Author Topic: PCB etching, photoresist method questions  (Read 18735 times)

0 Members and 1 Guest are viewing this topic.

Offline ArylTopic starter

  • Newbie
  • Posts: 6
PCB etching, photoresist method questions
« on: January 18, 2015, 04:21:13 am »
Made my first two attempts at using the photo method today, using a UV salon nail paint curing box as the light source.

The first one turned out looking pretty good but unusable because some 16mil traces got eaten. I used two layers of transparency sheets as a mask, the sheets were printed out by a local print shop as dark as their machines would go.

The second one I used three layers of transparency sheets, the board is usable but has some extreme pinholes compared to the first. No broken traces.

I'm having some confusion as to why the second one had so much more pinholes.

The first had two sheets, was exposed in 5 evenly spaced increments of 30 seconds each (It was supposed to be a test to figure out a good exposure time). When placed in the developer it was almost immediately developed, I only kept it in there for ten seconds or so.

The second one I exposed for 20 seconds (The entire board) and took several minutes in the developer.

The only explanation I can think of is that because I exposed it for less time and had to keep it in the developer for much longer the pinholes had time to completely develop with the rest of the board.

Same transparencies used, same alignment on the first two, I just added a third.

Pictures should be attached.

I've been scouring the internet for helpful information, I thought it would be worth asking how successful people on here have been and if anyone has any tips on avoiding all this pinholing, I can't find any better laser printers to use.
 

Offline sleemanj

  • Super Contributor
  • ***
  • Posts: 3020
  • Country: nz
  • Professional tightwad.
    • The electronics hobby components I sell.
Re: PCB etching, photoresist method questions
« Reply #1 on: January 18, 2015, 07:36:40 am »
Pre-sensitised, dry film, spray on, paint on, positive, negative... ie, what type of resist?

What type of developer?

~~~
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 ArylTopic starter

  • Newbie
  • Posts: 6
Re: PCB etching, photoresist method questions
« Reply #2 on: January 18, 2015, 07:58:01 am »
Presensitized positive MG Chemicals boards, MG chemicals 418 positive developer. According to the MSDS it's just sodium hydroxide.

I'm thinking this is just over etching as the board looked flawless after developing, but I only kept it in the etchant (Ferric Chloride) until the copper on the inside of the circle was dissolved. It was agitated the entire time, I was constantly brushing it with a plastic bristled paintbrush while it was submerged.
 

Offline sleemanj

  • Super Contributor
  • ***
  • Posts: 3020
  • Country: nz
  • Professional tightwad.
    • The electronics hobby components I sell.
Re: PCB etching, photoresist method questions
« Reply #3 on: January 18, 2015, 08:23:59 am »
Presensitized positive MG Chemicals boards, MG chemicals 418 positive developer.

This thread may help:
https://www.eevblog.com/forum/beginners/my-first-try-at-making-a-pcb/
~~~
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 :-)
 

Online wraper

  • Supporter
  • ****
  • Posts: 16798
  • Country: lv
Re: PCB etching, photoresist method questions
« Reply #4 on: January 18, 2015, 03:13:49 pm »
Presensitized positive MG Chemicals boards, MG chemicals 418 positive developer. According to the MSDS it's just sodium hydroxide.

I'm thinking this is just over etching as the board looked flawless after developing, but I only kept it in the etchant (Ferric Chloride) until the copper on the inside of the circle was dissolved. It was agitated the entire time, I was constantly brushing it with a plastic bristled paintbrush while it was submerged.
Very typical result. To develop with NaCl you need photomask with very dense blacks and precise exposure time. If blacks are not very dense, right exposure time becomes very tricky. Also you need VERY tight developer concentration and temperature control. Better use Negative photoresist or use silicate based developer for positive photoresist.
Read developing part here: http://www.electricstuff.co.uk/pcbs.html
 

Offline homebrew

  • Frequent Contributor
  • **
  • Posts: 293
  • Country: ch
Re: PCB etching, photoresist method questions
« Reply #5 on: January 18, 2015, 06:48:00 pm »
Just my 2 cents  as a hobbyist ...

