X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f X-Recipient: geda-user AT delorie DOT com Message-ID: <20170618223803.27683.qmail@rahul.net> Date: Sun, 18 Jun 2017 15:38:03 -0700 To: geda-user AT delorie DOT com Subject: Re: [geda-user] DIY PCBs In-Reply-To: References: <20170617081144 DOT 16928 DOT qmail AT rahul DOT net> <20170618153703 DOT D607D801F77A AT turkos DOT aspodata DOT se> X-Mailer: VM 7.19 under Emacs 23.4.1 From: conover AT rahul DOT net (John Conover) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: geda-user AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: geda-user AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk Not meaning to start an etchant war in the geda list, heating the etchant, (perhaps by setting the etchant container in a plastic container of 150 F hot tap water,) speeds up the etching. Every 10C increase in etchant temperature reduces the etching time by half. A 10cm X 7 cm, 50% interconnect can be etched in about 5 minutes with HCL/H2O2 etchant. 10X10 mil lines are a good baseline, since QFP packages require them. Its more complicated since how long a 10X10 mil, (in both directions on the PCB-they produce different as-made lines in each direction,) can be made is the issue. In the web page, I used two lines coiled up, (20 feet each,) on a 10X7 cm PCB, with no shorts or opens allowed as the certification. This required a 5X5 mil to be available over short runs, (several inches,) as minimum for the lithography. Note that it is different for positive and negative resist. In one, the lines go away, the other, the interstitial space goes away, as the limit of the lithography is reached. (On the web page, the 5 mil lines came out 6 mils, mostly due to artwork artifacts produced by a US $100 laser printer, using inexpensive cellulose vellum as the media.) What prompted the web page is that 3M is going out of the transparency business, (an artifact of the digital revolution-I used 3M religiously for many years,) and finding/certifying an alternative media required requalification of the entire process. As far as expense goes, its about US 4 cents, (plus the cost of the PCB,) in expendables, for a 10 cm X 7 cm PCB. The fast method processing time is 12 minutes and 19 seconds, (record time,) to the drill press. With that said, perhaps it would be appropriate to get back to geda(1) stuff. FWIW ... John BTW, I have no affiliation with any click through products on the page. The technique is used by trade schools in third world countries, and availability at distant places was a requirement. That's also why the entire system runs on 12 V, (as in solar charged lead acid batteries.) Also, why costs had to be minimized. John Griessen (john AT ecosensory DOT com) [via geda-user AT delorie DOT com] writes: > On 06/18/2017 10:37 AM, karl AT aspodata DOT se wrote: > > I usually use natriumperoxidisulfate for etching, would there be any > > advantage to use H2O2/HCl instead ? > > The advantage I like for middle of the road 10 mils line and space, (not attempting 6 or 4 mils), > is that the CuCl2 etchant is easily neutralized so it is not an environment problem. > When you add NaOH to make it neutral, the copper precipitates out as oxide dust > that is heavy, so easy to separate out of the bottom of a pan. The water is slightly salty > with NaCl after neutralizing, but OK to pour on the ground without killing plants. > > I've not found any recycler that buys copper oxide dust, but we have a household disposal > service by the city that will accept things as long as they are not industrial output. > > > It can be turned back to copper by heating in a retort with charcoal and no oxygen... > but I gave that kind of gas fired foundry work up in '02. Maybe later with a microwave > absorbing crucible in a 7X microwave, as in the "microwave steel" method... -- John Conover, conover AT rahul DOT net, http://www.johncon.com/