Want to start making my own circuit board where to start

[Beamin]

I guess first thing is a laser printer. Anything special about it? Or can I just go with a cheap one? Also how much am I looking to spend on it?
What do you print on? Staples had a box of 100 clear transparencies but they were 70 bucks! I know staples is a rip off but at that price this would be cost prohibitive.
anything special about the UV light? Will a black light work? I don't like the idea of eye damage. Ferric Chloride seems to be the best chemical except for its not safe to dump down the drain but I can manage safe disposal.

I also saw a youtube video where the guy prints on what looks to be a magazine page and irons it on the copper. Has anyone tried this?

[KL27x]

Plenty of people swear by UV. But it is not particularly hard to do toner transfer with very similar results. Unless you need smaller than say 7/7 traces, regularly, toner transfer can do it cheaper. And it is faster/easier in almost every respect, once you understand the entire process. It will do everything short of BGA (really can't do that with UV, either, unless you wanna plate thru holes, yourself?), although UV will perhaps have a slight edge in the minimum size of your vias.

If you wanna do toner transfer, I can offer some pointers.

[tggzzz]

I guess first thing is a laser printer. Anything special about it? Or can I just go with a cheap one? Also how much am I looking to spend on it?
What do you print on? Staples had a box of 100 clear transparencies but they were 70 bucks! I know staples is a rip off but at that price this would be cost prohibitive.
anything special about the UV light? Will a black light work? I don't like the idea of eye damage. Ferric Chloride seems to be the best chemical except for its not safe to dump down the drain but I can manage safe disposal.

I also saw a youtube video where the guy prints on what looks to be a magazine page and irons it on the copper. Has anyone tried this?

You need to give us more information, e.g. how you will design the PCB (cf manufacturing it) acceptable turn-around time, size of components and tracks, PCB size, single/doublesided, other attributes, and how you will assemble the components on the PCB.

You can read about my experiences making PCBs after a 30 year (gulp) gap at my blog; see the .sig. I was very pleasantly surprised at how easy it is nowadays. The major techniques are:
laser toner

• laser toner (works, just about OK unless you have practiced your technique)
• cheap Chinese PCBs (excellent, but there is a turn-around time)
• manhattan or dead-bug techniques over a groundplane (excellent for one-offs including medium speed digital and RF)

I haven't tried UV, so I won't comment. As for solderless breadboards: just say no!

[Rerouter]

for transfers grab a glossy magazine from a freind or find a cheap one, cut it to A4 sizes, and print on that. you will want dpi set as high as you can, and toner darkness as high as it will go.

Be aware that depending on how your printer interprets vector lines, it may not have seamless bends, e.g. the image attached, its 8/6 traces, but note how the 45 degree corners end up broken, its because my printer treated the segments seperatly, by going in to the printer settings and setting it to use raster images rather than vector i was able to fix it.

The image in this example was printed on a hp laserjet 4000, but my days of toner transfer are over. and i use a board house to punch it out, you would likely want a laser with better than 600dpi to handle below 10/10.

[rstofer]

I did this stuff back in the early '70s and I won't be doing it again.

I haven't learned how to use one of the more complete software packages like KiCad that will generate Gerber files and allow me to see off to China - nor would I want to wait.

I use ExpressPCB for everything.  It doesn't have pads for BGAs but I don't know how to solder them anyway.  Or, more to the point, I don't know how to inspect the solder joints.

ExpressPCB has a schematic entry program linked to a board layout program.  Everything is placed and routed manually, there is no automation, and it takes a while to get the layout just right.  Then a little online order and the boards are in hand in about a week.

I have been using ExpressPCB for about 12 years and I really like their service.

I have one of the Miniboards with a dozen or so DIPs and 100 or so discretes so this is a very capable service.

[Beamin]

For time I don't want to mail order, I want to do everything at my house. For size its mainly going to through hole or if surface mount its going to be the largest size packages I can find for analog parts. For IC's that have to be surface mount I will go with the largest size I can find, I don't like working with small things. I don't have the skill to create anything crazy: Think 555 timers and simple op amps. No crazy bga's or super fine pitch pins. Sorry I don't know the size numbers. Is the ironing a magazine tonner transfer method? Seems like a good way to simplify things.   

