My first try at making a pcb

[Fluxed Matter]

Well today was my first time attempting to make a PCB at home.  Here is my story. I have been planning and preparing for some time now. I chose to do the UV exposure of photoresist method to make my board. To start out with I am using the presensitized boards from MG chemicals. Since this is my first try I am trying a simple design that I found on the internet. I am using a nice EPROM eraser that I found at my local surplus to do the exposing. I wanted to try a more environmentally friendly way of doing the developing and etching. For the developing there really is not, but for the etching I decide to try using vinegar and peroxide for the etchant. I am using transparencies for the artwork mask that are for copy machines. I notice they distorted when run through the laser printer but since I was only using the the edge for a very small board I feel it was not significant enough that it did not matter. I was not sure how long to expose the the board so I did 10 minutes. Overkill I know, but I did want to take a chance. While I was exposing I mixed up the developer. 1 part developer to 10 parts water.  When the timer went off I developed the board, but in my excitement that the process was working I over developed it. On my second try I exposed for 15 minute, again overkill, but this time around I did not over develop it. I only left it in the solution for about 30 seconds. Hurray! Now it was time to etch so I mixed up the solution of 100 ml of vinegar and 100 ml of peroxide because this was based on what I had read was the mixture. In goes the board. The reaction is slow. I feel that it not working as well as I had read. During my research I read that adding salt helped. So after about 30 minutes I added some In hopes of speeding up the etch. I kept waiting and adding a little salt every now and then and after waiting over 2 hours I ended up overetching the board. I have attached pictures the fails. My question is how can I improve this? Has anyone used vinegar and peroxide as an etchant before, and if so how what were your results? I welcome any tips or feedback that are offered.

[Dave]

The first thing you should do is properly prepare the FR4 laminate by using a mild abrasive on the copper so you get it nice and shiny. I like to use a paste for scrubbing kitchen sinks and a kitchen sponge (just gritty enough to get rid of the oxide layer and mild enough not to make deep scratches). You then thoroughly rinse the board and wipe it clean with a paper towel soaked in paint thinner. Avoid touching the copper surface, so you don't contaminate it with oils from your fingers.

I would also suggest you change your etching solution. Mix water, hydrochloric acid (aka muriatic acid; 30%) and hydrogen peroxide (30%) in ratios of 5:1:1, respectfully. Mix the chemicals in the order I have written them. It might be difficult for you to acquire acid and peroxide in the concentrations written above, just use the best you can get (reduce the amount of water you mix).
The board should be done in a couple of minutes. Work in a well ventilated area, as the green fumes (cupric chloride - CuCl₂) produced by the etching process are quite irritating (I learned this the hard way). Use safety glasses, gloves and long sleeves when you are working. A disposable respirator wouldn't be a bad idea either.

[sleemanj]

The first thing you should do is properly prepare the FR4 laminate by using a mild abrasive on the copper so you get it nice and shiny.

No!  He is using pre-sensitised boards!

[sleemanj]

Quinn Dunki uses Vinegar, Peroxide and Salt and pre-sensitised boards, she wrote a detailed blog about it a while back: http://quinndunki.com/blondihacks/?p=835

[Dave]

The first thing you should do is properly prepare the FR4 laminate by using a mild abrasive on the copper so you get it nice and shiny.

No!  He is using pre-sensitised boards!

Yep, didn't really do a good job at reading his post.

[Fluxed Matter]

Dave thanks for your replies. If I do decide to switch the chemistry I will consider your recipe, but those concentrations may be difficult to get.

sleemanj thank you for the information. I will be checking it very soon. 

Thanks for the help guys.

[tautech]

Your etching brew will give you better results if heated, basin in basin of hot water.
The water, hydrochloric acid, hydrogen peroxide etchant does not require heating and will etch quickly and can be saved, reused and rejuvenated.

[Tac Eht Xilef]

A few pointers...

To start out with I am using the presensitized boards from MG chemicals. Since this is my first try I am trying a simple design that I found on the internet. I am using a nice EPROM eraser that I found at my local surplus to do the exposing.

The UV wavelength of EEPROM erasers is a bit short (~250nm) for most presensitised boards e.g. the MG Chemicals unit uses 375nm, and many people use UV LEDs or bugzapper / nail polish curing lamps which are typically ~390nm.

