Author Topic: Etching is very slow  (Read 5347 times)

0 Members and 1 Guest are viewing this topic.

Offline dsharp02

  • Contributor
  • Posts: 26
  • Country: us
Etching is very slow
« on: July 21, 2017, 02:23:54 pm »
I purchased some 1oz single-sided copper clad FR1 boards to try my hand at creating a PCB.  I cleaned the boards copper side with a scotch-bright pad and soapy water, degreased them with IPA, and then spray-painted the boards with black spray-paint.  Once the paint was dry, I then used a laser engraver to remove the paint where I wanted the board to etch.

I placed the board into the ferric chloride, and agitated the container until I got tired of standing around shaking it (maybe 10 minutes).  I looked at the board and could see no progress, so I let it sit for an hour.  After an hour, I could just barely start to see spots where the copper had been removed and I could see light through the board.  So I left the board in for another hour.  I could see more clear spots, but it was still mostly opaque.  So I waited another hour, still not complete.  So I left the board in the ferric chloride overnight (7 or 8 hours), and in the morning it appeared to be mostly complete, but there are still a few spots that are still opaque.

12-13 hours seems excessively long to etch a board.  The temperature was roughly 80F/27C in my garage the whole time.  Do I need to warm the etchant?  Could the spray paint I'm using be contaminating the ferric chloride and causing it to not work properly?  Or is this normal?  Do I need to build a rig to agitate the container?

Thanks,
Dave
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 14312
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: Etching is very slow
« Reply #1 on: July 21, 2017, 02:41:15 pm »
I don't use Ferric Chloride but for the etchants I use both heat and agitation speed the process some.
Some use fish tank heaters to keep the solution warm but with more active (faster) etchants the process is over before a preheated solution cools.
For agitation I use a fine air bubbler tube in the bottom of a vertical tank serving two purposes, agitation and removal of bubbles on the copper from the etching process. This bubble removal exposes the copper to faster and more even etching.
Avid Rabid Hobbyist
 

Online sleemanj

  • Super Contributor
  • ***
  • Posts: 2246
  • Country: nz
  • Professional tightwad.
    • The electronics hobby components I sell.
Re: Etching is very slow
« Reply #2 on: July 21, 2017, 03:00:23 pm »
Ferric works best hot, and agitation is necessary for any etching method (rocking, bubbling, pumping, sponging, brushing whatever you like, just got to keep etchant moving over all the surface).  When I used Ferric I would boil a jug and fill a pan with the hot water to make a bath, then put the etching container with ferric into that water bath.

