New to Self Etching

[Michael Tibes]

I've just recently received the first boards I ever got manufactured professionally and I'm so happy about not having to deal with the whole process and chemicals. Also the quality is so much better than anything I could do myself plus double sided is a no-brainer. I did self-etch a few boards over the past decades and the more recent tries utilising the toner-transfer method worked out ok. Nevertheless, it is incredibly time-consuming for me to do it and my time is probably better spent on producing and engineering ;-)

I used AISLER - Powerful Prototyping made in Germany because they are in Europe and guarantee high environmental standards. With 'far away' companies I do have a bit of a trust issue in that aspect, though they'll probably be even cheaper. The turnaround time was about two weeks. For my projects I'll probably not do prototype boards any more but start with a pcd right away, kicad and the manufacturing options make it so easy - and much more foolproof as a process.

Nevertheless, I remember etching a double sided pcb for a modem card for a C64 in the mid eighties. I am proud of having managed ;-)

Michael

 

[rock soderstrom]

Nevertheless, I remember etching a double sided pcb for a modem card for a C64 in the mid eighties. I am proud of having managed ;-)

That is the right spirit!

Does Eisler still have the minimum order quantity? What does a double-sided euroboard 160x100 cost, for example?

 

[rock soderstrom]

I used AISLER - Powerful Prototyping made in Germany because they are in Europe and guarantee high environmental standards. With 'far away' companies I do have a bit of a trust issue in that aspect, though they'll probably be even cheaper. The turnaround time was about two weeks.

I understand your argument about the environment.
2 weeks is ok, but if you want to do something small quickly on a weekend, it's a hell of a long time!

 

[Brian Roth]

In my recent times, I selected OshPark in the USA for my project(s). AFAIK, they are USA based for the fabrication. A few friends have used China shops with good results. One thing I like from the "real" fabs is that I get solder mask and my custom silk screening, which is WAAAY above what I ever did in my kitchen/self etch days.

http://recordingservicesandsupply.com/item/otari/output-driver-replacement-for-/lid=44716247
As for a quickie thing, I use perfboard/veroboard.

Bri

 

[Rob Flinn]

To stop the tracks rotting in the past I have sprayed with a thin layer of acrylic lacquer. When you solder this just gets burnt away. I have preamps I still use that I made like this20 years ago an they're still fine.

However, laziness has taken over & nowadays I use JLC pcb to get boards made, they're cheap have a pretty fast turnaround & the results are excellent. I have been doing a few boards with edge connectors & it's relatively cheap the get them gold flashed. However don't let me stop you having the experience !

 

[abbey road d enfer]

I looked at a slightly larger model at a buddy's place, the idea was to use it for labelling front panels as well.

Milling aluminium, even engraving is not easy. It requires either a sophisticated clamping device or a Bed Leveller. Then it takes a lot of time (and broken bits) to achieve good results.

Matching machines were more like 500€ upwards 5 years ago

Prices have gone down incredibly in just a few years.

and yes, they made fine dust and stench.

Dust, I understand, and it's not a big issue, even in a bedroom.
Stench, I don't get it. Well, it can happen if burning material (or bits) due to incorrect settings, but I never had any issue.

My etching bath doesn't do that.

For sure, but it has other cons, like disposing safely, or staining everything that comes in contact.

No doubt there are many useful ways of prototyping boards, my process fits my low demands as a hobbyist perfectly.

Many ways to skin a cat, for sure.

I just had a look at the model you mentioned. They are really very cheap. Are they any good?

They're good, as long as you don't expect too much.
Actually, the smaller machines (2418/3018) are pretty good because the bed is a single piece and the shorter the dimensions, the better the precision.
Actually, I don't use my CNC for PCB's for two reasons:

• Most of my PCB's are 2-layer, which is actually possible with a CNC machine, but requires better eyes and hands than mine (and patience).
• My machine is big one (9060) with a 3HP spindle, so not really adequate

But I have a friend who uses his 3018 for PCB's and gets excellent results. He's much cleverer than me...

Edit: If I were going in a similar direction, I would probably buy a laser engraver/cutter today.

Cutting copper requires a lot of power...

But they really stink!

... and burns the substrate, which stinks...

This is a big plus, because drilling is the most time consuming/ boring part in my process.

Indeed. And you break much less bits when the machine takes over. No problems of perpendicularity.

 

[abbey road d enfer]

I use JLC pcb to get boards made,

That's what I used to do, but recently the prices for shipping and handling have gone a little crazy (I believe due to the fact that the chinese gov't has reduced subsidization of export transport costs).
Now I use Aisler in Germany. I'm not sure it's a good move for those who are not in the EU...