I was not sure how long to expose the the board so I did 10 minutes. Overkill I know, but I did want to take a chance. While I was exposing I mixed up the developer. 1 part developer to 10 parts water.  When the timer went off I developed the board, but in my excitement that the process was working I over developed it. On my second try I exposed for 15 minute, again overkill, but this time around I did not over develop it. I only left it in the solution for about 30 seconds. Hurray!

In my experience, the MG Chemicals boards are remarkably difficult to overexpose, but fairly easy to overdevelop if you do expose them for a longer-than-optimum time. I suspect you'll find it easier to get better results by using a more appropriate light source and exposure times.

Trial and error is the only way to get a good ballpark for exposure in a particular setup. Grab one of the available PCB "test patterns" (or make your own), and try various exposure times. I usually cut a 2cm by 10cm strip off the end of the first board of any new type, and test exposure times up to 10 minutes (by covering it with cardboard and slowly revealing sections of the board & pattern) in 1 minute steps - there's no need for 30 second steps, unless the exposure time for that board type is known to be really short (e.g. Kinsten, which can overexpose in < 90 seconds).

(I know MG Chemicals suggest 10 minutes for their exposure kit; that uses normal daylight fluorescent tubes @ ~5" distance. My own experimental results range from ~7min with a cheap "black light blue" (~365nm) UV CFL @ ~2" distance, to 90secs~2mins with a ~385nm UV tube @ ~3")

I am using transparencies for the artwork mask that are for copy machines. I notice they distorted when run through the laser printer but since I was only using the the edge for a very small board I feel it was not significant enough that it did not matter.

Try using decent tracing paper - draftsman or good office-quality stuff, not the cheap school stuff. It may be a little harder to find than printer transparency sheets, but is usually cheaper (especially if you buy it in A3 & cut it down to A5 for use) and distorts a lot less.

Now it was time to etch so I mixed up the solution of 100 ml of vinegar and 100 ml of peroxide because this was based on what I had read was the mixture. In goes the board. The reaction is slow. I feel that it not working as well as I had read. During my research I read that adding salt helped. So after about 30 minutes I added some In hopes of speeding up the etch. I kept waiting and adding a little salt every now and then and after waiting over 2 hours I ended up overetching the board.ur results? I welcome any tips or feedback that are offered.

The only time I've ever been happy with that method is when I used >99.5% lab-grade acetic acid . Otherwise, it's just too slow - you get bored waiting for it, and you end up overetching.

I use 30% HCl + 6% H₂O₂ (it's a little hard to get stronger than that here in Aus) in a 50-50 mix. As mentioned, work in a well-ventilated area and wear eye protection (safety goggles at least), resistant gloves, and suitable (e.g. old ) clothing. I usually chuck the board in, agitate it gently by rocking the bath backwards and forwards, and wait for the copper to go beyond salmon pink - it's a bit hard to describe, but you'll see it go rough/dirty as the etchant really starts to 'bite'. At that point I lift the board out of the etchant, grab a bit of small foam sponge, and gently wipe the etchant up from the bath and across the board, turning occasionally. Small 3x3cm board usually take me just over 2 mins to etch, while larger boards can take ~5-7min.

As for the "environmentally friendly" aspect ... well, whichever way you look at it and regardless of the etchant used, you're making copper salts - which are mostly fairly nasty things from an environmental PoV.

• Ferric chloride + copper = CuCl(II) (i.e. copper chloride) + FeCl(II) (i.e. a different ferric chloride)
• Ammonium persulphate + copper = copper sulphate + SO₄^-2 (i.e. sulphuric acid, as soon as it finds any water)
• HCl + H₂O₂ + copper = one or more of the copper chlorides + chlorine gas + water
• Vinegar (acetic acid) + H₂O₂ + copper = copper acetate + water + (probably CO or CO2 or carbonic acid); as soon as you add salt (NaCl) to speed it up you're probably back to making copper chlorides (and something with sodium - sodium carbonate? - as well?)

So regardless of the etchant used, you really need to dispose of it properly (or at least neutralise & stabilise it) rather than just flushing it down the drain...