(I don't use Ferric any more)
~~~
EEVBlog Members - get yourself 10% discount off all my electronic components for sale just use the Buy Direct links and use Coupon Code "eevblog" during checkout.  Shipping from New Zealand, international orders welcome :-)
 

Offline matseng

  • Frequent Contributor
  • **
  • Posts: 564
  • Country: se
    • My Github
Re: Etching is very slow
« Reply #3 on: July 21, 2017, 03:26:14 pm »
I've read that people sometimes have problems with the ablated paint re-settling on the board again in a very thin and mostly invisible layer and this causes problems with the etch.

Wiping down the board with IPA after the laser ablation might help to remove the residue.
 

Offline chickenHeadKnob

  • Frequent Contributor
  • **
  • Posts: 722
  • Country: ca
  • doofus programus semi-retiredae
Re: Etching is very slow
« Reply #4 on: July 21, 2017, 04:45:23 pm »
I've read that people sometimes have problems with the ablated paint re-settling on the board again in a very thin and mostly invisible layer and this causes problems with the etch.

Wiping down the board with IPA after the laser ablation might help to remove the residue.

My thought as well. To discern if the ferric chloride is weak clean a small test coupon of PCB with no paint or laser ablation steps and see if it etches promptly. Heat the FeCl and sponge the PCB gently every 10 seconds or so.
 

Offline BurningTantalum

  • Frequent Contributor
  • **
  • Posts: 280
  • Country: au
Re: Etching is very slow
« Reply #5 on: July 21, 2017, 08:45:46 pm »
Many years ago I left a nice job where I had access to a spray etch tank. I then had to resort to etching in a plastic tray like the OP. I found that a solution of Ferric had to be as strong as could be mixed, and I used to etch the PCB upside down as I found that it was quicker and I had fewer 'spots' of unetched copper. The etchant had to be discarded when it became too contaminated- it started to take ages to etch.
Those were the days of 'Daler' pens... I should never have left the job!
BT
 

Offline capt bullshot

  • Super Contributor
  • ***
  • Posts: 1402
  • Country: de
    • Mostly useless stuff, but nice to have: wunderkis.de
Re: Etching is very slow
« Reply #6 on: July 21, 2017, 09:31:57 pm »
The ferric chloride must be heated to at least 40°C to work
Safety devices hinder evolution
 
The following users thanked this post: BrianHG

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 2775
  • Country: ca
Re: Etching is very slow
« Reply #7 on: July 21, 2017, 09:50:50 pm »
At around 70°C, with agitation, it should only take around 4-5 minutes depending on copper thickness, though, there will be some fumes.  At this speed, with some heavy duty rubber gloves, I used to just sit and wave the PCB up and down in the fluid maintaining a consistent flow, checking every minute to verify I wasn't overdoing it.

***Remember, copper flood fill all unused area on your PCB, including the borders and edges, whether it is connected to GND or not is not important.  This will save the life of you ferric chloride chemicals a 5-10 fold (seriously, many PCB just have thin traces and all that extra copper kills the ferric chloride.), this save you money on purchasing more ferric chloride and you also help the environment by not throwing out all of that wasted copper saturated ferric chloride.  I used to use a heavy duty rubber tub with lid for document storage I purchased at office depot and just leave the ferric chloride in there.  Under the tub, I had a weak plug-in radiator heater (I know this is dangerous) which I turned on 1 hour before doing the etch.  The ferric chloride would then be nice and warm and ready to etch multiple boards...
« Last Edit: July 21, 2017, 09:52:36 pm by BrianHG »
__________
BrianHG.
 

Offline BurningTantalum

  • Frequent Contributor
  • **
  • Posts: 280
  • Country: au
Re: Etching is very slow
« Reply #8 on: July 21, 2017, 10:19:46 pm »
The machine that I used in the late 70s/early 80s definitely did not heat the ferric. It consisted of a large plastic welded cabinet with lift up door and two oscillating spray bars. It was large enough to take a PCB of at least 600 x 600 mm.
I never heated my hobby etchant either- I think that heating speeds up the etch time considerably but is not essential for hobby use. The ferric is dangerous enough without having to worry about heating it too!
BT
 

Offline dsharp02

  • Contributor
  • Posts: 26
  • Country: us
Re: Etching is very slow
« Reply #9 on: July 21, 2017, 10:28:02 pm »
Thanks everyone for the advice. I didn't realize that the ferric chloride needed to be heated.  The instructions on the bottle don't mention this.

I will probably place the container on one of my 3d printers, set the heat bed to 45-50C and then "print" a g-code file that moves the bed back and forth to agitate.

Regarding the flood fill.  I did do a copper pour attached to GND so I'm really only removing a tiny bit of the copper.  It also removed the need to route GND connections, so that was nice.





 

Offline chickenHeadKnob

  • Frequent Contributor
  • **
  • Posts: 722
  • Country: ca
  • doofus programus semi-retiredae
Re: Etching is very slow
« Reply #10 on: July 21, 2017, 10:54:33 pm »
Agitation helps but what really works is to take a small  disposable foam brush you can buy cheap at any hardware or paint store and daub the surface. That helps remove the converted copper still clinging to the surface and exposes the fresh layer underneath.
 

Offline uwezi

  • Supporter
  • ****
  • Posts: 271
  • Country: se
    • GreenPhotons
Re: Etching is very slow
« Reply #11 on: July 21, 2017, 11:00:57 pm »
Ferric chloride does not need to be heated!

It should normally etch 1oz copper layers in less than half an hour, but it is a messy business. Sodium persulfate on the other hand needs to be heated.