[KL27x]

I have done toner transfer for over a decade. It's only once every few months, but sometimes I do a lot of boards at a time.

Here are the two important bits you might not find on a google search.

1. paper. The Pulsar paper they now sell at Mouser and Sparkfun and all sorts of places is amaze-balls. Yes, you can use magazine paper. I've done it. Yes, you can use laser printer label backing. Yes, even regular paper works. But I'd suggest you at least buy 1 pack of this Pulsar for reference. Because it works as good or better than any of them, and it is by far the most likely to give you good results if your technique is not great. It releases easy, leaves no fuzz around the edges. BTW, magazine paper is pretty good, but it distorts a little, which is a problem only on large double-sided boards. I think it shrinks or wrinkles when exposed to high heat.

2. Pre-etch. This is incredibly important. After cleaning the boards off with soap and steel wool, you still benefit (GREATLY) from dunking your blank board into your etchant for about 10 seconds. Rinse it in a bucket of water. Then let it dry (I recommend a heat gun to speed it up). Very important: Do NOT WIPE THE BOARD. There will be a layer of precipitate that is loosely adhered to the surface of the board, all over, and it is the reason why you want to do this. If you wipe the board down with a damp paper towel, you will remove all this dust.  The dust gives the surface nooks and crannies for the toner to stick and to seep in, rather than flattening out and/or falling off. If I don't do a pre-etch, my traces can come out anywhere from normal to about 2.5x as fat as they're supposed to (this means bridging between pads/traces and general wavi/uneven-ness). And they are more prone to falling off. With a good preetch, you really can't mess up.

As far as the printer, I know firsthand that Brother printers which take the TN350 cartridge will not work. Most any laser printer other than those will more than likely work. Including other Brother printers.

If you want to make it easy on yourself, you can buy GBC Creative laminator for maybe $30. And a 1200W heat gun for another 20-30. What I do is take the case off the laminator and shoot the board with a heat gun while it's going through. The amount of heat depends on the size of the board. I have run a lot of boards, and my first laminator is still trucking. Stored outside, near a corrosive tank, ten years.  The only other mod besides removing the plastic case and screwing it to a plank of wood for a base was to egg out the holes for the roller, so that a full thickness 0.064 board will fit through. Just be careful. The metal roller is juiced up, so don't feed a copper board in barefoot or with wet hands. I've gotten a tingle, before.

Thickness of the copper: this is usually described in "oz". Stick with 1 oz, or even 1/2 oz pour, for the most part. These will etch much more accurately than 2oz. If you need 2 oz, just take that into consideration when you decide what pitch you can handle. 

If you just want to make a tiny board that probably requires some bodging, just to say you did it, and you dont' want to spend any time or money to do it right, watch Ben Heck's toner transfer video. 

[tggzzz]

For time I don't want to mail order, I want to do everything at my house. For size its mainly going to through hole or if surface mount its going to be the largest size packages I can find for analog parts. For IC's that have to be surface mount I will go with the largest size I can find, I don't like working with small things. I don't have the skill to create anything crazy: Think 555 timers and simple op amps. No crazy bga's or super fine pitch pins. Sorry I don't know the size numbers. Is the ironing a magazine tonner transfer method? Seems like a good way to simplify things.   

You haven't indicated the essential benefit from limiting yourself to doing everything at home.

BGAs are relatively difficult to inspect and rework; I'd avoid them.

Apart from that, anything down to 0603 and 0.5mm pitch can be assembled easily by hand with limited equipment.

If you only use toner transfer, I think you are limiting your options.

[Zero999]

I've done toner transfer before but it can be a pain and lots of experimentation is required to get it right.

