bubble vs spray etch pcb

[Siwastaja]

Yeah, we have used Dupont Riston off Ebay successfully. But we happened to have two big rolls from the industry from the early 90's, which should be badly out of date, but it still worked perfectly so we have been using that free stuff without any issues.

You can wet laminate it with a squeegee (spray a little bit of water - stick it to the clad - squeegee all the water out - oven at 80 degC for a few minutes IIRC), and we started with that, but then we rearranged our process so that we CNC drill first, and wet lamination stopped working due to the holes. (Drilling first makes alignment easy, and allows you to compensate printing the mask to actually fit holes, instead of destroying the complete board you have worked with for an hour by accidentally drilling it with an offset.)

In the industry, the dry film is laminated using heat and pressure - it's basically a big-ass heavy-duty office laminator!

Which got us thinking, maybe a standard super-cheapo office laminator (meant for laminating paper in plastic pouches) would work.

And yes, it worked perfectly! This is easier than wet lamination, as well.

Biggest PCBs we do are A4 size (around 210x290 mm).

We built the point source UV unit just using random Ebay UV LEDs... No issue there. Sometimes the Ebay LEDs are crap, though, so we were lucky.

IIRC, 380nm would be the optimal wavelength for dry film materials, but 405 nm works well enough (requires maybe 3-4 times more power), but 405nm is at least ten times cheaper per watt so that you obviously want to use that.

[CopperCone]

Do you think its feasible to make multiple layers at home using this dry film method?

For some reason I always had alignment problems with my projector sheet print outs on double layer boards, if I used the default trace thicknesses and stuff from altium, I ended up having to really tape the bottom sheet down and then kinda stretch the top sheet over it under light, then staple it, so it had some tension in it, to get good alignment where I can correctly line up a DIP package or VIA. I thought it was because the ground plane had alot of block on it, so it got thermally warped from the laser printer during printing? I notice this more on big PCB, biggest I made was about 10x6 inches. Like I think I would tape the top side to a sheet of glass then stretch the bottom on it.. it was tricky business that I was frustrated with. Took forever to make it look good when I held it up to a light source to check for alignment. 

Does it work better with inkjet? I don't expect to make multilayer RF boards, but things that operate in the ultrasonic range at low noise levels with top and bottom ground plane/shields, analog signal layer, power plane, digital plane, etc.

Also what glue do you recommend for bonding PCB's together?

[David Hess]

The good bubblers I have seen used an array of tubes covering the entire bottom of the tank with a spacing of about an inch or less between the holes and the tubes.  The holes were large enough and the air volume high enough to roil the surface of the enchant so a cover was a good idea.

[CopperCone]

so you are supposed to use big ass bubbles?

[David Hess]

so you are supposed to use big ass bubbles?

If the idea is to agitate the tank, then yes.  I am just reporting what we used but I doubt bubble fizz is going to work nearly as well.

[Siwastaja]

CNC drill the holes first, then expose one side at a time, so you can align them separately, which will be much more accurate and less risky than trying to join the two masks together and get them register reliably.

Each printer will require you to find compensation (resize) parameters so that the masks align with the CNC'd holes. I output hi-res images, then resize in GIMP or similar before printing. For example, the resize factors may look like this: X=99.95%, Y=99.80%

This takes a bit of trial and error.

Getting good masks printed is always the most difficult part.

I wouldn't try multilayer.

[tautech]

so you are supposed to use big ass bubbles?

If the idea is to agitate the tank, then yes.  I am just reporting what we used but I doubt bubble fizz is going to work nearly as well.

Well yes a byproduct of bubbling is agitation but it's not its primary function which is to remove etching reaction bubbles from the PCB. Too much air cools the the etchant and that's undesirable for any of the heated etchants.

[CopperCone]

uh so what bubble size should I go for?

[tautech]

uh so what bubble size should I go for?

Study:
https://www.eevblog.com/forum/manufacture/bubble-vs-spray-etch-pcb/msg1708010/#msg1708010
 

[CopperCone]

uh so what bubble size should I go for?

Study:
https://www.eevblog.com/forum/manufacture/bubble-vs-spray-etch-pcb/msg1708010/#msg1708010
 

but someone else is using a diffuser and I don't know if their results are bad because of it.

[tautech]

uh so what bubble size should I go for?

Study:
https://www.eevblog.com/forum/manufacture/bubble-vs-spray-etch-pcb/msg1708010/#msg1708010
 

but someone else is using a diffuser and I don't know if their results are bad because of it.

Tank design will be the decider and the reason why I built mine skinny so a single bubble tube is all that's needed.
If you've got a wide tank and can do multiple PCBs of course you'll need either multiple bubble tubes or some sort of bubble diffuser/spreader.

[David Hess]

so you are supposed to use big ass bubbles?

If the idea is to agitate the tank, then yes.  I am just reporting what we used but I doubt bubble fizz is going to work nearly as well.

Well yes a byproduct of bubbling is agitation but it's not its primary function which is to remove etching reaction bubbles from the PCB. Too much air cools the the etchant and that's undesirable for any of the heated etchants.

More heat!

We were using ammonium persulfate purchased in drums so the tanks were heated.

[james_s]

The biggest challenge I had building a bubble tank is getting the bubbles to be evenly distributed. I suppose that could be a challenge with spray etch too but it seems it would be simpler to get an even flow. I went with bubbles because I wanted a compact vertical tank and didn't like the idea of HCl spraying around.

[jmelson]

Do you think its feasible to make multiple layers at home using this dry film method?

Yes, but...
Laser printers are not accurate enough to do two-sided boards of any significant size.  I built a laser photoplotter that is accurate to a few thousandths of an inch over the whole film size.  So, I print out the two artworks and have alignment marks on them.  I take a piece of scrap PC board material and glue it to one edge of one artwork.  I then slip a piece of Plexiglas the same thickness as the board between the two films and align.  Then, I glue the 2nd film to the PC board scrap, making a kind-of 2-page book.  To print the board, I just slip it between the "pages" and expose both sides without releasing the vacuum in the double-sided exposing frame.

But, for double-sided boards over 100 mm, I just don't know what others can do, without an accurate source of artwork.  To check yours, make a grid of lines on 1" / 25 mm spacing, and print one normal, and one mirror-imaged on your printer.  Then, put them on a light table and see if the lines align.  You will likely be very unpleasantly surprised at how far off the two sheets are.

Jon

[KL27x]

The biggest challenge I had building a bubble tank is getting the bubbles to be evenly distributed.

Those of us who have discovered the "bubble wall" wand are very happy with it. The main drawback is you need higher pressure than what you can get with a fish tank diaphragm pump.

I am no stranger to doing detailed work, but getting the bubbles to work evenly by drilling holes in a tube was a fruitless task. The bubbles are huge, and half the air wanted to go through a few of the holes.

The bubbles you get from a bubble wall tube are almost microscopic. I once tried to take a picture, but they didn't show up. I can't see them until I almost have my nose against the tank. They're so small, they don't even perceptibly rise to the surface... so the etchant level in my tank will rise the better part of an inch when the bubbler is on and there's a huge volume of air in there, and a ginormous surface area of contact between etchant and air which works great for cupric. It is super impressive. It feels like the bubbles should coalesce, more. I have speculated that by shooting enough air in there and adding a tiny dash of soap, you could etch with 100% foamed etchant. I doubt it would be a benefit, but as long as you are intent to overcomplicate things (like anyone who wants to make a spray etcher), this might be of interest to you.

Poked holes in a tube just make big bubbles that rise to the surface in 1 second and splash more etchant, which is fine if your etchant just needs the bubbles for agitation.

Well yes a byproduct of bubbling is agitation but it's not its primary function which is to remove etching reaction bubbles from the PCB. Too much air cools the the etchant and that's undesirable for any of the heated etchants.

Maybe this is more important for ammonium based etchants. IME, ferric and cupric doesn't produce significant bubbles on the pcb , unless you have added peroxide. With pure ferric with a low/zero free acid, I think the agitation is important to displace the solid sediment?

[CopperCone]

I realized why I never had serious alignment problems: I always use a solid ground plane.

When I get more transparancies I will test the alignment method by the way you describe.

[CopperCone]

Ugh, testing my tube carefully, I noticed its difficult to do this right.

What I see is that when you increase the pressure to get significant agitation, the bubbles form a mushroom cloud type thing in the tank, where they are pulled in towards the middle. It seems to work ok with very small pressure with the bubbles going strait up but I found that over the last 2 days the pex tubing self sealed alot of the drill holes.

I think the proper way is to make a square bubbler out of acrylic and put it on the bottom of the tank and drill holes uniformly at the base, then to carefully adjust the gas flow to prevent the convection like effect with the bubbles being pulled in towards the middle to make a trunk. Buncha small holes drilled for nothing. Maybe you even need to stuff it with some kind of wool material to make a pressure drop across it so the bubble emission is uniform.

I see basically no bubbles on within a few inches of the sidewall of the tank. If I get enough flow to make a few centimeters of agitation on the top you get the trunking effect with significant deadzones on the edges. I don't know how uniform the trunk is, if its fairly uniform you can make a wider tank I suppose, but the tank needs to be like 40% wider then the PCB in my case. I did not see this with the shitty bubbler I had before for some reason.

[tautech]

Yeah and that's why I decided to use the PVC gutter guard mesh as a PCB carrier because it helps disperse the bubbles rather than have them just rise vertically to the surface. The mesh rests on the bubble tube so the PCB is right near the bubble path.

Well yes a byproduct of bubbling is agitation but it's not its primary function which is to remove etching reaction bubbles from the PCB. Too much air cools the the etchant and that's undesirable for any of the heated etchants.

Maybe this is more important for ammonium based etchants. IME, ferric and cupric doesn't produce significant bubbles on the pcb , unless you have added peroxide. With pure ferric with a low/zero free acid, I think the agitation is important to displace the solid sediment?

Yes but there's the agitation benefit of the bubble path too that keeps solution from stagnating in one place and that can of course speed the etch.

[CopperCone]

do you notice the convection/mushroom cloud effect in your setup ?

How much wider is your tank then the PCB? I can only use like 60% or less of my tank.

[tautech]

do you notice the convection/mushroom cloud effect in your setup ?

No, I'm careful to control the air flow so to not create a mist of etchant however there is a tiny bit of splatter.

How much wider is your tank then the PCB? I can only use like 60% or less of my tank.

From an old thread:
Here's mine, knocked up from stuff I had lying around.
Base: plastic road marker 
Tank: custom thermoformed 150mm waste pipe  (using large propane torch and timber and wedges to get to flat form)
PVC solvent adhesive for plastic gutters.
Bubbler tube (not visible): extruded (straight) semi soft plastic tube, multi pierced with a stiff needle (fine ground file tang)
PCB holder: folded gutter guard


Fits 220 x 150 mm PCB. 750 ml etchant capacity.

[KL27x]

Another 2 cents:
I try to leave at least 4" empty at the top in my vertical tank. But I can go all the way up when I need to. When I etch larger boards, I lay a bit of paper towel over the edges that are uncovered to soak up the overspray.

There is zero splashing with the microbubbler I use for my tank. The bubbles form a head where they coalesce into larger bubbles. The popping of the bubbles is what shoots tiny bits of etchant up about 4".

This is one reason why virtually all commercial vertical etch tanks are much taller than they are wide. Mine is  square because of the logistics of buying small amounts of plexiglass and not wanting to spend more than I had to. 

[CopperCone]

do you notice the convection/mushroom cloud effect in your setup ?

No, I'm careful to control the air flow so to not create a mist of etchant however there is a tiny bit of splatter.

How much wider is your tank then the PCB? I can only use like 60% or less of my tank.

From an old thread:
Here's mine, knocked up from stuff I had lying around.
Base: plastic road marker 
Tank: custom thermoformed 150mm waste pipe  (using large propane torch and timber and wedges to get to flat form)
PVC solvent adhesive for plastic gutters.
Bubbler tube (not visible): extruded (straight) semi soft plastic tube, multi pierced with a stiff needle (fine ground file tang)
PCB holder: folded gutter guard


Fits 220 x 150 mm PCB. 750 ml etchant capacity.

No, you misunderstood me. I meant a bubble mushroom cloud under water, where the bubbles seem to be pulled towards the middle of the container in a column, rather then flowing laminarly upwards. I basically get no contact with the PCB within like 2 inches of my container walls. If I drastically decrease the effect the problem goes away.

It seems there is water recirculating near the walls of the tank in some kind of eddies or some shit, so the bubbles are pushed away from the sides.

It seems like its going to be a pain in the balls to make this right.

[CopperCone]

like this



They are literarly pushed away into a lower pressure region or some shit, I can see sharp deflection angles.

[tautech]

No, you misunderstood me. I meant a bubble mushroom cloud under water, where the bubbles seem to be pulled towards the middle of the container in a column, rather then flowing laminarly upwards. I basically get no contact with the PCB within like 2 inches of my container walls. If I drastically decrease the effect the problem goes away.

It seems there is water recirculating near the walls of the tank in some kind of eddies or some shit, so the bubbles are pushed away from the sides.

It seems like its going to be a pain in the balls to make this right.

You can see as mine is so narrow and the PCB is central and in the mesh carrier I have none of those issues.

If you have a strong enough bubble flow the agitation will cause central column movement upwards while the edges of the tank will have a downward flow.
Naturally the bubbles will want to remain central in the tank.

[KL27x]

I've tried 1.5" tank. Went to 3/4". Copper, do you hang multiple boards in your tank with some sort of spacer? I'm curious to see a solution for this.