Have you tried to etch a tiny piece of blank pcb which you did not expose to the paint/ablation process? The only reason that I can see for your problem is an incomplete ablation.
 

Offline dsharp02

  • Contributor
  • Posts: 26
  • Country: us
Re: Etching is very slow
« Reply #12 on: July 21, 2017, 11:16:41 pm »
I actually ran the laser over the copper twice to remove any paint that had resettled (sadly my Legend 24TT doesn't do bottom up raster engraving), and then I used 90% IPA and gently wiped down the board (was afraid of damaging the paint, so I didn't scrub too hard)>

 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 9366
  • Country: us
Re: Etching is very slow
« Reply #13 on: July 21, 2017, 11:27:41 pm »
Thanks everyone for the advice. I didn't realize that the ferric chloride needed to be heated.  The instructions on the bottle don't mention this.

No, it doesn't need to be heated.

Ferric chloride should etch a board in minutes at room temperature, but you do have to agitate it. Where etching is happening you should see darker colored liquid floating away from the etched surface. If etching is too slow, make sure the ferric chloride is the proper strength. Be sure to read the instructions if you are buying the dry crystals and making it up yourself.
I'm not an EE--what am I doing here?
 

Offline alanb

  • Frequent Contributor
  • **
  • Posts: 328
  • Country: gb
Re: Etching is very slow
« Reply #14 on: July 22, 2017, 12:24:10 am »
I would try etching a small sample of cleaned, de-greased board before you have added the paint. It should then be possible to see if the issue is with the etchant or the rest of the process.
 

Offline rdl

  • Super Contributor
  • ***
  • Posts: 2611
  • Country: us
Re: Etching is very slow
« Reply #15 on: July 22, 2017, 12:41:24 am »
Just a few words of caution. I would not allow ferric chloride anywhere near valuable equipment such as 3D printers or anything with exposed metal you value. Be very careful when agitation is used, particularly airstone/bubbler types. The small aerosol particles formed can travel long distances and will corrode everything in sight.
 
The following users thanked this post: tooki, BrianHG

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 2775
  • Country: ca
Re: Etching is very slow
« Reply #16 on: July 22, 2017, 02:26:12 am »
@rdl +1, I corroded the outside of my lead plumbing in my room a few decades ago...
__________
BrianHG.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 9366
  • Country: us
Re: Etching is very slow
« Reply #17 on: July 22, 2017, 03:24:00 am »
By agitation I only mean gentle stirring with a plastic implement, or rolling the liquid around the etching tray. There is no need for anything vigorous like shaking or bubbling.
I'm not an EE--what am I doing here?
 

Offline rdl

  • Super Contributor
  • ***
  • Posts: 2611
  • Country: us
Re: Etching is very slow
« Reply #18 on: July 22, 2017, 04:33:38 am »
As I said, just words of caution. The stuff is damned insidious. In my personal experience the finish on one bathtub faucet was ruined and another kitchen sink fixture nearly so, using nothing but gentle rocking of the tray for agitation. The damage was not immediately obvious and I was surprised it happened because I thought I was being careful. I stopped using it indoors after that, and eventually decided to quit using it completely.
 

Offline jpanhalt

  • Frequent Contributor
  • **
  • Posts: 634
  • Country: us
Re: Etching is very slow
« Reply #19 on: July 22, 2017, 04:38:49 am »
Re. etch speed:   Etching is an oxidative reaction in which the metallic copper is oxidized.  Thus in spray etching, the ferric chloride can be considered saturated with air/oxygen and works faster at lower temperature than bath etching in which not only may the etchant not be saturated with air, but there is a laminar layer on the copper surface.   Gentle rocking helps, but so does heat.   As a rule of thumb, reaction rate doubles for every increase in temperature of 10°C.   I etch in a bath at 40° to 60°C.   It generally take about 15 minutes.   There is an "edge" effect, which means the edges will etch faster than the center of the board, and fine details (remember viscosity and laminar layer) will etch slower.   On occasion, I will use a soft sponge with the etchant to carefully rub over those areas.

Re. your resist:  What wavelength is your laser?  Presumably near IR?  I would consider a dye that absorbed that wavelength in preference to a pigment.   I would also try to find materials that ablated without leaving an ash.  Some "paints" will be worse than others.  Acrylics and cellulose-like lacquers (e.g, clear model dope) are probably better than enamels.  I would not use water-based latex paints.

After ablating, the etchant should wet the areas to be etched and tend to bead up on the painted areas.   If you don't see the wetting, you probably have some residual resist.   The initial etching reaction is very fast.   I use a photoresist.  Sometimes, after development, I may doubt whether more development is necessary.   I very quick dip in or drop of ferric chloride will turn clean copper a salmon color.  If I don't see wetting or the color, I develop a little longer.  As you get your system standardized, that will rarely be an issue, but when starting out, it is a good test and a lot quicker than waiting for hours to find almost no etching.

John
 
The following users thanked this post: dsharp02

Offline Brumby

  • Supporter
  • ****
  • Posts: 8591
  • Country: au
Re: Etching is very slow
« Reply #20 on: July 22, 2017, 11:43:45 am »
Thanks everyone for the advice. I didn't realize that the ferric chloride needed to be heated.  The instructions on the bottle don't mention this.

No, it doesn't need to be heated.