The photographic process is the way to go, in my opinion. You can buy plain copper clad board and spray it with the photoresist but I wouldn't bother, just get pre-sensitised board, as it's much easier. UVA and the shorter visible wavelengths (indigo and violet) will work best. Don't worry about eye damage. These wavelengths are safe, as long as you're sensible and don't stare into the beam. You can buy cheap LEDs with a peak wavelength of 400nm which should work quite well but I haven't tried them myself. I'd recommend a black light compact fluorescent lamp. Don't bother with incandescent lamps: they're inefficient and get hot enough to burn/melt the transparency.

https://www.amazon.com/13W-CFL-Blacklight-Bulb-Vibe/dp/B01IYDSGFK
http://www.ebay.co.uk/itm/20W-BLACKLIGHT-ULTRA-VIOLET-UV-BULB-BC-B22-BAYONET-FITTING-BLACK-LIGHT-LAMP/252742747657

[janoc]

I guess first thing is a laser printer. Anything special about it? Or can I just go with a cheap one? Also how much am I looking to spend on it?
What do you print on? Staples had a box of 100 clear transparencies but they were 70 bucks! I know staples is a rip off but at that price this would be cost prohibitive.
anything special about the UV light? Will a black light work? I don't like the idea of eye damage. Ferric Chloride seems to be the best chemical except for its not safe to dump down the drain but I can manage safe disposal.

I also saw a youtube video where the guy prints on what looks to be a magazine page and irons it on the copper. Has anyone tried this?

Laser is good for doing toner transfer (in fact, inkjet will not work for that at all). That's the part with the magazine paper and ironing. Search for "toner transfer" on Google, there are tons of tutorials on it.

If you want to do UV photolithography, then laser printer is actually not the best option. You want as opaque black as possible and cheap lasers rarely have sufficient toner coverage on larger areas. Office grade printers usually do, but unless you can find one second hand, they are not cheap. There are some tricks, like stacking two printouts to get more opaque black, but it is both more costly (you need two transparencies) and can be a royal pain in the butt to align.

For doing lithography your best bet is a cheap inkjet. Or have the transparencies printed in the nearest copy shop.

UV light - black light may work, depends on the wavelength. You do not want a germicide/disinfection lamp (that's where the eye damage can occur). It is not necessary for this job. If you don't want to buy/build an exposer using UV tubes, look for a small UV lamp used by ladies to cure their glue-on nails. Those are sold cheaply in drugstores.

Another option is to use UV LEDs. It does work surprisingly well, there are plenty of instructions on how to build one online (e.g. http://www.instructables.com/id/UV-LED-Exposure-Box/ ) and the UV leds are fairly cheap.

However, in a pinch even sunlight works, it just takes a while.

[janoc]

You can buy plain copper clad board and spray it with the photoresist but I wouldn't bother, just get pre-sensitised board, as it's much easier.

Actually, the cheapest and easiest method is to use the dry film resist. You only need a cheap office laminator to laminate it on the board. Much cheaper than pre-sensitized boards and you can use any scraps or pieces of the laminate you may have laying around, not only the few sizes the distributor carries. Spraying I really don't recommend neither - messy and next to impossible to get an even layer without dust on it.

[tautech]

For time I don't want to mail order, I want to do everything at my house. For size its mainly going to through hole or if surface mount its going to be the largest size packages I can find for analog parts. For IC's that have to be surface mount I will go with the largest size I can find, I don't like working with small things. I don't have the skill to create anything crazy: Think 555 timers and simple op amps. No crazy bga's or super fine pitch pins. Sorry I don't know the size numbers. Is the ironing a magazine tonner transfer method? Seems like a good way to simplify things.   

Good on you for wanting to give it a go.
There's a bit to learn and some expenditure to get all you need.

Stick to SS boards to start with. I use Paxolin rather than FR4 as it's cheaper and easier to work with.
Ironing works for some but a laminator makes the process more repeatable.
Glossy papers are best for release of toner to the copper but a lot depends on your printer and its settings.
Then there's etchants.......start with something you can buy off the shelf until you get some experience.
Just start with a tray for etching but sooner or later you might want a etching tank.....not too hard to make.

Don't try anything too complex to start with, keep good wide clearances and say 1/2 mm (20 mil/thou) traces. Wider for power. Copper plane fills reduce the amount of copper that need be removed but unless you have solid prints there will likely be some porosity.

Do lots of reading on the subject, start with a forum search for 'toner transfer' in the Search button not the box at top right. (it's useless). Getting the toner transfer bit sorted is the key, printer, toner, paper and the transfer process each/all have an influence on the result. Doesn't look right, scrub the transferred toner off the copper and try again.
Investment in time getting a reliable/repeatable process is the key here and when you do you can knock boards out from start to etched in under and hour.

[Zero999]

Laser is good for doing toner transfer (in fact, inkjet will not work for that at all). That's the part with the magazine paper and ironing. Search for "toner transfer" on Google, there are tons of tutorials on it.

If you want to do UV photolithography, then laser printer is actually not the best option. You want as opaque black as possible and cheap lasers rarely have sufficient toner coverage on larger areas. Office grade printers usually do, but unless you can find one second hand, they are not cheap. There are some tricks, like stacking two printouts to get more opaque black, but it is both more costly (you need two transparencies) and can be a royal pain in the butt to align.

The problem with incomplete coverage with laser printers can also be an issue for toner transfer too. Inket isn't ideal either, as it can run and bleed, causing lines to loose their definition and be thicker than they should be. Personally, I haven't had an issue with incomplete coverage with laser printers, when using the photographic process. I've found it to be more of a problem with toner transfer.

UV light - black light may work, depends on the wavelength. You do not want a germicide/disinfection lamp (that's where the eye damage can occur). It is not necessary for this job. If you don't want to buy/build an exposer using UV tubes, look for a small UV lamp used by ladies to cure their glue-on nails. Those are sold cheaply in drugstores.

Another option is to use UV LEDs. It does work surprisingly well, there are plenty of instructions on how to build one online (e.g. http://www.instructables.com/id/UV-LED-Exposure-Box/ ) and the UV leds are fairly cheap.

I agree with you about not using germicidal lamps. It's completely the wrong wavelength and is totally unsuitable for exposing photoresist. There's a bad tutorial on another site which recommends germicidal tubes: if you happen to come across it avoid it and any other project on that site.

Tubes from insect lamps also work quite well, along with the sort used in fraudulent bank note detection lamps.

Wow, a decent project on instructables: they're not many of them on that site!

[janoc]

The problem with incomplete coverage with laser printers can also be an issue for toner transfer too. Inket isn't ideal either, as it can run and bleed, causing lines to loose their definition and be thicker than they should be. Personally, I haven't had an issue with incomplete coverage with laser printers, when using the photographic process. I've found it to be more of a problem with toner transfer.

It likely depends a lot on the printer used. E.g. my Samsung ML-2010 is marginal. It works for toner transfer but large areas are not covered well, there is a lot of visible banding and pitting. I have to do a lot of manual "postprocessing" with a felt marker before etching and even then it could end up being crap.

I have also tried to use it to make transparencies using tracing paper* and unless I have stacked at least two, it was not sufficiently opaque for the dry film resist - even covered parts were getting exposed slightly. Two stacked produced lines that were blurry in some places because of misalignment or maybe the glass I have used to compress it didn't put even pressure on the entire surface.

(* You don't have to use real plastic transparencies - paper is enough, because it is mostly transparent for the UV light. Another advantage is that it warps less in the laser printer than a transparency, even those that are actually intended for lasers)

On the other hand, the larger color HP LaserJet printer/scanner combo we have at the office produces shiny, very smooth and almost completely opaque black, so it is likely that that one would work very well. Will need to check which toner it uses and see whether I could get a suitable second-hand printer off eBay.

Wow, a decent project on instructables: they're not many of them on that site!

I have actually built a board with UV leds based on that one to test it - it works pretty well for exposing the dry film. I am definitely more comfortable playing with that than having to deal with fragile tubes and high voltage.

[sleemanj]

You can buy plain copper clad board and spray it with the photoresist but I wouldn't bother, just get pre-sensitised board, as it's much easier.

Actually, the cheapest and easiest method is to use the dry film resist. You only need a cheap office laminator to laminate it on the board.