(BTW, don't trust my chemistry there - it's probably roughly right, but I'm an ecologist / environmental scientist, not a chemist - but they all end up with environmentally unpleasant compounds...)

[LukeW]

If, after you've developed the photoresist, you find that the photoresist image is stuffed up, for example underexposed or overdeveloped, just clean all the photoresist off the board, and that leaves you with a blank piece of copper clad board, so it's not entirely wasted. You can potentially re-use it with toner transfer or re-apply spray or film photoresist.

[Fluxed Matter]

tautech thank you for the suggestion. I will look for a way to heat the solution that is not to much trouble. The double boiler method could be messy and requires extra equipment.

Tac Eht Xilef thank you so much for the great response and the chemistry lesson at the end. I always wear PPE when I do this type of work and old clothes because safety is first.  I am not planning to just flush it down the drain but want to dispose of it properly. One of the reasons I wanted to use this method is so that disposal of the waste would be easier, or at least easier to neutralize/breakdown than ferric chloride or ammonium persulphate. I am looking for tracing paper that I can run through the laser printer so I can try that. As the boards get bigger I think it will be required to use. And too bad about the eprom eraser but at least it was only $20 US. Maybe I can change that lamps since the ones that are in it are long tubes with filaments at each end, but that another project for later. 

LukeW thanks. I always keep the fails/scraps for later use. They come in handy sometimes.

[tautech]

Few tips here: https://www.eevblog.com/forum/reviews/laser-printer-transparent-paper-for-pcb-making/

[rob77]

my 2 cents..

don't use HCL + H2O2 inside your lab ! etch your boards outside ! when using hydrochloric acid, the fumes will settle all over your lab and every single piece of iron/steel will start corroding after few months of using HCL inside (even with good ventilation) . been there, trust me

[Psi]

my 2 cents..

don't use HCL + H2O2 inside your lab ! etch your boards outside ! when using hydrochloric acid, the fumes will settle all over your lab and every single piece of iron/steel will start corroding after few months of using HCL inside (even with good ventilation) . been there, trust me

yeah, its far too tempting to etch around your bathroom/kitchen/workshop sink. A splash/drop of HCL will totally ruin the shiny sink finish and require much polishing to restore it.

[Fluxed Matter]

tautech thanks for the link.

rob77 I read that somewhere about it corroding all metal in the area. I have an area outside where I will be doing all the developing and etching.

Psi my wife would never allow me to etch anywhere inside the house. She has allergies and is sensitive to odors.

[Zad]

I have never used the method you describe, it sounds like some experimentation will be useful.

The first thing I did when I started with UV PCBs was a simple masked off piece of PCB (I used a coin to mask it).  Expose it for 30 seconds, take it out and have a look at the resist, then repeat in 30 second intervals. When you notice a distinct colour change, then that's probably "done". Make a note of the total exposure time. It won't be spot on but it will give you a good guide for the next step. Dunk it in developer. If its not exposed enough then it'll be reluctant to develop. When developed, wash it off and put it in the etch. Keep an eye on it until you can see that the copper has etched back to the shaded area. Again, this is just approximate to give you a first guess. Because it the repeated exposures etc it will be a mess anyway. It should help to give you a feel for exposure and etch times, and how the copper behaves.

The second thing I did was to make a strip of test subjects, approximately 15mm per box, with a some lines, dots and characters of various sizes. Nothing fancy. Using your initial guess/calculation as a guide, start off exposing one cell (mask the rest with aluminium cooking foil), then maybe 30 seconds later the next strip, then the next, so you have a time progression of cells. Hopefully the middle will be perfect, to one side will be over exposed, the other side under exposed.

It seems a waste of copper clad, but in reality making controlled tests actually saves resources in the long run, by zoning in on the optimal times and conditions.

[sleemanj]

The second thing I did was to make a strip of test subjects, approximately 15mm per box, with a some lines, dots and characters of various sizes

I made such a stepped artwork:

https://github.com/sleemanj/pcb-test-pattern

Download Zip

worked well to tune in my current rig.

[Fluxed Matter]

Thank you everyone for your suggestions and experiences in rolling your own pcb.  I wanted to post this update since it's been a few weeks. I have two problems really, overdeveloping the board and then overetching afterwards if it even gets that far.

So I have been working on the exposure/developing part first. I had been unsuccessful in what I thought was the overdeveloping problem, but the real problem was with the exposure of the artwork. First thing I did was change the bulbs in the EPROM burner. When i bought it the burner had G8T5 bulbs installed and I replaced them with F8T5BLB bulbs. I did a few more test runs but still the image was overdeveloping and washing away. I was only leaving the board in the solution for 20-30 seconds. I have attached some pictures of each try.  Something was wrong because on the last board I exposed it for 30 minutes.  I could not figure it out, and I kept researching more about the method, the boards I was using, and the lights.  Everything kept swirling around in my head, but I continued looking and in the MG chemicals FAQ it mentions that they develop the boards for 10 minutes at 125 mm away using 15 watt tubes. I was exposing my boards at a distance of 13 mm and not even directly under the lamp. My next step was finding a way to modify my EPROM eraser. I used four 100 mm spacers at each corner and then used cardboard to fill in the 100 mm gap created by the spacers. The last two pictures are the resulted the most recent exposure at a spacing of around 110 mm. I exposed one more test board for my next testing phase which is getting a better etch.

Stay tuned for what happens next. 

[Tac Eht Xilef]

I did a few more test runs but still the image was overdeveloping and washing away. I was only leaving the board in the solution for 20-30 seconds. I have attached some pictures of each try.  Something was wrong because on the last board I exposed it for 30 minutes. I could not figure it out, and I kept researching more about the method, the boards I was using, and the lights.  Everything kept swirling around in my head ...

They way you've written that, it reads to me like you were increasing exposure times in an attempt to reduce overdeveloping. I hope I'm either reading that wrong, or you now realise that's the wrong way around ...

(edit: Ah, I think I get it now - by 'last board' you mean the one in your earlier post that you were exposing under an EPROM eraser, yeah?)

... but I continued looking and in the MG chemicals FAQ it mentions that they develop the boards for 10 minutes at 125 mm away using 15 watt tubes.

Did you notice that they're talking about normal white daylight tubes there, not UV? The exposure time will be considerably shorter for even a "blacklight blue" UV tube.

But that last pic is looking pretty good, and should etch nicely.

So you ended up with 110mm distance - how many tubes / what total wattage? What was your exposure time for that successful attempt?

(BTW I hope that's just a random test pattern, or that 45 degree track is meant to short 2 pins on the bottom RH of the SOIC-14)

[Fluxed Matter]

Tac Eht XileF your right, I should have been more clear about what I was doing. Yes I was increasing the exposure time hoping to alleviate the overdeveloping that was occurring. I went from 10 minutes to just over 30 minutes hoping that this would solve the problem. It did not.

All the board were exposed under the eprom eraser. The boards in my first post were expose with eraser in its stock configuration, and why the second worked is beyond me. After that the next three board were exposed after I switched to the BLB lamps, but with eraser un-modified. And, the last two are with the BLB lamps and with the eraser modified with 100mm spacers.

I guess I should have read the FAQ better because I missed the part about them using daylight florescent tubes. I using two 8 watt tubes 12 inches long (11.3 really) in the eraser. They are Bulbrite F8T5BLB.

Yes it is meant to short the two pins together.

I am going to experiment with different exposure times now to see what is optimal.

[Fred27]

I found that it takes a while to get your timings, etc. sorted but once you do you can get great consistent results - far better than toner transfer.

I use a cheap UV nail light from eBay. This is the correct frequency for presensitised boards. As you found out, EPPROM erasers are not. I only have 2 bulbs in it to help a little with collimation and 4 minutes is perfect. It may look like cheap tat (and it is) but it works.
http://www.amazon.co.uk/Dryer-Shellac-Gelish-buffing-Freeshipping/dp/B0088KJA0A

Use a metasilicate based developer, not sodium hydroxide. It's almost impossible to overdevelop. I give it 3 minutes plus and a little agitation. Rinse well before etching.

Another tip - Ziploc vacuum bags do a great job holding the transparency down firmly on the PCB.
http://www.amazon.com/Ziploc-Vacuum-Starter-3-Quart-1-Pump/dp/B003UEMFUG

When you're done, reuse that UV lamp for Dynamask 5000 dry film soldermask. You'll need a laminator but your PCB will look very professional. This guy (not me) made some good instructables on it.
http://www.instructables.com/id/Dry-Film-Solder-Mask/

[Fluxed Matter]

Just a quick update on my attempt to etch a pcb using basic household chemicals. Taking into account some of the suggestions given here and others I read about. I started to use a hot plate for a water bath to heat the solution of vinegar and peroxide while etching. I also increased the amount of salt to a tablespoon and added it to vinegar I preheated in the microwave. I place the board into solution and immediately it started to fizz and bubble but over time the reaction slowed down and even adding more salt did not seem to help revive it. After a while I noticed the resist started lifting off while in solution. Below are some pictures of the board in solution and after allowing it to soak for 1.5 hours in the solution. Most of the information I found claimed to work in forty five minutes to about an hour, but this was well over that and still not completely etched. I understand it’s a slow etch process and I can deal with that, but if my resist is going to start lifting after being in the solution for some time then this is a problem. In all the reading about using vinegar and peroxide most say to use a 50/50 mixture, but there is no mention of the volume used. I have been use 200 ml of both for a total of 400 ml for my recipe.

Recently during my search for more information I stumble on to a different method to etch the copper. This way is called contact etch or direct etch. It uses a small amount of ferric chloride and a sponge. So I decided to give it a try. Wow! I did a board in 8-10 minutes using only 10 ml of ferric chloride. It was very easy and leaves less hazardous waste then the more common way of using ferric chloride. The last two pictures are of the etched board and the rinse water used, and then again afterwards to rinse the sponge in.

This method worked great the only problem was that the I exposed the artwork image wrong. It should be flipped horizontally.

[Fluxed Matter]

Quick question here. Can liquid tin be recycled? Can it be used again on other board? Can you save in a separate container and reuse it at a later time? Thanks for the input guys.

[Tac Eht Xilef]

I always poured it into a second container & reused it. After a few go-rounds it becomes either too slow or too full of crud to use.

That was the stuff that comes as a liquid - I believe the stuff that comes as a powder & mix with water goes off fairly quickly.

(Disclaimer: I don't use it anymore; I use solder-through lacquer. I found it didn't play well with flux - it's fine with fluxed solder since the flux all boils off / floats to the top of the joint, but if I used added flux & didn't clean it off within a few hours (e.g. if I was building & testing section-by-section) the flux would take the tin off...)

[IanB]

Below are some pictures of the board in solution and after allowing it to soak for 1.5 hours in the solution. Most of the information I found claimed to work in forty five minutes to about an hour, but this was well over that and still not completely etched. I understand it’s a slow etch process and I can deal with that, but if my resist is going to start lifting after being in the solution for some time then this is a problem. In all the reading about using vinegar and peroxide most say to use a 50/50 mixture, but there is no mention of the volume used. I have been use 200 ml of both for a total of 400 ml for my recipe.

Recently during my search for more information I stumble on to a different method to etch the copper. This way is called contact etch or direct etch. It uses a small amount of ferric chloride and a sponge. So I decided to give it a try. Wow! I did a board in 8-10 minutes using only 10 ml of ferric chloride. It was very easy and leaves less hazardous waste then the more common way of using ferric chloride. The last two pictures are of the etched board and the rinse water used, and then again afterwards to rinse the sponge in.

I think if your etching doesn't complete in about 10 minutes or so then it's not a good recipe.

Where hazardous waste is concerned, it is the copper that is the environmental hazard. Whichever etching solution you use, the amount of copper in the waste is the same, it depends only on the area of the board to be etched and the amount of copper removed.

Although ferric chloride seems nasty, it is not really an environmental problem. Ferric chloride is even added deliberately in water treatment plants as a flocculant to help remove cloudiness from the water. Ferric choride contains iron and chloride. Iron occurs freely in nature, and chloride is so common as to be harmless.

So in reality, using peroxide and vinegar is not "nicer" than using ferric chloride. After you have used either solution you will have dissolved copper in the solution, which is a disposal problem either way.