From: Don_Schmitz@transarc.com Date: Mon Aug 3, 1998 Obligatory Disclaimer: The information contained in this message was contributed by individuals, who, unless otherwise indicated, speak only for themselves and not the institutions or buisnesses they are associated with. The author(s) and editor(s) of this material make no warranties as to the correctness of the information provided. This material should be considered copyright by the author. This material may be redistributed for non-commercial use without explicit permission of the author(s) as long as the text is used exactly as is (except for reformatting) and the author is given full written credit for the material. Commercial use requires explicit permission of the author.
Part 6
Part 6 ** How to make your own photo-etched parts. This is a single question with several long answer. [Q] How do I make my own photoetch parts? If you have web access, check out the following related urls: phoenix.bath.ac.uk/~justinw/two_mm/etch/etch.html home.cdh.net.pcboard [A] schmitz@transarc.com (Don Schmitz) Here is the summary of the "do-it-yourself" photoetching article that appeared in the Sept. 1993 issue of "Fine Scale Modeler". This is a nicely done article, written by Rusty White , who runs a small business producing photo-etched ship details. I've tried to extract the useful information in the article so as to minimize both my typing and the chance that I would violate someone's copyright. I want to point out that I haven't used this technique myself - I just happened to pull this magazine out to consider giving it a whirl for a project I'm starting, when the request for info showed up here. I've double checked all of the details, but please let us all know if you detect something that looks wrong. Finally, I've added editorial comments based on information bouncing around my head that I can't attribute sources to - other articles, former r.m.s postings, and a few of my own thoughts. Some of these are speculation on my part, and I want to make sure that they are recognized as such, and not taken with the same authority as the information contained in the FSM article. I've identified these comments with square brackets ([]s) in running text, and >>> where I ramble on at length. ------------------------------------------------------------------------ The article outlines the procedure for producing photo-etched parts from standard hobby store .005 inch (0.13 mm) thick sheet brass. The article states emphatically that the etchant described (ferric chloride) will destroy aluminum sheet, and seems to imply that it will have no effect on stainless steel. >>> This process should also work with similar sized nickel-silver >>> sheet if you can find it. You can probably push the thickness of >>> the sheet up to .02 inches (0.5 mm), but you will tend to get >>> less crisp edges - as usual, some experimenting is in order. 1. Designing the parts. The articles suggests that you first draw a draft of the parts you plan to produce in pencil on drafting paper, working in a reasonably large scale - typically 3 to 5 times actual size. You should design the pattern much like a sprue of plastic parts, with an outside frame of metal supporting the desired parts via small "runners". The runners are needed to keep small parts from falling into the depths of the etchant tank [unless have a source for the adhesive-rubber backed brass sheets that are used for many commercial photoetched parts]. When designing the sheet of parts, you should also try to minimize the amount of brass that must be removed from the sheet (in order to extend the etchant life and minimize etching time). This implies filling in large open areas with "dummy" parts, or drilling holes in the open areas. >>> Speaking as a one-time mechanical designer/draftsman, you have to be >>> careful when working at 5x actual size to not design parts that are >>> ridiculously small in actual size. For example, say you plan to >>> produce a car grill, and draw perfectly reasonable looking 1/16-inch >>> wide grill slats at 5x actual size - corresponding (assuming >>> a 1/24th scale model) to a .3 inch thick slat on the 1:1 original. >>> This corresponds to a .01 inch (.25 mm) wide feature on the 1:1 >>> photo-etch sheet - about the width of a thin pencil line. My guess >>> is that such a feature will simply disappear when etched. >>> >>> The article doesn't mention it, but I expect that the resulting >>> etched parts will be slightly smaller than the artwork, by about the >>> thickness of the brass sheet being etched. If this source of error >>> matters, you will have to experiment to find out how much width you >>> lose to etching and exaggerate small features to compensate. Once you have the design for the entire photo-etch sheet mapped out, you should transfer the pattern to frosted mylar drafting paper (available at art/drafting stores) using a #0 black technical pen. Color in the parts (corresponding to the brass not to be etched) using a #2 technical pen. You then take the mylar artwork to a shop that can make a "film-positive" from the mylar original. Supposedly most shops that make blueprints and reproductions of architectural drawings can do this [another article claimed a cost of roughly $7.00 US for one such reproduction]. When you have the positive made, you should specify the amount of reduction you need so that the positive is the actual size you want the etched parts to be. The article suggests you draw a "scale" on your original that