Don't even need that, I use a clothes iron, set very low.  After cleaning the pcb, spritz it with some water, remove the cover film from inside-of-curl of the dry film in the usual manner and lay it on the wet pcb, rub it down from the center outwards with a small wad of paper towel to expel the water (water makes it easy to do without bubbles and wrinkles, and you can lift it off to grab any bit of dust that got trapped before you heat it), and then use the iron to rub and heat it which will fix it to the copper.  Iron must be quite cool to avoid blistering the film, 50-70 degrees (c) works well.  You can also use hotair - sometimes I will just grab that instead of fishing out the iron, I have heard a hairdryer will do the job, but I don't have enough hair to own one :-)  Leave pcb to cool before exposing.

[ion]

I've not had much luck with toner transfer - I've made a few boards but they always took several attempts to get the toner released from the paper intact.
In my opinion photolithography is a far better process.  UV tubes sold for insect killers work quite well for exposure.

I have also tried to use it to make transparencies using tracing paper* and unless I have stacked at least two, it was not sufficiently opaque for the dry film resist - even covered parts were getting exposed slightly...

I also use tracing paper to print the masks on and I find the results to be really good if I use the best quality print on my inkjet.  I think HP black ink is pigment based, which would help.  I seem to remember using a laser printer for my first couple of attempts - the results weren't as good.

A reminder... Keep in mind that doing solder mask at home isn't that easy as etching copper, and without solder mask, god help you on soldering surface mount parts...

If you use the dry film solder mask it isn't that hard, you apply it just like the dry film resist.  It just takes a bit of experimentation to get the exposure and curing times right.

[Zero999]

I have also tried to use it to make transparencies using tracing paper* and unless I have stacked at least two, it was not sufficiently opaque for the dry film resist - even covered parts were getting exposed slightly. Two stacked produced lines that were blurry in some places because of misalignment or maybe the glass I have used to compress it didn't put even pressure on the entire surface.

(* You don't have to use real plastic transparencies - paper is enough, because it is mostly transparent for the UV light. Another advantage is that it warps less in the laser printer than a transparency, even those that are actually intended for lasers)

Just for clarity's sake, you are talking about tracing paper there, rather than ordinary copier paper? I agree, tracing paper is perfect and will work with most printers better than film. I don't know about ordinary copier paper, which I'm pretty sure is opaque to UV, as it fluoresces a greenish blue, under UV, although that might still be a short enough wavelength to expose the photoresist.

Wow, a decent project on instructables: they're not many of them on that site!

I have actually built a board with UV leds based on that one to test it - it works pretty well for exposing the dry film. I am definitely more comfortable playing with that than having to deal with fragile tubes and high voltage.

Another possibility is black light LED strip, which can be cut up into pieces and stuck to the inside of a suitable enclosure. I don't know if it's more expensive per watt of UV, than using 5mm LEDs, (the SMD LEDs in the strip are more powerful than 5mm units) but it will save a lot of soldering.
http://www.ebay.co.uk/itm/5M-12V-REAL-UV-Ultraviolet-Strip-Light-Tape-Light-SMD5050-LEDs-Blacklight-/192119454976

[janoc]

Just for clarity's sake, you are talking about tracing paper there, rather than ordinary copier paper? I agree, tracing paper is perfect and will work with most printers better than film. I don't know about ordinary copier paper, which I'm pretty sure is opaque to UV, as it fluoresces a greenish blue, under UV, although that might still be a short enough wavelength to expose the photoresist.

Yes, tracing paper, not regular copier papier.

Another possibility is black light LED strip, which can be cut up into pieces and stuck to the inside of a suitable enclosure. I don't know if it's more expensive per watt of UV, than using 5mm LEDs, (the SMD LEDs in the strip are more powerful than 5mm units) but it will save a lot of soldering.
http://www.ebay.co.uk/itm/5M-12V-REAL-UV-Ultraviolet-Strip-Light-Tape-Light-SMD5050-LEDs-Blacklight-/192119454976

Good idea! Haven't seen those before.

[KL27x]

I've not had much luck with toner transfer - I've made a few boards but they always took several attempts to get the toner released from the paper intact.