 

[Rob Flinn]

That's what I used to do, but recently the prices for shipping and handling have gone a little crazy (I believe due to the fact that the chinese gov't has reduced subsidization of export transport costs).
Now I use Aisler in Germany. I'm not sure it's a good move for those who are not in the EU...

I have found they do different postage levels & by selecting the cheapest option it's not so bad. But I will certainly look at Aisler as an option because I prefer to do business in Europe than Asia.

 

[RuudNL]

Do you have to keep sodium persulfate at a certain temperature during etching or does it work well at room temperature?

Sodium persulfate works on room temperature, but if you want to speed up the etching process, you can heat it up a little. (40 C.) Sodium persulfate dissolved in water is a colorless solution that gets light blue if copper is in the solution. The color gives a good indication how much the solution is saturated with copper.

https://www.reichelt.nl/nl/nl/etsmiddel-natriumpersulfaat-600-g--etchant-600g-p23424.html?r=1

 

[JohnRoberts]

This "old fart" will share some more stories about DIY etching PCBs.

Back in the 70s it was common practice for small companies to etch our own prototype PCBs... Indeed heating the etchant speeds up the process. We put a glass bowl full of ferric chloride on an electric hot plate to get our prototype boards etched quicker. Apparently the bowl was not happy with direct heating and the bowl split apart, spilling the load of ferric chloride onto the red hot heat element.

The vaporized etchant was very aggressive reacting with everything metal in the room. We had a steel drill press in that room that instantly completely rusted over. Luckily nobody was in the room when that accident happened, I can't imagine breathing in that toxic cloud.

====
While I was a technician working at the MIT instrumentation lab, an engineer I was supporting wanted a quick simple PCB. I painted the pattern of PCB traces on some copper clad board stock with blue tool ink (used to make scribe marks ). Nowadays they sell conductive ink pens that would be even faster, but this was over half a century ago, so making a PCB from scratch in a couple hours was quick.

====

For those who want to hear yet another Peavey story, Hartley was/is frugal and encouraged all of his PCB artists to not leave significant bare areas on finished PCB designs, to conserve etchant life. They were advised to fill up the bare PCB spaces with copper patterns. Some of the PCB layout artists got pretty creative. I vaguely recall issues with disposing of the spent etchant on a commercial scale.

JR

 

[abbey road d enfer]

I have found they do different postage levels & by selecting the cheapest option it's not so bad.

Interesting.
It seems that there is no option for me, only DHL Express.

 

[Jarno]

Etched a lot of PCBs, but always hated drilling a f*ckton of holes, even though I did buy a little Proxxon pillar drill for it.
Most of the PCB's I etched with a mixture of muriatic acid and peroxide. These chemicals are easily obtainable, at least in NL. It also works on aluminium, used it on guitar stompboxes and frontpanels.
Works at all temperatures, so I have etched with the basin propped up on the snow outside. Especially with aluminium, pretty exothermic.

But these days, cheap chinese PCB houses for the win, double sided, plated, soldermask, silkscreen. Unless you desperately need to test out a circuit on sunday afternoon, no brainer.

 

[Rob Flinn]

Interesting.
It seems that there is no option for me, only DHL Express.

I checked out the German manufacturer & it was significantly more expensive for me, but thanks for the tip, always worthwhile checking these things out.

 

[rock soderstrom]

Sodium persulfate works on room temperature, but if you want to speed up the etching process, you can heat it up a little. (40 C.) Sodium persulfate dissolved in water is a colorless solution that gets light blue if copper is in the solution. The color gives a good indication how much the solution is saturated with copper.

https://www.reichelt.nl/nl/nl/etsmiddel-natriumpersulfaat-600-g--etchant-600g-p23424.html?r=1

Okay, I know this stuff, it's called Natrium persulfate in Germany. The transparency and color is great, but unfortunately the etching result was not so good, it was extremely temperature dependent. Below 40 C almost nothing happened but above 55 C it began to crystallize. The etch pattern was not pretty either, I had problems with partial under etching of the tracks. Strange, maybe a bad batch, but maybe also user error.

I went back to ferric cloride.

 

[MidnightArrakis]

@justinheronmusic

How do you plan to create what we used to call the "artwork" used to make the "mask" for the blank copper clad material? I can remember/ramble on. Bishop Graphics "stickies" anyone? lol

Kicad here and send Gerbers to a vendor.

@MidnightArrakis I await your comments! <G!!>

Bri

[I await your comments!] -- OK!!! I'll throw in a few of my personal comments on this topic for everyone to read.....

WA-A-A-A-A-A-YYYY before I actually became a "PCB Designer", I was a "Private" in the military stationed at a -- literally -- SECRET base located in the middle of the Utah desert!!! There wasn't - ANYTHING - around this base for 100-miles in every direction but flat desert and Salt Lake City was just over 90-minutes away, while driving at stupid high-speeds on unpopulated desert roads!!! OH, YEAH!!! To be young and foolish during your early 20's!!!

This military base is where the U.S. Army conducted a lot of their CBRN (Chemical/Biological/Radiological/Neurological) tests and some of these test sites are/were several square miles in size. So.....imagine a radio-type broadcasting tower being several hundred feet high as a "center-point" and then try to imagine thousands of "receiving test stations" located in dozens of spaced concentric circles or rings, with each "test station" located maybe 25-feet or so from one another. Since each concentric ring of "test stations" becomes ever larger and larger and with the furthest ring being a mile or two away from the "center-point", you can only imagine how many "test stations" there had to be overall. It was all staggering to me!!!

These "test stations" all fed their data back to the "test-site central command" building that contained, at the time, an IBM main-frame computer. The IBM computer would receive all of this data, analyze it, compile it and record it on those fast-spinning tape decks that you now see in "SCI-FI" movies. So.....what does all of this have to do with "home-etching PCB's" you ask??? Well.....this:

The engineers who were involved with these tests complained about the incoming data becoming corrupted and ruining their test results. "What can be done to fix this?", they asked. The "Head Director" of this testing group asked an Electrical Engineer who was from Michigan if he could design any circuitry that could clean-up the data signals as they were coming into the IBM computer. The Electrical Engineer guy said, "Yeah.....Sure"!!! Then.....the "Head Director" looked at -- ME -- and said, "Well, Williams.....his circuitry is going to need to be put onto some Printed Circuit Boards -- AN-N-N-ND -- the circuit boards will need to be put into some kind of an enclosure. So.....why don't you just figure out on how all of that needs to be done, OK"??? WOW!!!

So.....here it is in the year of 1972, I had hardly even seen a Printed Circuit Board at that point in my life at that point in time and the only "kind of" mechanical design I had ever done was during my "Drafting" class when I had "Shop Class" in the 9th-grade!!! OK.....This will certainly be interesting!!!

So.....one day the Electrical Engineer (Dave) and I get into a military Jeep and drive-off into the desert to visit several of the actual "test stations" so he can make a series of electrical measurements of them in order for him to know what type of data they are outputting and all of their parameters. As mentioned earlier, since these test sites were each several square miles in size, it took us several hours to not only drive from site-to-site, but also to drive around each test site to a number of the test stations. And, being out there in the middle of the Utah desert, under the direct desert sun, it was a rather hot, dirty and sweaty job!!!

After Dave had collected all of the electrical information he had needed, we drove back to the "Central Command" base and Dave set about to analyze his measurements against the schematics of the "test stations" and began to design circuitry that would "clean-up" the data being fed into the IBM computers. All of this was just "the first step" of the process.

While Dave was designing his circuitry, I was quickly forcing myself to learn about what it would take to "manually hand-tape" Printed Circuit Boards, since I had -- NEVER -- done this before!!! From somewhere, I got some kind of a catalog that had the materials needed to "hand-tape" PCB's and I ordered a bunch of it. In addition, I also had ordered some large rectangular glass baking bowls for cakes, some etching solution, photosensitive PCB laminates and other materials. In addition, I also cleaned out some sections of a large "locker" assembly (like used in a gym to store your clothing and personal items) and created a small and cramped "dark-room" with a swivel-lamp to expose the artwork onto the laminate material. You really had to be "creative" back then when things didn't exist or could be just bought like you can today!!!

After a couple of weeks, Dave finished his circuitry design and he had come up with 8-circuits that he felt would clean-up the incoming data from the desert "test stations". Now, it was up to me to convert his schematics into 8 different PCB-designs, which was something I had -- NEVER -- done before!!! This'll be good!!!

Because I was generating all of the PCB layouts "manually" with no photo equipment, my layouts had to be done on a 1:1 basis of the actual size. And, since this was 1972 and being located in the middle of the Utah desert, pre-set footprints like those that were available from "Bishop Graphics" during the later 1970's just weren't available to me then and I had to manually create all of the component pin-spacings and pad sizes, in addition to coming up with a workable components placement and routable PCB.