I did a lot of boards myself and for some time I had access to a professional UV exposure unit. I always ended up with results like yours, because I couldn't produce "films" that were "black" enough in the UV spectrum.

What I ended up with was a 500W Halogen Flood lamp, mounted 40cm above the PCB. And I used two transparencies on top of each other, both ink-side down and a thin glass on top of it to hold everything down ...

Yeah, this setup is very unprofessional in every aspect: Wrong peak wavelength, a (crazy) lot of heat, quite long exposure times (8-10min) etc. etc. but it always gave me near perfect results ...

I used standard Bungard FR4 material and NaOH as a developing agent.

But anyway, I completely stopped etching my own boards because of the hazards (got a small splash Fe3Cl in one eye - luckily just painful but without injury), the problematic heavy metal waste and all the obstacles of doing proper two-layer layouts with alignment, drilling and vias, solder-mask ...

You can now get a professionally made PCB (100x160) for around 50 EUR (or even cheaper) ...  but yes, it takes a week or two ...

Happy etching!
« Last Edit: January 18, 2015, 07:02:13 pm by homebrew »
 

Offline ArylTopic starter

  • Newbie
  • Posts: 6
Re: PCB etching, photoresist method questions
« Reply #6 on: January 19, 2015, 03:44:12 am »
Thanks for the advice everyone, found my problem, it was over etching due to not agitating the board enough. It was just eating through the resist.

This time I bought some foam brushes, scrubbed at the area to be removed for the entire time it took to etch, focusing on the area around the traces first. I also heated up the etchant with boiling water and changed the exposure time to one minute.

It's as perfect as I could have ever hoped for. A handful of pinholes in the ground plane, but that doesn't matter. The traces are perfect, the logo I designed came out better than expected...this is really cool. It's a huge leap from messes of wires on a breadboard.
 

Tac Eht Xilef

  • Guest
Re: PCB etching, photoresist method questions
« Reply #7 on: January 19, 2015, 05:09:40 am »
Thanks for the advice everyone, found my problem, it was over etching due to not agitating the board enough. It was just eating through the resist.

This time I bought some foam brushes, scrubbed at the area to be removed for the entire time it took to etch, focusing on the area around the traces first.

That's not normal, particularly with MG Chemicals board. In my experience it's damn near impossible to overexpose or over-develop (though I think that's exactly what's happened), and I've never seen the resist etched through. I might believe it if it were something like Kinsten, which (again, IME) is softer and easier to overexpose/over-develop.

Your first couple of results look like what you get from using inkjet-printed transparencies or a laser that really needs its drum replaced, coupled with overexposure. And while scrubbing certainly speeds up etching, I've never needed to do it with MG Chemicals board; gently rocking the tray is enough.

I think you've stumbled on a way of balancing out the problems, rather than something reliable. My guess is the real problems are (a) poor quality mask, and (b) gross overexposure from way too high light intensity at the board surface.

Does your UV curer use lamps or LEDs? What sort of printer was used for the transparencies? Did you use glass or something to sandwich the transparency printed side down to the board?

I'd try (a) getting a decent laser print on good tracing paper, and (b) increasing the distance between light source and board - a good 2"-3" inches at least.

FWIW, I use a single layer of laser printed transparency paper (Brother HL-2270, Sihl 'Utoplex' tracing paper). 12-16 thou is easy with MG board, and 8-10 thou is reliably possible with a bit of care.
 