I swear people make it harder than it needs to be, just so they can build a UV LED box. I don't know how cheap is UV dry film. But the cost of Pulsar is insignificant compared to the process. And while I have not done UV, it seems obvious that UV requires more space and time and effort.

Once you have got it down, it doesn't take several attempts. I haven't had to redo a toner transfer in probably 4 years. Whenever Pulsar hit the market. Scores of boards. First try is the only try. I could do it with my eyes closed and be confident in the result. There is no guesswork or maybe in the process. It's very straight forward when you don't skip steps.

[janoc]

A reminder... Keep in mind that doing solder mask at home isn't that easy as etching copper, and without solder mask, god help you on soldering surface mount parts. You will end up with using tons of flux and a very messy work space.

Naaaw, come on. Few people if anybody does solder mask at home and people solder even large TQFPs and QFNs on home made boards, no problem. Even 0.4mm pin pitch parts are doable (done it). Yes, without solder mask it is a bit harder but hardly impossible. Just don't use huge blobs of solder and do use plenty of flux. I seriously don't care if it is messy - it is easy to clean off with a bit of alcohol and these are usually prototype boards anyway.

[KL27x]

^ that's what I do. Exposed copper tracks are not an issue for me. If I make some error, which is rare, I can clean off excess solder with solder wick, if it matters.

If I need corrosion resistance, I clean the board after assembly with alcohol brush and blotting with paper towels. And then I paint it with silicone conformal coat.

without solder mask it is a bit harder but hardly impossible.

I don't agree, even with this.
I have found essentially no advantage in populating a board that has proper soldermask vs nothing. Even commercial FPC don't have soldermask between IC pads. The soldermask is more properly functioning as "trace cover," to me. To prevent shorts if mishandling and to protect from corrosion. The only place it matters is where I have thermal pad to a plane. Without mask, I need to take more care to prevent a big blob. That is the biggest "issue," if you can call it that. One biggest problem, IMO, is that without the proper magnification and lighting, looking at unmasked board is harder on your eyes. Copper and components and solder blobs can all kind of blur together.

[tautech]

I've not had much luck with toner transfer - I've made a few boards but they always took several attempts to get the toner released from the paper intact.

Which indicates your TT process wasn't well enough developed.
I've had that too and when I'm in a hurry and want to get the PCB etched, just soak it in water for a bit until the transfer paper becomes soggy and rub the paper away with your fingers leaving just toner on the PCB.
One has to get all/any paper coating off the PCB too before going into the etching bath.

[KL27x]

Even recent article on Hackaday on toner transfer totally omits the pre-etch part which I deem just THE MOST IMPORTANT PART.

http://hackaday.com/2016/09/12/take-your-pcbs-from-good-to-great-toner-transfer/

"I have no idea how the Internet rumor that you should turn up your iron “as hot as possible” came about. But I do know what the result of doing so is — smeary transfers that are extremely sensitive to the amount of pressure applied. That’s the exact opposite of what we’re looking for here. Instead, the goal is to keep the toner in the glass transition, with the temperature as low as possible so that it will fuse with the copper at our roughly constant maximum pressure."

I am here to tell ya's, you can't use too much heat if you do the pre-etch. The liquefied toner is absorbed into the layer of precipitate dust. There is no fine tuning. Get it hot enough, press it a bit, and there is no way to mess it up. You can't get it too hot. You can't press it too hard (although it takes very little pressure once you get the toner totally liquefied. The pre-etched board will suck up the toner like a sponge). The only top limit to the temp is when the FR-4 bubbles/burns and the copper delaminates! I also notice he adjusts spacing of his tracks so the space between them is 2 mils larger than the tracks, themselves. This is probably an adjustment he makes for the slight fattening of traces he gets, because he doesn't pre-etch the boards.

I'm sure he makes great boards. If you want to make great boards, but you don't want to have to fine tune things and do hours of experimentation, it is very, very easy. Pulsar paper. Pre etch. If you are gonna remember only 2 things, make it these two things. You can get the temp so hot that a layer of toner melts into the paper underneath the dextrin. And you get trace outline on the paper and on the copper. This is wayyyy hotter than you need. And the board will still come out fine.

