I was being overly conservative. I have just test etched a small scrap of 1 oz PCB in 7 minutes.
Temperature makes a large difference to etching rate. I float my sloshing container in hot tap water.
I don't do heaters, so there's that. But in the summer, I'm peaking out at upwards of 110F.
Love the internet: hot tap water should apparently be no hotter than 120F in my country.
I also run my tank lean on acid. Higher acid increases the etch rate fairly significantly, IIRC. But it also decreases the etching accuracy. My internet source was some paper by a guy name.. Sechel? Something like that.*** My own experience is limited and biased, but it seems to match. I routinely run my tank devoid of acid. I can tell because the etching hits a wall, but also the head on the etchant grows and bubbles over like crazy (adding acid at this point looks like a Gas-X commercial). The etchant will also locally turn blue where you squirt in plain water, when the acid is really low/out. I use maybe a turkey baster of 30% HCl per 6x9" panel, depending on the coverage. I routinely add some acid just before etching each successive board, and/or during etching if the process stalls out.
*** I found it
http://techref.massmind.org/techref/pcb/etch/CuCl2.htm ; but in fact, Seychell's entire paper entirely ignores accuracy and is focused on speed, alone; the only con he seems to mention regarding excess acid is the amount of fumes. Adam Seychell's focus on etching speed may have got some people off track. And his estimations for amount of recharged etchant needed to etch a given area of 1 oz copper appear to be extremely optimistic to me, but maybe that was affected by acid concentration.*%* From the opening statement, it seems like he wrote this specifically with the idea of a hobbyist using a vertical bubble tank for strictly air-regeneration, but he never describes what is requires to achieve this in practice. Many people believe you throw an aquarium bubbler in there, and you're good to go. That will work if you use a 2-3 gallon tank for a small board once every day or 2. For a vertical etching tank, 2-3 gallons is pretty big. And when you're done with the corrected revision, a day or 2 is a long time to wait to spin the new board. Most of us might only etch boards once or twice a year, if that. But when we are doing it, we're sometimes burning the candle on both ends to solve problems that need to be solved. Air regeneration is simply not practical with an aquarium pump. It would be appropriate for tiny-scale, regular production, of a same-board-a-day, or whatnot. But we don't work like that. A big part of the allure of toner transfer is you are completely free to make mistakes and immediately fix them. Ordering boards, there are a lot of details you might forget and then remember after submitting the order. Or you might not notice your obvious mistake until you receive the boards, week or 3 later.
I measure the temperature of the backing paper (vinyl foil is laminated to regular or tracing paper), so while accuracy is questionable, the consistency should be there and I shouldn't be getting reflections from metal surface.
No, I don't get smearing unless I overdo it with the heat gun, but at that point I'm approaching the bubbling point of FR4. Since I apparently don't need such high temperature anyway, I don't risk it.
Can you leave it unsupervised for majority of the 30-45 minutes etch, or do you need to babysit it? Do you get undercutting?
Do you need to bubble the etchant after/before the etch to revitalize it, or is it enough to bubble it during etching alone?
I have never even set a timer; the boards are fine to sit in there for an extra hour, even. The traces start to get thinner, but it is a long time before there are any problems with my 8/8 boards. The traces get thinner, but thinner all over, and not more ragged. It's crazy. I have forgotten a board overnight, once. Obviously, the air ran out after the first 1-1.5 hours. The board was ruined and most of the resist was floating at the top. But a lot of the traces were still there, only they were 2 mils thick. I will never get anything looking like a rat chewed on it; it doesn't happen.
It's enough to bubble it during etching, alone. But if you wanted to, it takes maybe (wild guess) 5 to 7 minutes for my fine bubble blaster to turn the cola in my 0.4L tank back to mountain dew.* So if you wanted to get a little head start on the next board, I suppose you might bother to regenerate it. In a matter of weeks, it will regenerate on its own, just sitting there.
Adam Seychell: "If you want to be pedantic then the solution can be aerated after each etching job until the color is a completely transparent bright green, i.e. Cu1+ concentration << 1 g/L."
*So my system isn't going to win any races, but it finishes a board on one setup/air-tank, unsupervised, and that's where I stopped caring. So I don't know how much faster you could get with better regeneration. The original plan, using a top end fish tank bubbler, it did not do the job. I think those only get up to 4 psi, or so. Using an aquarium pump to put air through either a tube with holes or stone or wand, nope. That doesn't regenerate the cupric for squat. It took hours to recharge a tank vs minutes with the higher volume of finer bubbles I get with compressed air/wand. I couldn't practically etch anything without adding peroxide, using wimpy bubblers... yield was terrible on fine pitch boards... and yet it took me months or maybe even years before I was convinced of this, myself, that the majority of the internet is completely wrong, thinking you can etch with cupric... then recharge it, after; uhhh, no. Unless you use a relatively enormous quantity of etchant for the board.
edit:
http://www.chemcut.net/wp-content/uploads/2015/03/Chemcut_Bulletin_8_Cupri_Chloride_Proces_-Parameters.pdfAccording to this info, increasing free acid increases the etch speed, but it also proportionally increases the amount of relative undercutting. Temperature changes the etch speed, but it has essentially no effect on undercutting. So running at 130-160F will increase the speed, if your equipment can handle it. But for a home user doing one-offs, a heater might not matter as much. The specific gravity apparently hardly effects speed at all, as long as you're within a wide band. And it, also, has almost no effect on undercutting. According to this link, you can etch a 1 oz board in something around 100 seconds** using cupric chloride in a heated spray etcher. Cu2 vs Cu1 is measured with an electrical probe in order to determine the amount of regenerative agents added in the recharging loop (chlorine gas or 35% peroxide), and the tank is run with some Cu1 left in there, to prevent excess reagent from being added (excess chlorine gas would escape into the atmosphere; excess H2O2 would turn the etchant into mountain dew acid peroxide). So for a home user trying to get the best possible accuracy and not concerned too much with absolute speed: keep the acid down, don't put peroxide in there, and be patient. Heater can technically increase speed, but air regeneration rate will inevitably be the weak link, anyways. As long as the etchant is not frozen, it will etch boards at a rate at least comparable to warm ammonium persulfate, at the worst; it'll take longer when it's cold, but accuracy will not be affected; it won't be practically unusable, like cold ammonium persulfate etchant.
** I think with spray etching, you will perhaps be able to keep a higher concentration of Cu2:Cu1 on the surface of the board? I mean, you are continually flooding the surface of the board with near-fullly charged etchant. It would also be reasonable to assume there will be a higher rate of atmospheric oxidation of copper; hence there could be more etching via acid directly dissolving copper oxides.
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
*%* Let's examine closer. From his paper, Seychell states you should be able to etch half-coverage 1 square decimeter of 1oz copper with 1.6L of charged etchant. Who knows if this means etch time of hours or more, due to scraping the bottom of the barrel at the end? But anyhow, a square decimeter is 15.5 square inches. A 4x6" board is 48 square inches, double-sided. So it should take... 4.9 L of charged cupric to etch an average 4x6" double-sided 1 oz board. This is 1.3 US gallons.
6x9" double-sided (I do this size, regularly) would take 11.15L, or 3 US gallons. My tank holds ~3.4% of that. With my setup, it's all about the regeneration, because this is the only way it is practical (i.e. 99% of us should stick with ferric or acid peroxide; and yeah, you can call it etching with cupric, if you like). With 3.4% of the volume, I can crank out these boards in succession if and when needed. Refill the air, swap the boards, add some acid and water (1 turkey baster of acid vs 3 more gallons of green koolaid).