But.....I persevered and managed to layout all 8 of the circuit boards. Dave and I kludged together an etching station which also included a small motor assembly that rocked the glass enclosure back and forth to help the etching solution do its thing. When all of the boards had been etched, just like what someone else here in this thread had mentioned, I had to sit down in front of a drill press and manually drill-out all of the component pads so I could mount and solder all of the parts. There were a whole lot of holes to drill out!!!

Since there were no silkscreens on the boards, I had made some kind of an "Assembly Drawing" of each board so I knew where each part went and their orientation. Amazingly, all of the boards and their parts soldered very easily and the finished boards looked really great when each one was completed.

Mechanically, I had designed a small card-cage enclosure for all of the circuit boards. However, for some unknown reason, my memory of the mechanical design details about this project just "aren't there", so I will have to skip that part here.

However, I do remember that the finished mechanical card-cage enclosure and its 8 circuit boards were all designed to fit within a small empty hole within a mound of electronic equipment were contained within a large radome that was mounted onto a flat-bed semi-trailer truck outside of the building which housed the IBM main-frame computer. I had to crawl inside the radome and then crawl into and then behind all of the equipment in order for me to wire-up my little enclosure to the IBM computer wires. It was really weird for me doing all of this as I "knew what I was doing, but I didn't know what was really going on"!!!

A few days after I had installed my little enclosure into the radome equipment and the Electrical Engineers who conduct the tests there at this military base had determined that electrically everything seemed to be a "GO", they then conducted some "Mustard Gas" test out at one of the test sites and all of the engineers very eagerly closely watched all of the data streaming into the IBM computer. Tensions were high in that computer room for the first few minutes!!! Then..........they all looked around at each other and everyone started hooping and hollering and screaming and yelling in JOY because.....for the first time, > ALL < of the data coming into the IBM computer was "clean as a whistle" and was perfectly recorded onto those fast-spinning computer tapes!!!

With me being only 23-years old at the time, I stood there within that computer room and I was amazed to realize that my efforts of not only coming up with the mechanical design of a small card-cage enclosure, but also the PCB-designs of several small circuit boards of circuitry that did "something" to this electrical data coming into the IBM computer made all of this important stuff actually work!!! This experience showed me that -- I -- can "make a difference" in a bigger picture of things. AMAZING!!!

Fast-forward about 20-years.....and I then find myself sitting in an R&D laboratory at the NASA/Goddard Space Flight Center designing 8- to 16-layer Printed Circuit Boards on a computer using CAD-software that costs $20,000 per seat!!! Some of the PCB's are so complex with thousands upon thousands of nets that it takes an auto-router an entire weekend working 24-hours a day just to route these circuit boards!!! And, to think.....I had designed my first 8 circuit boards with absolutely no previous experience or knowledge of what I was doing other than using my "common sense" on how I thought things should be done.

Now.....I have nearly the same PCB-design software as I had at NASA installed here-at-home on my home computer. Life's good!!!

/

 

[justinheronmusic]

Whoa, this thread really took off! I am excited to go back and read some of these ol' war stories of the days where people were self etching PCB's in an underground bunker after collaborating on a circuit design via morse code!

For real though, I would normally go the route of Eagle/Vendor fabricating, but the design I am building is actually a Nixie clock on instructable, where self etching is the only option. The design is fairly complex and would require more time copying it into Eagle/ect. I felt like it was a good opportunity to teach myself something new.

 

[RuudNL]

Okay, I know this stuff, it's called Natrium persulfate in Germany. The transparency and color is great, but unfortunately the etching result was not so good, it was extremely temperature dependent. Below 40 C almost nothing happened but above 55 C it began to crystallize. The etch pattern was not pretty either, I had problems with partial under etching of the tracks. Strange, maybe a bad batch, but maybe also user error.

Funny, in the more than 40 years I am making my own PCBs I never had this problem!

 

[rock soderstrom]

Funny, in the more than 40 years I am making my own PCBs I never had this problem!

As already mentioned, probably my mistake.
I think the most important thing is to find a process that works well for you. Then optimise it a bit over time and stick with it.

I am so experienced now with my process that I could almost do it with my eyes closed.

 

[abbey road d enfer]

I am so experienced now with my process that I could almost do it with my eyes closed.

I am so experienced in doing things wrong, I could probably do them with my eyes closed and one arm tied behind my back!

 

[rock soderstrom]

I am so experienced in doing things wrong, I could probably do them with my eyes closed and one arm tied behind my back!

That sounds good, we should start a business together!

Funny, in the more than 40 years I am making my own PCBs I never had this problem!

Ruud, did you see that your post links to this page? You would also fit into our new company LOL!

Edit: Screenshot deleted