[janoc]

Even recent article on Hackaday on toner transfer totally omits the pre-etch part which I deem just THE MOST IMPORTANT PART.

You don't need to pre-etch - scrubbing the board clean with a scouring pad and then properly degreasing it will do the same job, cleaning and roughing the surface for the toner to stick to. Pre-etching cannot harm but it is not really necessary and it saves you one messy chemical pass.

I have never had issues with toner not sticking to the laminate except when I have rushed the job through the laminator and the board remained too cold for the toner to fuse properly.

I haven't tried the Pulsar paper, 12-15 euro before tax for 10 sheets (Farnell or Mouser) is heck expensive for what it is! I am normally using regular paper for non critical things and for very fine stuff I am using self-adhesive vinyl. Stick a piece on a regular paper as a carrier, print on the vinyl and then laminate it on the board. The vinyl will cleanly release the toner once hot and soft. That way you won't get any paper "hairs" and much less pitting and hairline breaks in the traces so common with paper. On the other hand, the vinyl is much more prone to smearing.

Dry film resist is a lot cheaper - first link I got on eBay:
http://www.ebay.fr/itm/PCB-Photosensitive-Dry-Film-for-Circuit-Production-Photoresist-Sheets-30cm-x-5m-/221849849115

5m roll for about 5 bucks.

[KL27x]

scrubbing the board clean with a scouring pad and then properly degreasing it will do the same job,

No, it doesn't. The scouring/sanding degreases the board. It might add some very little bit of surface texture. I clean/degrease the board, too. This is an unrelated step. (I have used scotchbrite in the past, but steel wool and dish soap is the fastest and easiest to me.)

I'm not talking about the toner just falling off. It will also FATTEN the traces making them 2 or even 3x thicker than the original print. This will cause bridges.

When I pre etch in my cupric, this does not just put little micro pits/texture on the surface of the board. It leaves a fine, porous, fragile surface. It binds the toner and prevents it from squishing out laterally. It sucks up liquid toner. This fattening/smearing issue is why this guy has spent so much time to fine tune the exact temp in his process. He is striving to get the toner just hot enough to transfer without liquefying. This is a fools errand because gauging the temp is difficult. The exact settings and time will depend on the size and thickness of the board. 

In my process, there is no upper limit. There is no calibration. The sky is the limit. And I can get down to 6/6 when needed. For 8/8 and higher, there is no careful temp or pressure regulation. There are no inspections and redo's. It works. Across the entire board. Every time. And it doesn't work like this without the pre-etch.

Initially, it took me several months to appreciate the importance of the pre-etch. When you do board only once a month or more, you miss things. You shrug things off. Your board is maybe different size, different pitch, different everything. That must be how it always happens, you think. And beyond that, it took me 8 years to figure out WHAT the pre etch actually does. I eventually tried to pre-etch a stack of boards in advance. Well, in storage, I found they release some dust. I wiped one of the boards off with a damp cloth... and lo and behold my traces came out all wavy and up to 2x the width in spots. (The other 7 pre-pre-etched boards were almost as good as a freshly pre-etched board). Luckily, this was a 12/12 board. And I salvaged it by scraping out the 2 or 3 bridges that resulted. With my normal routine, I do 8/8 and don't even look at the board. I don't inspect it at any stage, because I know what I'm going to see. The pre-etch leaves the surface so fine and porous that it corrodes into dust just sitting there. Even gently wiping the board with a damp paper towel will remove the effect. And it probably even leaves dust fragments in the surface of the board from the start. You cannot replicate that with a Scotchbrite.

As for cost of the paper, this is nothing to me. I don't care about the upfront cost of a laminator and a laser printer. I don't care about the cost of the paper. I don't care I spent over $100.00 on my etch tank. I care about my time. Even if UV method is free, it takes more time and space and effort. I'm probably a high volume user compared to most people. I buy the pulsar 5 packs at a time. Probably spend around $100 a year. I'm not rich, yet, but this is nothing to me. The process is simple and fast. There is no darkroom, no developer, no exposure box. No chemicals to pour or mix or store. Just a laminator, a heatgun, and an etch tank that is always ready to go.