should map to 1 of your favorite units of length when the reduction is done - eg. if you're working at 5x actual size, draw a 5 inch (or cm) line along one edge of your artwork. You can then double check the reduction by measuring the line on the film positive - it should be 1 inch (or cm). >>> A frequently asked question is whether you can prepare the artwork >>> yourself, using a reducing photocopier to get the size right, and >>> then photocopying onto a transparency to produce a "film positive". >>> Or, alternatively, using a CAD package to generate the original >>> and printing directly on a transparency. At least one poster to >>> r.m.s has reported success with this basic approach. A few possible >>> complications: >>> >>> - many photocopiers and printers introduce (slight) distortion in >>> their output ie. straight lines on the original are slightly curved >>> on the copy. This is probably not significant for small parts that >>> don't have tight fit requirements. The distortion tends to be >>> worse near the edges of the page, so you may be able to work around >>> this by centering your artwork in the middle of the page. >>> >>> - most photocopiers and laser printers have a hard time producing >>> large black areas, especially on transparency plastic - the centers >>> tend fade to white and there are often small white marks scattered >>> about. You may have to do some touch up with black paint. 2. Transferring the design to brass The next step is to turn the artwork on the film positive into a resist pattern on the brass. The first step is to thoroughly clean the brass sheet as if you were going to paint it - wash with detergent, rinse and air dry. Spray both sides with a few coats of photoresist and allow to dry for 24 hours. The source listed in the article was from a company in the UK, it is: Electrolube RP50 Positive Photoresist Maplin Electronics PO Box 3 Rayleigh Essex SS6 8LR England Approx. cost $25 for a large spray can, including shipping and international money order. An alternative, US source was given as: G.C. Electronics 1801 Morgan St. Rockford, IL 61102 (free catalog) Note: This photoresist does not change colors when exposed and developed as described in the following text. Next you expose the photoresist. Take a sheet of 1/4 inch plywood or similar stiff material, add a thin layer of foam rubber, follow with the resist coated brass, then the film positive, top with a sheet of glass, and rubber band the whole thing together. You need a UV source to expose the photoresist, the FSM article suggests direct noon day sun for 2 minutes, add another 2 minutes for cloudy days. Other articles suggest plant grow lamps or sun lamps. Once exposed, you should remove the plate from any light source until it can be developed. With the RP50 photoresist, a correct exposure results in a yellow tint. Next, you develop the brass - ie. remove the exposed parts of the photoresist. Developing is done with a lye (sodium hydroxide) solution - a chemical supply house is suggested as a source for the lye. Create a "stock" solution using 20 grams of lye to 250 milliliters of water, and store this in a bottle clearly marked "poison". For the actual developing work, dilute 1 part of the "stock" solution with 7 parts water. Put the exposed plate in a plastic or stainless tray, and cover it with the developing solution for a few minutes. Rock the container to agitate the solution. Remove the plate (wear gloves and use tongs/tweezers) and check the development process - the previously exposed areas of the resist should wash away, leaving the brass that is to be etched away exposed. If you use the RP50 the remaining photoresist should turn a green color. When all of the exposed resist is gone, rinse under tap water to halt the development process. If there are minor flaws in the resist image, you can touch it up with a technical pen - the ink is an effective resist. If you overdevelop the plate, the photoresist image will be fuzzy and unusable. Put the plate back in the developer solution until all of the resist is gone, and start over. 3. Etching the brass. The etching solution is ferric chloride. The article suggests you buy dry ferric chloride from a chemical supply store, and create a solution of 250 grams of ferric chloride to 600 milliliters of water. The article warns that ferric chloride will stain just about everything, and is a nasty corrosive chemical - so wear gloves and treat the stuff with respect. [I think Radio Shack sells ready mixed ferric chloride solution in 32 oz bottles for about $3, probably much more convenient to buy and use]. You need a plastic container that will hold the sheet of brass at a roughly 45 degree angle - a margarine tub is suggested. Place the brass sheet in face down (so that the etched brass falls away from the surface of the sheet), and pour in the ferric chloride solution. It will bubble and fume, so this isn't something to do on the kitchen table. Remove the plate and check the progress every 15 minutes or so, until the brass is fully etched. Typical etch times are given as 3 hours with new etchant, longer with used. Rinse with lots of water to stop the etching process. The etchant can be saved and reused, but it will gradually lose its effectiveness as more brass is dissolved in it. Again, you should label the etchant as poison and treat it as such. >>> The Auto-Etch photo-etch system you may have seen advertized in >>> recent model magazines uses the same etchant, and includes a heater >>> and circulating pump for the etchant to speed up the etching process. >>> If you're reasonably handy, you can probably duplicate these features >>> with a fish-tank heater or hot-water bath for the etchant tank, and a >>> fish-tank air pump bubbling the etchant to keep it circulating. 4. Finishing up. You now remove the photoresist from the parts by placing the brass sheet in some of the "stock" (full strength) lye solution. Rinse with lots of water, and your photo-etched parts are done! >>> Something to think about: unless you need to produce many identical >>> parts, you could skip the entire photo transfer process and draw/paint >>> the parts with a suitable resist right on the brass, perhaps using >>> a stencil or masking tape to get clean lines. I think lacquer paint >>> is a fairly good resist, and is readily available in spray cans. >>> The paint/resist can then be removed from the finished parts with >>> your favorite stripper - eg. oven cleaner or brake fluid. [A] RANDY.GORDON-GILMORE@ocr.com (Randy Gordon-Gilmore) Date: Fri, 15 Jul 1994 10:21:41 GMT I have been doing my etching for about a year now, and did not see the FSM article referenced. There are a few points which I'd like to add (pardon if they were covered in the article): -ALWAYS use distilled (deionized - DI for short) water for rinsing brass and making chemical solutions. Some water (such as in my location) contains enough minerals to mess up the processes. -ALWAYS use rubber gloves when handling the brass up to the point when the photoresist is dry. One fingerprint will ruin the photoresist adhesion, and unless you are MUCH more coordinated than I am, you WILL touch the surface of the brass, no matter how careful you are. -DO NOT use a cleaner that contains chlorine or chlorides to clean the brass. It will slowly react with the brass and make bubbles/porosity in the photoresist coating. After experimenting with pumice powder (which abrades will but is very hard to rinse all the particles completely from the brass surface), I ended up using 3M brand "Scotchbrite" abrasive pads sold as a non-metallic replacement for steel wool for fine wood finishing. I got the dark maroon grade at a hardware store, and cut it up in 1 x 2 inch (about 3 x 6 cm) pieces, and use one piece per etching session. I also make up a very weak ferric chloride solution (maybe one part of the etching solution to 100 parts of DI water) to chemically clean the part after mechanical cleaning. My cleaning regimen is thus: =1= Put on the rubber gloves. =2= Scrub both sides of the brass under running water using the abrasive pad (tap water is OK for this step). =3= Squirt off the piece thoroughly with DI water from a spray bottle. =4= Using plastic tweezers, dip the brass sheet in the weak acid solution for 10 seconds or so. =5= Thoroughly squirt off the piece again with DI water. =6= Shake off most of the water, rub off most of the rest with a paper towel (don't try to dry completely with the towel; it will leave similar to water marks on the surface), and dry completely with a hair dryer on low setting. If you leave too much of the water on the surface (even DI) before blow drying, it will leave water marks on the brass. =7= Coat with photoresist as soon as possible (within 10 or 15 minutes for best adhesion). Actually, after trying for a few weeks with the spray-on liquid resist, I was unable to get an even coating across the whole sheet, no matter how flat the sheet seemed. GC Electronics (who sell photoresist, along with the ferric chloride solution) suggested spraying both sides of the sheet, hanging vertically until dry, then spraying the sheet again and hanging vertically again, but "upside down" from the first step. The theory is that the coating will form a thin wedge, with the thickest coating at the bottom of the sheet. When you spray again and invert the sheet, the second wedge compensates for the first one. I never got it to work to my satisfaction (one of my projects is N-scale [1:160] railroad passenger cars with a ribbed outer surface, with .006" (.15mm) ribs separated by like spaces), and ended up using dry film photoresist made by Morton Dynachem. This material is .001" (.025mm) thick, and covered on both sides by a clear cover sheet. You take off one cover sheet and laminate the resist onto the brass with heat and pressure. I bought a surplus Canon fuser unit (like in a photocopier) that works perfectly for this function. It actually took a bit of experimentation to get bubble-free, wrinkle-free lamination, but I can coat up to .025" (.60mm) sheet this way. I also progressed through the static (no circulation) etching bath, through a bath with an aquarium heater and recirculating pump, to a homemade spray etching tank, made from acrylic (Plexiglas, Perspex) sheet with a small laboratory pump and two plastic spray nozzles that mist both sides of the vertically-suspended brass sheet, and the aquarium heater. I had known that commercial photoetchers use spray-type units, and decided to try it out myself, but was not prepared for the major leap in etching speed. Apparantly, even with agitation in a submerged bath, a thin layer of acid-with-etch-byproducts clings to the surface of the brass where the etching is taking place, whereas in a spray unit, there is a continual supply of fresh solution, together with a continual washing away of the byproducts (which are then evenly dispersed through the etchant). I had been diluting the GC ferric chloride by 1/3 with DI water upon their suggestion (this moderately slows the etching but gives a smoother etched surface) and it would take about 1-1/4 hours to etch through the .025" (.60mm) sheet, working from both sides. This was with heating the bath to about 100oF (40oC). With the spray tank, this same concentration ate through the same brass in about 25 minutes. I now dilute the ferric chloride by 1/2, and it takes about 30 minutes for the same thickness. I suspect that a mist-type spray bottle (with no metal parts [yes, ferric chloride will attack even stainless steel, although slower]) might work well for applying the ferric chloride, for someone with the patience to spritz the brass every 10 seconds or so. If you are etching from one side only you can economize on resist by coating the back side of the brass, after cleaning, with lacquer. If you do coat both sides of the brass with resist (and are using the type that remains where exposed to light) remember to expose the back side also. Randy Gordon-Gilmore [A] frank@bnkl01.astro.ucla.edu (Frank Henriquez 4/95) In article <3nj37h$937@martha.utk.edu>, Scott Kirkham <kirkham@utkvx.utk.edu> wrote: > I use Radio Shack computer board etching fluid for making custom > photoetched stuff.. Highly caustic stuff, but it works pretty well. The Radio Shack etching fluid is Ferric Chloride, a truly nasty, expensive chemical. It stains just about everything, and it's very difficult to control the etching process (since it's an almost opaque liquid). It's also very hard to tell when the Ferric Chloride is "used up". A far better "solution" is to use Ammonium Persufate. It's very, very cheap - a 5 lb jar of the crystals is roughly $20; Ferric Chloride crystals are at least 3x - 5x more expensive. Ammonium Persufate maskes a clear liquid that slowly turns blue as it dissolves the copper/brass. It's very easy to control the etching process, and Ammonium Persufate doesn't have to be heated for best etching (Ferric Chloride should be heated to around 100F for best results). It's still a dangerous chemical, but not as toxic or dangerous as Ferric Chloride. Most chemical supply houses will have Ammonium Persufate; use it. Take the Ferric chloride to a hazardous liquid waste disposal site - it just isn't worth the bother. > the special Radio Shack resist pen, available at finer Shacks in your > part of the world, and drew the artwork I wanted. Draw it a few times > until you get what you like. Then cut it out, leaving a small border of > raw brass around the edges. Most of my etching has been to make printed circuit boards, but the process of making photoetched parts is the same. I use a laser printer friendly "paper" called Press N' Peel Blue. It's a thin plastic sheet, 8 1/2" x 11", coated with a flat blue material. You put the PNP Blue in the printer and print your design; the toner sticks to the flat blue surface. Next, you place the brass (or copper) material on a heat resistant surface and the PNP Blue sheet on top of the brass, with the design facing, and in contact with, the brass. You then use a regular household iron to IRON the design onto the brass - it takes a few minutes, but the toner heats up more than the surrounding blue material, melts and sticks to the brass. The result is a brass sheet with your design in blue (the toner peels off the blue material, which acts as a tough resist layer to protect the toner). Next, etch the brass in the Ammonium Persufate. You may want to use masking tape on the unused areas of the brass sheet, and on the backside, if it doesn't have a design. You don't want to eatch more brass/copper than you need to! After the brass has been etched, remove the brass from the etching solution, rinse it off and use it. The blue backing and toner are pretty hard to get off; I usually scrub it off my PCBs with a copper and brass cleaning powder (or similar abrasive cleanser) and it takes quite a bit of work. It's probably best to leave it on delicate photoetched parts, or look for a solvent that'll remove the toner and backing. Since most toners are plastic, maybe a solvent like laquer thinner would work...haven't tried it, though. Press N Peel is pretty expensive; about $30 for 20 sheets. The company that sells the material is: Techniks Inc. PO Box 463 Ringoes, NJ 08551 (908) 788-8249. Their ad is on page 111 of the April '95 issue of Nuts & Volts. They also sell a material called Press N Peel Wet that can be used to make decals (they show how in their instruction sheet). I did a test by scanning in a decal sheet from a 1/72 scale Hasegawa kit and printing it (in black) on a printer. The results were dissapointing - even at 600dpi, the laser printer (and scanner) just didn't have the resolution to capture all the fine details of the original. This is a problem with photoetched parts, too. Don't expect to catch the fine details of a commercial photoetched part. For instance, forget about reproducing an instrument needle in the dial of a 1/72 scale instrument panel. Larger scales probably wouldn't have this problem, though. Hope this helps. Frank BTW - if anyone out there has used the Press N Peel stuff * AND KNOWS OF A CHEAPER SOURCE OR EQUIVALENT * please let me know! [Q] Are there any better etchants than the messy ferric chloride stuff? [A] jbober@ida.org (James T. Bober 7/95) The best etchant I've found is based on perchloric acid and ceric perchlorate. It is available from GFS Chemicals in Columbus Ohio as product # 5910. Price is $25.50 for 4 liters.
rec.models.scale FAQ, part 7
FAQ Table of Contents