Offline kolbep

  • Frequent Contributor
  • **
  • Posts: 598
  • Country: za
    • ShoutingElectronics.com
Re: PCB etching, photoresist method questions
« Reply #8 on: January 19, 2015, 05:21:19 am »
Well done.
Now, the question we all really want to know,
What is it????
====================================
www.ShoutingElectronics.com Don't just talk about Electronics, SHOUT ABOUT IT! Electronics Blog Site and Youtube Channel
 

Offline ArylTopic starter

  • Newbie
  • Posts: 6
Re: PCB etching, photoresist method questions
« Reply #9 on: January 19, 2015, 06:43:05 am »
I'll try moving the lightsource further away, it's about an inch or so away, less along the edges. The bulbs are fluorescent, two at the top and two angled on the sides, I'm using this, it has 4 9-watt bulbs:

http://www.amazon.com/Salon-Edge-Acrylic-Shellac-Curing/dp/B006QO4BRM

the board is sandwiched to the transparencies with glass, I'm using a 5 x 7 picture frame with a spring loaded cardboard back so it keeps them sandwiched pretty well. The transparencies are pretty good, with three stacked I can only see one small speck of light on the ground plane when held up to a bright light source. They were printed by a local printshop on laser printers.

The board looks perfect after developing, it's only near the end of etching that the pinhoels/pitting happens.


Well done.
Now, the question we all really want to know,
What is it????

Hah, get ready for a bit of a read.

I designed and am trying to make a new instrument cluster for my car (1986 Saab 900), making all the gauges and everything myself so it fits in the stock housing.

This board is the front of the tachometer, the outer ring holds 24 WS2812B RGB LEDs (Adafruit Neopixels), the inner has 16 for either a clock or boost/vac gauge, have not decided yet. The other pads near the leds are for 0805 decoupling capacitors.

Each gauge was designed to have a small AVR powering it so I can change their function whenever I wish, all the gauges communicate via i2c to a "master" atmega328 hooked to a 20x4 LCD screen so I can have exact readings from sensors.

Each gauge is made of a "stack" that plugs into eachother, so it's somewhat modular and if I want to change something I don't have to redesign the whole thing, just one of the three plates. First plate is the display portion, second is the AVR, oscillator, ISP header and the third is the input.

For the tachometer the input plate consists of an R/C filter, optocoupler, schmitt trigger and a few other components to convert the 12v NOISY square wave into a clean 5v signal to trigger an interrupt on the AVR.

I figured it would be a good first real project. Not dead simple, but complicated enough that it's still a challenge. I've had to learn quite a bit to get it all to work, totally worth it.
« Last Edit: January 19, 2015, 06:46:19 am by Aryl »
 

Offline Fred27

  • Supporter
  • ****
  • Posts: 726
  • Country: gb
    • Fred's blog
Re: PCB etching, photoresist method questions
« Reply #10 on: January 19, 2015, 10:25:29 am »
I'd definitely recommend a silicate based developer. I also use the same sort of nail lamp. I removed the two side bulbs to reduce the risk of light getting under the transparency and it still work well.

I get great result using a Canon MG5250 inkjet printer for the transparencies in T-shirt printing mode. It gives really thick coverage and also automatically mirrors it.

Another recommendation - Ziplock vacuum food bags to hold the transparency down tight against the PCB.

http://0xfred.wordpress.com/2014/02/28/further-pcb-improvements/
 

Offline kolbep

  • Frequent Contributor
  • **
  • Posts: 598
  • Country: za
    • ShoutingElectronics.com
Re: PCB etching, photoresist method questions
« Reply #11 on: January 19, 2015, 12:43:20 pm »
Thanks for the info.
It sounds like a great project  :clap:
====================================
www.ShoutingElectronics.com Don't just talk about Electronics, SHOUT ABOUT IT! Electronics Blog Site and Youtube Channel
 

Offline poorchava

  • Super Contributor
  • ***
  • Posts: 1672
  • Country: pl
  • Troll Cave Electronics!
Re: PCB etching, photoresist method questions
« Reply #12 on: January 20, 2015, 06:59:45 am »
Dry film is also nice, but can be a bitch to apply so that there are no air bubbles trapped underneath.