Ferric chloride should etch a board in minutes at room temperature, but you do have to agitate it. Where etching is happening you should see darker colored liquid floating away from the etched surface. If etching is too slow, make sure the ferric chloride is the proper strength. Be sure to read the instructions if you are buying the dry crystals and making it up yourself.

^^^ This has been my experience.

I would suggest you don't need things like bubble tanks and heating for ferric chloride.  Those are necessary for ammonium persulphate etching, but ferric chloride is nasty effective enough without them.

If you have fresh solution made up at the right strength, then at the temperature you are talking about, with some simple agitation - the only impediment I can think of is surface contamination.  12-13 hours should be long enough to etch through a copper pipe.

To test this, I would suggest you take a small piece of PCB and simply do the scrub and clean.  No paint, masking or laser stuff - just a piece of plain board.  Then just drop it into your solution and, with a little bit of agitation, see how long it takes to etch.

If that takes anywhere near 12-13 hours, then your process would seem exonerated - but your ferric chloride would be highly suspect.
 

Offline dsharp02

  • Contributor
  • Posts: 26
  • Country: us
Re: Etching is very slow
« Reply #21 on: July 23, 2017, 01:13:27 pm »
Quote
Re. etch speed:   Etching is an oxidative reaction in which the metallic copper is oxidized.  Thus in spray etching, the ferric chloride can be considered saturated with air/oxygen and works faster at lower temperature than bath etching in which not only may the etchant not be saturated with air, but there is a laminar layer on the copper surface.   Gentle rocking helps, but so does heat.   As a rule of thumb, reaction rate doubles for every increase in temperature of 10°C.   I etch in a bath at 40° to 60°C.   It generally take about 15 minutes.   There is an "edge" effect, which means the edges will etch faster than the center of the board, and fine details (remember viscosity and laminar layer) will etch slower.   On occasion, I will use a soft sponge with the etchant to carefully rub over those areas.

I got a foam brush instead of a sponge, and that seems to help a bit.  The etchant is 42 baume, whatever that means. I finished one board, but I there were some broken tracks (from undercutting during the long bath overnight?).

Quote
Re. your resist:  What wavelength is your laser?  Presumably near IR?  I would consider a dye that absorbed that wavelength in preference to a pigment.   I would also try to find materials that ablated without leaving an ash.  Some "paints" will be worse than others.  Acrylics and cellulose-like lacquers (e.g, clear model dope) are probably better than enamels.  I would not use water-based latex paints.

I think it's 10600 nm.  It's a CO2 laser.

Quote
After ablating, the etchant should wet the areas to be etched and tend to bead up on the painted areas.   If you don't see the wetting, you probably have some residual resist.   The initial etching reaction is very fast.   I use a photoresist.  Sometimes, after development, I may doubt whether more development is necessary.   I very quick dip in or drop of ferric chloride will turn clean copper a salmon color.  If I don't see wetting or the color, I develop a little longer.  As you get your system standardized, that will rarely be an issue, but when starting out, it is a good test and a lot quicker than waiting for hours to find almost no etching.

Thanks for your help.  I may try the photoresist method at some point.

Dave
 

Offline tronde

  • Frequent Contributor
  • **
  • Posts: 310
  • Country: no
Re: Etching is very slow
« Reply #22 on: July 23, 2017, 02:09:58 pm »
Have you got ferric (III) chloride (brownish-yellowish) or ferric (II) chloride (greenish)?

Ferric (III) chloride does not need much heat or agitation. I remember from very long time ago I got ferric (II) chloride. Was extremely slow. Don't remember if it ever managed to finish the etching.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 9366
  • Country: us
Re: Etching is very slow
« Reply #23 on: July 23, 2017, 02:13:05 pm »
The etchant is 42 baume, whatever that means.

I think it means you bought the product ready mixed as a solution in a bottle. Is that correct? Also, we presume the product was sold specifically for etching copper?
I'm not an EE--what am I doing here?
 

Offline orin

  • Frequent Contributor
  • **
  • Posts: 425
  • Country: us
Re: Etching is very slow
« Reply #24 on: July 23, 2017, 02:25:28 pm »
Many years ago I left a nice job where I had access to a spray etch tank. I then had to resort to etching in a plastic tray like the OP. I found that a solution of Ferric had to be as strong as could be mixed, and I used to etch the PCB upside down as I found that it was quicker and I had fewer 'spots' of unetched copper. The etchant had to be discarded when it became too contaminated- it started to take ages to etch.
Those were the days of 'Daler' pens... I should never have left the job!
BT


I use the pre-mixed Ferric Chloride solution from MG Chemicals in a photographic tray, float the tray in a sink of hot water and rock it back and forth.  Takes 15 to 20 minutes.  When it's done, I pour as much as possible of the FeCl back in the bottle.

(To prepare the boards, I use the MG Chemicals positive resist boards, laser or inkjet printer on transparencies and a single UV tube about 6 inches over the board for their recommended exposure time, then their developer diluted as recommended.)

If the etching (or developing come to that) gets slow, it's time to replace the solutions.

So, if etching is taking longer than 30 minutes at room temperature, there is something really wrong...
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf