set up for the job - UV light box, and a 'PCB Lab' - that develops, etches and tin plates PCB, but I've not used it for donkey's years - it even incorporates the kitchen sink 

If anyone is dead keen to set themselves up get in touch as it's just taking up space.

This is it, made by Mega :

[vtsteam]

lots of stuff for photo resist still available:

https://www.ebay.com/b/photoresist/bn_7024915100

And a video showing DIY methods of producing circuit boards that way:

[AdeV]

I figured I'd have a really quick go at etching an offcut of ali plate I had lying around this morning.... so, quick print job, print to vinyl, abuse the laminator to transfer; it worked ish, I got about 98% of the toner across - the ali wasn't perfect, having a few dents & vice marks in it, which didn't help; then set it up in a mystery strength solution of sodium hydroxide. All was going well for a bit, but then the etching undercut the thin letters & they all floated off! Also, some of the etchant got underneath the thicker lines in a couple of places, so the result was extremely meh....

So... instead of doing a productive day's work, I scratched my head & thought about the problem.... and I think I may have a solution. Unfortunately, for the majority of people, it's not going to work because you need a couple of specific pieces of workshop equipment which few will have....  I'm thinking, I'll use my CNC engraver to carve the message (or whatever) into a plate. I can then use my falling die spark eroder to "engrave" the opposite onto a sheet of metal - i.e. leave raised letters. If the engraved text is deep enough, I shouldn't lose any definition over the, say, 20 thou/0.5mm depth I'd anticipate engraving.

I reckon that's got to be worth wasting a weekend trying! 

[vtsteam]

Ade it's a disease!  I've totally stopped productive work on my ELS and lathe to concentrate on this stupid etching thing to make a metal version of simple label I can print and attach in a few seconds, if I were satisfied with paper. This is just bling. But ya know, I just hate it when I find I can't do something, that I know otherwise CAN be done, no matter how trivial the ultimate purpose is.
 

ps, and now I've gone on Ebay and ordered a sheet of the the film type PCB photoresist, some overhead projector film (laser printer compatible), 5 sheets of Press-N-Peel transfer paper, and a new black laser printer toner cartridge (but again not OEM -- they are $100). Fingers crossed that at least one of these things will make a reasonable difference.

ps, Ade I think lye is a first rate paint remover, so even though it does a good etching job on aluminum, I'm guessing it's hell on resist -- as happened in your trial. Even though aluminum is much more resistant to acids than a base, the electolysis method may speed the acid's effect up. That's just a theory -- possibly contradicted by reality. But the reason why I'm trying vinegar/salt and volts.

[sorveltaja]

Does the plate have to be of aluminum? If sodium hydroxide was used, it has some kind of thermal(or something like that) reaction with alu, that causes uneven etching and pitting. 

In general, aluminum is a bugger to etch any delicate figures, be it chemically or electrochemically. But that's just my experience; there could well be ways to get better results.

     

[AdeV]

Does the plate have to be of aluminum?

I think we were trying aluminium because it's cheaper than brass! I'm fairly sure the final plate will be brass. I only have bits of aluminium (or steel) sheet lying around; or copper-clad circuit board - but I know I can etch that stuff using laser toner as resist, but that doesn't make for a nice plate to go on a machine...

ps, Ade I think lye is a first rate paint remover, so even though it does a good etching job on aluminum, I'm guessing it's hell on resist -- as happened in your trial. Even though aluminum is much more resistant to acids than a base, the electolysis method may speed the acid's effect up. That's just a theory -- possibly contradicted by reality. But the reason why I'm trying vinegar/salt and volts.

I think the resist was OK - the letters just floated off & were sitting on top of the lye (on the bubbles) - complete, but detached... However, my sum total prep on the aluminium was to give it a scrub with some dry scotchbrite, I didn't try to degrease it or surface prep it any better than that; so it's possible the toner wasn't stuck down as hard as it should have been. Also, I did have a fair bit of trouble getting the plate through the laminator, it's a bit thicker than PCB (2mm vs. 1.6mm) and I suspect that was a bit too much for the poor thing. It only cost me £10 (about $13), so it's no big loss if I do kill it.

I took some photos of the process, including the epic fail, I'll put them on a different thread though, I've polluted this one enough with my nonsense! 

[Pete.]

Ade I have a huge sheet of reclaimed brass if you need some. I think it's 18 or 16 gauge. IT's got cotact adhesive one side and (unbelievably) brown paint the other but it comes up well with some careful scraping of the glue and rubbing with scotchbrite.

[vtsteam]

For me the final hoped for plate is brass, and I have a small piece large enough for only two tries. So that's why I'm starting out with aluminum. I have yet to get a qood resist copy except by the CNC/laser paint removal method -- and that one wasn't ideal. So until I get the resist part of the equation done, there's no point in going to brass yet.

As for the etching of aluminum with good detail, if you watch the first video I linked above, you will see a very clean edged aluminum etch with the vinegar and salt electrolysis method.

As for sodium hydroxide and resist, in the second video I linked above, sodium hydroxide is used to remove the film based photoresist after the etch is over. It does actually float up the resist, it doesn't dissolve it.

When I was 7 my mother got a 55 gallon barrel, set it up on bricks and put a load of wooden house shutters in it. Then she put water and lye in ( I was very impressed with the label's skull and crossbones) and lit a wood fire under it. This removed all of the century-old enamel paint layers from the shutters in a very short time. This was in the fifties before modern paint strippers were common where we lived.

[vtsteam]

Yesterday I tried one more method of printer resist transfer -- heating the plate under rather than ironing through on the top. The results were better than before, but not nearly good enough. This is again with the materials at hand. The new materials are due here by the first of Feb or so.

My suspicion is that the film photoresist method has the best chance of solving the problems encountered so far in doing this fine-lined job. I believe that the other methods tried work well for larger area images -- artwork and signs and such. But this particularly demanding thread chart application isn't well suited for those.

[AdeV]

Ade I have a huge sheet of reclaimed brass if you need some. I think it's 18 or 16 gauge. IT's got cotact adhesive one side and (unbelievably) brown paint the other but it comes up well with some careful scraping of the glue and rubbing with scotchbrite.

Hi Pete,

I would definitely be interested in a bit of that at some point.... I've got a few brass plates on the lathe that need re-making, due to wear and tear, I'll ping you some sizes over in a day or so, they're only little...

[mattinker]

Adev,

If your making MkI Edgewick plates, could you make two of the one on the top of the saddle, the thread indicator one? I will of course pay!!

Cheers, Matthew

[PekkaNF]

Apparently white vinegar works for aluminium:

I too have ben working with some laser print transfer materials. I used to use standard HP printer with standard HP color cadridge and no paper would work with it. However brand print trancparency film worked fine, but need a clothes iron to transfer image. Standard laminator just does not have enough heat, I read from several sources that you need 180C to fuse the dye.

I still have the same old lasr printer and bought a new color cadridge, not HP and not the cheapes one, but mid priced generic. Transparency film does not work any more, but fun fact is that super glossy and coated paper works almost fine with a the same laminator.

I got some chinese yellow transfer paper and it seems to work most of the time ok. Not great for surface mout but ok with trough hole system.

I am thinkking of getting a 100€ T-shirt picture transfer press for this purpose, they should provide enough heat and pressure to get work done....or try UV option again...or try gardening instead.

[vtsteam]

I received the blue Press-N-Peel sheets yesterday. These are supposedly designed for laser printer transfer to printed circuit boards. Pricey stuff, $3+ per sheet. I bought 5.

Well that was a mistake. The first sheet I tried disappeared into my Dell 1250c laser printer (Xerox Phaser 600b engine) and never returned.   

Removing all the user access covers for clearing paper jams failed to disclose its location. 

Shining a strong flashlight into the back cover grill showed a thin sliver of blue trapped in the (well guarded) fuser rollers. The entire sheet was compressed into a space a couple mm wide (see I can use metric).

After a fair amount of internet research uncovering some approximate anecdotal advice I managed to take off the retaining bezels defeating most of the devilishly clever hidden plastic retaining hooks catches and pivot points, only breaking one or two in the process. And finally loosening the entire top case of the printer for access.

The plastic grill (still attached) covering the fuser roller was bent back far enough to insert a hemostat and clamp onto one bit of the blue transfer sheet. I gradually worked it out through the top of the grill.

Released and expanded it looked like the side panel of an accordion. I have never seen a worse paper jam. Well it's not paper, actually, it's very thin plastic film, super shiny on one side. Who in their right mind thought that this kind of thing would pass through a laser printer intact? Including me? Why didn't I go......hmm, that'll never work, let's not take a chance? But nope, I just ran it through, thinking gee this stuff says it works..... I'll soon have a brass etched plate, now.....Duh! 

Okay, folks..... don't buy Press-N-Peel transfer sheets unless you enjoy "Pressing" hidden plastic fasteners in your laser printer's construction a"N"d "Peeling" crumpled plastic film out of its fuser rollers. 

Luckily my printer wasn't damaged (other than the broken catches) and I was able to print again on real paper after the fiasco. What a blivet. 

Caveat emptor.

[RussellT]

Hi Steve

I feel a bit guilty now as I suggested the Press n Peel.

I've had success with it and no problems like that.  However I don't think I've ever put a whole sheet in the printer at one go.  It's far too expensive for that.  My technique is as follows.  First I print out the design on plain paper so I know exactly where it appears on the sheet, then I cut a piece of Press n Peel slightly bigger than the design with a bit of excess at the top, I tape that along the top edge to the piece of paper I first printed so that it covers the design.  Then I feed that through the printer again and it prints the design on my scrap of press-n-peel.  I use ordinary sellotape but I generally cut the ends rather than tear them so that they lie flat.

Russell

[Pete.]

Ade I have a huge sheet of reclaimed brass if you need some. I think it's 18 or 16 gauge. IT's got cotact adhesive one side and (unbelievably) brown paint the other but it comes up well with some careful scraping of the glue and rubbing with scotchbrite.

Hi Pete,

I would definitely be interested in a bit of that at some point.... I've got a few brass plates on the lathe that need re-making, due to wear and tear, I'll ping you some sizes over in a day or so, they're only little...

No problem. Let me know when you have your sizes. They'll be cut over-size with a grinder though as I have no shear.

Pete.

[vtsteam]

Hi Steve

I feel a bit guilty now as I suggested the Press n Peel.

No worries at all, Russell, I had actually ordered it before I saw your suggestion. And your (above) suggested method makes a lot of sense. I will try it (again on my own decision, don't worry!). I think that will solve the slippery back face problem.

I have actually printed homemade model airplane "decals" on fine tissue paper by doing a similar paper backing method, and never had a problem.

Thanks for the explanation of what you've done.