You can enchance your laserjetted exposure masks by putting a teaspoon of acetone or a nitro thinner into a jar, hanging the mask on a string, closing the lid and heating the jar A LITTLE BIT (so that it feels warm to the touch). Solvent vapors will be absorbed by plastic particles in the toner making them sticky, but also swell a little which causes the black to become almost completly opaque.
I love the smell of FR4 in the morning!
 

Online mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 13695
  • Country: gb
    • Mike's Electric Stuff
Re: PCB etching, photoresist method questions
« Reply #13 on: January 20, 2015, 11:06:41 am »
+1 for silicate developer

Use thick tracing paper, not transparency film.
Youtube channel:Taking wierd stuff apart. Very apart.
Mike's Electric Stuff: High voltage, vintage electronics etc.
Day Job: Mostly LEDs
 

Offline poorchava

  • Super Contributor
  • ***
  • Posts: 1672
  • Country: pl
  • Troll Cave Electronics!
Re: PCB etching, photoresist method questions
« Reply #14 on: January 20, 2015, 11:15:47 am »
Yep, transparency likes to shrink a little bit form the heat when fusing the toner. That amount of shrinkage is not a problem for slideshows and such, but can be a killer when doing PCBs, especially if you're using finer pitch stuff (like 0.5mm QFPs and SOPs)
I love the smell of FR4 in the morning!
 

Offline ElektroQuark

  • Supporter
  • ****
  • Posts: 1244
  • Country: es
    • ElektroQuark
Re: PCB etching, photoresist method questions
« Reply #15 on: January 20, 2015, 11:16:34 am »
CAS number for the silicate developer?
Maybe I will try it.

Offline Solder_Junkie

  • Frequent Contributor
  • **
  • Posts: 318
  • Country: gb
Re: PCB etching, photoresist method questions
« Reply #16 on: January 20, 2015, 12:05:12 pm »
Using direct laser print transfer is much easier than using UV light boxes, I've made several boards in recent months using shiny paper (eBay item 261640812271) which costs less than 12 GB pounds for 100 A4 sheets delivered, and an ordinary clothes iron to transfer the toner. My laser printer is a Samsung M2020, which was one of the cheapest I could find. Laser toner is very resistant to warm ferric chloride, acetone removes it afterwards. Don't forget to print your tracks as a mirror image and to print on the shiny side of the paper...

For a quick try, you can print onto glossy inkjet photo paper with a laser printer. You will find it needs soaking off in water then quite a bit of rubbing with a finger under a running tap to remove all the paper residue. With the special paper, it just peels off and needs no washing. However, I have used glossy photo paper with perfect results on SMD boards down to 0805 sized parts.

Double sided boards with tracks on both sides are possible, although for the most part I use SMD boards which are single sided or double with just an earth plane and a handful of wire links through for ground connections.
 

Offline madires

  • Super Contributor
  • ***
  • Posts: 7695
  • Country: de
  • A qualified hobbyist ;)
Re: PCB etching, photoresist method questions
« Reply #17 on: January 20, 2015, 12:23:59 pm »
I did a lot of boards myself and for some time I had access to a professional UV exposure unit. I always ended up with results like yours, because I couldn't produce "films" that were "black" enough in the UV spectrum.

Take tracing paper (90-95 g/m²), print layout with a laser printer (disable toner save mode) and use something like CRC Solvent 50 (less expensive than special toner sprays) to get a nice solid black.
 

Offline ArylTopic starter

  • Newbie
  • Posts: 6
Re: PCB etching, photoresist method questions
« Reply #18 on: January 21, 2015, 12:50:13 am »
I'll try some tracing paper for the next board. Surface mount soldering sure is a lot easier than I thought it would be, probably helps I ditched my old firestarter iron for a new hakko 591. This has got to be the nicest tool I've ever owned.

First try, and it all works  :D
 

Tac Eht Xilef

  • Guest
Re: PCB etching, photoresist method questions
« Reply #19 on: January 21, 2015, 12:52:43 am »
Using direct laser print transfer is much easier than using UV light boxes

Each to their own, but I'd disagree. I use both, but with UV & presensitised board it's (usually) easier to get the process right initially (& it's more consistent when you do), no need to retouch tracks when the toner fails to stick, no scrubbing to clean the board beforehand or remove the resist afterwards, and I can go from hitting "Print" to having a ready-to-solder board in my hand in 10 minutes.

The biggest downside is cost - decent +ve resist board costs ~50% more than blank PCB.
 

Offline ElektroQuark

  • Supporter
  • ****
  • Posts: 1244
  • Country: es
    • ElektroQuark
Re: PCB etching, photoresist method questions
« Reply #20 on: January 26, 2015, 09:28:26 pm »
CAS number for the silicate developer?
Maybe I will try it.


Anyone?

Offline akis

  • Frequent Contributor
  • **
  • Posts: 981
  • Country: gb
Re: PCB etching, photoresist method questions
« Reply #21 on: January 26, 2015, 09:56:12 pm »
I use tracing paper, over 100gsm, like this http://www.amazon.co.uk/gp/product/B009LHXTX8?psc=1&redirect=true&ref_=oh_aui_detailpage_o00_s00

It does not crease too much in the laser printer, and most of the times the toner sticks to it fine.

Large areas of black are a problem. I have tried placing two transparencies on the UV box, but aligning them is very hard.

You must experiment with your UV box and the proper exposure time for the type of transparency, the type of the board, the UV box etc. Mine is like 7 mins.

A very important factor is the UV box temperature. As it warms up it gets stronger so for example when I say "7 mins" I mean with pre-warmed up UV box (2-3 mins at least and immediately before exposure).

Doing two sided is the same issue - if you get one side different to the other, it's probably the UV box being warmer on the second side!

Development is extremely temperature sensitive, I use a warm water bath to keep the developer at a constant 24C-25C and depending on its strength it takes about 45-90 seconds with vigorous agitation throughout.

I etch using ammonium persulphate, so I can splash it everywhere (not intentionally) without worries. It will not eat through the resist. It is quite slow though, and must be kept at 55C (so I use a hot water bath for it). But it is safe and etches really well.

 

Offline Michal_S

  • Contributor
  • Posts: 11
Re: PCB etching, photoresist method questions
« Reply #22 on: January 26, 2015, 10:44:56 pm »
I use SENO 4007 developer (don't know it contents, but it doesn't contain NaOH). Used NaOH earlier, and it was extremely difficult to get it right, even using laboratory scale, for example it often was to strong and ruined the board immediately. With SENO 4007 I get consistent results with no worries.
 

Offline george graves

  • Super Contributor
  • ***
  • Posts: 1257
  • Country: us
Re: PCB etching, photoresist method questions
« Reply #23 on: January 26, 2015, 11:16:52 pm »
Yep, transparency likes to shrink a little bit form the heat when fusing the toner.

I get a small amount of shrinkage with the toner transfer method.  Not sure at what point it's happening at.  But a 10 pin header will be ever so slightly too small.  Someday I'll figure it out and compensate for it.

Online tautech

  • Super Contributor
  • ***
  • Posts: 28142
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: PCB etching, photoresist method questions
« Reply #24 on: January 26, 2015, 11:42:53 pm »
Yep, transparency likes to shrink a little bit form the heat when fusing the toner.

I get a small amount of shrinkage with the toner transfer method.  Not sure at what point it's happening at.  But a 10 pin header will be ever so slightly too small.  Someday I'll figure it out and compensate for it.
Your paper is shrinking when passing through your printer.

Print to plain paper first, then stick your cut to size tranfer paper over the image and print again.
That should fix or reduce the problem for all but big PCB's.
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf