Polarizing Film Question

Go To Last Post
30 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

When etching circuit boards at home. I notice there is a small amount of bleed with the UV light and the mask.

I was thinking this could be sorted out by straightening the light, with a polarizing film. But I've never used any polarizing films before, and I'm just not sure what degree of film to use, or how many.

 

0 and 90 degree films seem to be the most popular, but if I'm correct, rotating the film will change the degree? I guess they are just labeled like that because they are for LCD screens, so you cant really rotate them.

 

Does anyone know what would be the best way would be, to only let light through that is going directly towards the PCB?

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Do polarizers pass uv?

#1 This forum helps those that help themselves

#2 All grounds are not created equal

#3 How have you proved that your chip is running at xxMHz?

#4 "If you think you need floating point to solve the problem then you don't understand the problem. If you really do need floating point then you have a problem you do not understand." - Heater's ex-boss

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yup. I have a DLP 3d printer, and I noticed there is a polarizer under the bed, between the LCD screen. That's what actually gave me the idea.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Light waves from incandescent and fluorescent light sources, as well as sky light, are randomly polarized. 

 

A polarizer is INTENDED to pass light waves which are polarized in a certain direction and reject others. A polarizer should have rather little effect on edge bleed. It MIGHT improve things on edges that are at one specific angle with respect to the  polarization plane passed by the polarizer, but, on a circuit board, there are edges in all sorts of  orientations. 

 

Some of that "bleed" is due to diffraction and some is due to the non-parallelism of the light rays. Non-parallelism happens because the light source has a large area compared to the distance between the mask and the light source. Lenses can improve this - it is called collimate. Another solution to this is to put a mask with a near-pinhole opening in front of the light source, but this greatly increases the time needed for exposure. lens Diffraction can be avoided only by reducing the distance between the mask and the photosensitive surface.

 

There is a Freaks member, Ross, who makes circuit boards commercially. He may be able to help more.

 

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

Last Edited: Wed. Dec 19, 2018 - 09:36 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

LittleRain wrote:
When etching circuit boards at home.

Why, they are so cheap now from commercial sources and so much better then you can make!

 

Click Link: Get Free Stock: Retire early! PM for strategy

share.robinhood.com/jamesc3274
https://www.onegold.com/join/7134f67c2b814c5ca8144a458eccfd61

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:

on a circuit board, there are edges in all sorts of  orientations. 

 

I'm not sure I understand you on this one. Cause I'm sure you know, but I'm only dealing with the top plane, and I just want to allow the light that is perpendicular to the plane to be able to pass through.

 

 

Seems like you do know your stuff though!

 

But I still think it should help quite a bit, at least with my specific setup.

I'm not 100% sure, but it seems like the bleed is caused from having to double up the masks, as well as from the thickness of the transparency film.

If I don't double up the mask, I get light bleed everywhere, not just on the edges.

I do always make sure that the toner is on the board's side, but its still not enough to stop it, and so I need the second layer.

But because there is 2 layers, one of the masks is farther from the board, so when the light is coming in from an angle, it manages to pass right beside the top mask, and will hit the bottom one's edge. This is why (I believe) I get some edge bleed.

 

Here is an illustration of what I think is going on.

 

I'll admit that the bleed isn't really that bad at all, I'm still able to make 0.2mm traces, with 0.2mm spacing.

But when I'm not consistent with how I hold and move the light, it can cause some traces to have sections that don't etch.

Building something like a curing box would solve my problem, but I think this would be a good quick fix, and probably help the box as well.

 

 

ki0bk wrote:

LittleRain wrote:
When etching circuit boards at home.

Why, they are so cheap now from commercial sources and so much better then you can make!

 

 

Cause I like to be able to have a prototype by the end of the day.

And most of the time, something is really wrong on my first board, so it really helps me make sure I'm sending a working design to a fab house.

It would really suck to spend $50, wait a week or 2 and find out I made a huge mistake.

 

I actually just made a board 3 days ago, unfortunately one of the parts I made had its mounting holes in the wrong spot(Its hard to measure conductive rubber pads with a caliper), so I fixed the mistake and remade it last night.

 

Also, I'm getting really good at making boards. The only thing I really need to improve on is solder mask.

I'm going to design a little CNC machine that squeegees it on, then I should be able to make boards as good as a fab house.

Last Edited: Wed. Dec 19, 2018 - 10:31 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Stacking masks will kill you. The mask closest to the photoresist will be a lot better than a mask farther from the photoresist. 

 

Do you really need to stack?  0.2mm is getting pretty fine, just under .008" (8 mil). If you are doing that sort of work and expecting a useable yield, then I would shift to a single mask.

 

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 it is called collimate

Agreed.

 

Although the wavelength is a little bit different, the concept is the same with taking medical X-rays and getting a sharp image.

 

These days X-ray machines all use a Bucky-Potter grid to clean up the image being produced from the full range of "polarities" of the X-rays generated by the tube, and further rotated / deflected by their interaction with the tissues being X-rayed.

 

Know that the Bucky plate moves during the exposure, otherwise one gets banding / hot spots.

 

JC 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I don't see why a polarized filter would help.

 

But some questions :

 

How do you make the film, for me it sounds like is not "really" black.

 

What is your light source ? (How big)  

 

for test of "really" black try to over expose 5 or 10 times, does everything behave as exposed?

with bad films (I have been there 30+ years ago) perhaps under expose a tad, or use a black touch to add color to the thin lines. (for nice films we used some thin tape that was made for that 

you could make bends with it). And for chips we had transfer paper (not sure if it's the correct name) with the correct footprint (and for dip's it made the hole clean so it was easy to drill).  

Last Edited: Wed. Dec 19, 2018 - 11:10 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You are using toner methods to expose your board?  How good of a resolution do you expect?  I am wondering, since I've never tried it, but would think there is some fuzzy limit to edge definition using toner.

 

You could always just make a board to "take a look" for for & layout, then toss it in the trash & order your real, plated-through boards--nice and cheap .  Try to use 8 mil traces only when absolutely necessary!

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

As Jim said what you are looking for is collimation.

 

Look at think and tinks DIY box

 

https://www.thinktink.com/stack/...

 

I built something along those lines - works great.

 

You still need a good phototool and a vacuum table to make sure it is in perfect contact.

 

For a cheap way into phototools - use an inkjet printer on multilayer OHT films.  Lasers suck for this unless you have a small car priced laser.  The inkjets on the correct film do fantastic.  Pigment black is the safest bet for good UV opacity.

 

KiwoJet is the film I use - but others brands are fine as long as they are not the crap gelatin coat style.

 

Also get a stoufer guage so you can get the exposure perfect.

 

Next down the list is a spray etcher if you are keen.

 

Going up to plated through holes is a major step and probably not for most people.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:
There is a Freaks member, Ross, who makes circuit boards commercially. He may be able to help more.
Well Jim, some of that statement is correct. cheeky

 

Yes, I make pcbs here at my place in support of commercial work, but to express it in your shorter form could be misleading to the reader. I do proof of concept prototype designs and extremely short production runs (largest was 8 evaluation units for the Canadian Government). I always make the first unit here before sending the (sometimes modified) design off to one of the quick turn Chinese pcb houses. This way I have a working unit to check mechanical aspects (clearances, ease of assembly, etc) and to start software. It is very useful to be able to photograph an assembled prototype within a couple of days to show the client and obtain feedback. I have also done a "show and tell" via Skype.

 

Now back to the topic. I produce a reversed life sized pdf image which I then take to my favourite printing centre for a transparency. They keep their machines in great condition at no extra cost to me. Like the OP, I then place the toner side against the presensitised pcb and cover that with a sheet of glass and place the combination in a frame to ensure a flat even contact. I have a bank of UV leds which is suspended at about 50mm above the glass/film/pcb assembly. The bank of leds is always larger in area than any pcb that I am making. I have two sheets of baking paper in front of the leds to diffuse the uv light so that the pinpoint light source effect is removed. I control the exposure time based upon experience to prevent the undercutting that the OP is witnessing. The finest I have needed/produced is 10mil traces.

 

I can have a pcb etched in around an hour from finishing the pdf file, including driving to the printing centre. Assembly of course is then subject to parts availability and drilling, and soldering. I use a chemical dip to tin the copper traces. No solder mask or component overlay at this prototype stage.

 

Hope that helps.

 

Ross

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Presensitised PCBs would probably mean positive photoresist (Kinsten).  Positive photoresist although convenient is not very high performance and is easy to scratch.  If you want to break the 8/8 thou line/space barrier go for negative resist.

 

A cheap pigment inkjet will get equal or better performance for this task than the giant clean print centre laser.  Orifice Works in Oz have Epson Workforce pigment printers under $100.  Other parts of world are probably cheaper.  So if you are doing them often or you are not close to a print centre with a well maintained commercial class laser printer, that is an option.  Home/SMB class lasers - the kind that cost $50 to $4000 are not as good as a $100 inkjet for this task.

 

Unless you have deep pockets then LEDs are not the best option for a UV light source.  Most cheap LEDs are 405nm.  UV Resist is optimized for the Mercury "I-Line" of 365nm.  365nm LEDs are much dearer.  Also to get best results you need to have enough power to do the primary polymerization in around 1 minute.  Slower is fuzzier.  Black light bulbs (Disco lights / Bug Zapper lights) are a cheap source of 365nm light.

 

Cheap 405nm LEDs will work.  They are not optimal.

 

Using a sheet of baking paper as a diffuser is the exact opposite of collimation (what the OP mistakenly thought was the polarization he needed).  It will make more light at an higher incident angle to get under the traces.  Again if you are aiming for 10/10, 12/12 or larger you can get away with it no worries.  If you are aiming at under 8/8 thou try to make your light as parallel as possible.  Sunlight is very parallel :)  It is what I used when I made my first UV PCB 1/4 a century ago.  Back then I probably did 20 though traces than necked down to 16 to get between DIP IC legs.  These days I need to make PCBs for things like FPC connectors that have 0.3mm pitch so need 6/6 though trace/space.

 

A sheet of glass for the contact can work fine.  Especially if you are making a small PCB and there is less risk of things being warped causing issues.  Best practice is a sheet of cheap (not UV stable) plastic with a vacuum pump pulling it into contact with the PCB phototool sandwich.

 

The other thing to take into account.  I am saying "If you want reliable 8/8 or under rules" you need all this stuff.  When I say reliable I mean a whole PCB with 8/8 all over it and doing that every time.  If you have a PCB that just has some parts that are fine and the rest of it is 16/16 - then you can just pay attention to the important part to make sure it is OK.  Or if it is a tiny PCB that is <8/8 then just make 16 copies of it on the one blank and one of them will turn out fine even with LEDs and average phototools.

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The polarizers you mention are the same, but just cut in different orientations out of the same base material.

Watch some youtube vid's about polarisers and polarized light.

 

But it won't get you anywhere.

As mentioned beore 2 printed masks won't work well. You need the ink/toner side flat against your PCB to get sharp edges.

 

You can also create more distance between your (point) light source and your PCB to get sharper shadow edges.

 

As an alternative you can use a small CNC machine and an UV laser to write directly into your uncured solder mask.

Marco Reps has been experimenting with this:

https://www.youtube.com/watch?v=gm5P74vcB84

Doing magic with a USD 7 Logic Analyser: https://www.avrfreaks.net/comment/2421756#comment-2421756

Bunch of old projects with AVR's: http://www.hoevendesign.com

Last Edited: Thu. Dec 20, 2018 - 09:13 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Big clive has a bunch of youtube videos describing his tests with home PCB etching. Toner masks don't give good results. You might get better results with inkjet dye or pigment on transparency:

 

https://www.youtube.com/watch?v=...

 

And his tests for different transparency material:

 

https://www.youtube.com/watch?v=...

 

- S

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I just remembered one ting. 

 

If you print isn't black enough on a plastic sheet (and there are also many different types) try on paper, and then give the paper some oil (like WD40) then it will work as a film. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The second video is the type of tranny material I recommend.  It is far superior to the gelatine based ones.

 

mnehpets wrote:

Big clive has a bunch of youtube videos describing his tests with home PCB etching. Toner masks don't give good results. You might get better results with inkjet dye or pigment on transparency:

 

https://www.youtube.com/watch?v=...

 

And his tests for different transparency material:

 

https://www.youtube.com/watch?v=...

 

- S

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for all the replies guys. I will definitely try everything that was advised.

Please excuse any shotty writing, it was a lot to reply to!

 

ka7ehk wrote:

Stacking masks will kill you. The mask closest to the photoresist will be a lot better than a mask farther from the photoresist. 

 

The problem is 1 mask will partly cure the traces as well.

 

sparrow2 wrote:

I don't see why a polarized filter would help.

 

But some questions :

 

How do you make the film, for me it sounds like is not "really" black.

 

What is your light source ? (How big)  

 

for test of "really" black try to over expose 5 or 10 times, does everything behave as exposed?

with bad films (I have been there 30+ years ago) perhaps under expose a tad, or use a black touch to add color to the thin lines. (for nice films we used some thin tape that was made for that 

you could make bends with it). And for chips we had transfer paper (not sure if it's the correct name) with the correct footprint (and for dip's it made the hole clean so it was easy to drill).  

 

I use a 2400dpi printer, and yeah the real problem is that its not as black as I need it to be.

 

Not really sure how to describe my light, but its this exact one.

I have plans to make a curing box, already have all the LEDs, but I just haven't found the time to get around to it.

 

avrcandies wrote:

You are using toner methods to expose your board?  How good of a resolution do you expect?  I am wondering, since I've never tried it, but would think there is some fuzzy limit to edge definition using toner.

 

You could always just make a board to "take a look" for for & layout, then toss it in the trash & order your real, plated-through boards--nice and cheap .  Try to use 8 mil traces only when absolutely necessary!

 

I'm using photo-resist, but the masks I make to cure the photo-resist are just transparent sheets with toner.

Its not really about resolution, its about the consistency. A lot of the time my whole board will be almost perfect, but there will be one or two traces that get over cured. Then I have to use either an exact-o knife, or a CNC router to fix it.

The toner methods never worked well for me, it would always leave residue from either the paper, or possible the stuff that's mixed with the toner while printing, on the board. Everywhere that residue was, would not etch, which was literally everywhere.

 

I was using tiny rivets for my vias, but I just recently got everything I need to electroplate them.

Literally all I need to do (after I get the consistency down) is make something that will spread solder mask evenly, and then I should be able to make near fab house quality boards.

 

andrewm1973 wrote:

As Jim said what you are looking for is collimation.

 

Look at think and tinks DIY box

 

https://www.thinktink.com/stack/...

 

I built something along those lines - works great.

 

You still need a good phototool and a vacuum table to make sure it is in perfect contact.

 

For a cheap way into phototools - use an inkjet printer on multilayer OHT films.  Lasers suck for this unless you have a small car priced laser.  The inkjets on the correct film do fantastic.  Pigment black is the safest bet for good UV opacity.

 

KiwoJet is the film I use - but others brands are fine as long as they are not the crap gelatin coat style.

 

Also get a stoufer guage so you can get the exposure perfect.

 

Next down the list is a spray etcher if you are keen.

 

Going up to plated through holes is a major step and probably not for most people.

 

Sounds expensive haha. But anyway, I did briefly look it up, but couldn't really find anything that I could use. What I really need is a film.

I'm still pretty sure a polarizer would help, if they are using them in DLP printers there must be a reason(other than the fact LCD's use them).

It may take a bit longer since it would be blocking some of the light, but in the end this would be just an experiment.

 

A curing box may also be the answer too, my light source is basically a point, so no doubt that someone more evenly distributed would help. 

I was planning on making one as I said to sparrow. Already have the LED's, and now I'm just waiting for some free time to open up.

 

Unfortunately I don't have an inkjet printer, but I'll definitely try that out. My dad has an ink jet, but I just hope it has a decent enough DPI.

 

Ahh! That stoufer gauge looks pretty awesome. I've narrowed it down to about 5 minutes, but the whole time, I'm having to hold the light at a certain height and move it around constantly.

 

What exactly does a spray etcher do, constantly spray etching solution on the board?

I just looked at an Instructable, as well as a YouTube video, but they haven't really said what it does exactly.

 

As for plated through holes, I used to use 0.9mm rivets, but just recently I bought everything I need to do electroplating.

I've been waiting to do it, but just haven't needed to do any double sided boards since I got it all.

The container is all ready to go, and all I really need to do is mix the ink and prepare the bath.

 

valusoft wrote:

Yes, I make pcbs here at my place in support of commercial work, but to express it in your shorter form could be misleading to the reader. I do proof of concept prototype designs and extremely short production runs (largest was 8 evaluation units for the Canadian Government). I always make the first unit here before sending the (sometimes modified) design off to one of the quick turn Chinese pcb houses. This way I have a working unit to check mechanical aspects (clearances, ease of assembly, etc) and to start software. It is very useful to be able to photograph an assembled prototype within a couple of days to show the client and obtain feedback. I have also done a "show and tell" via Skype.

 

Ahh very cool, I'm from Canada, but I wonder why they would send it off to Australia to be done.

I see that you see the value in making a quick board before sending it off.

There are always people who say "Why don't you just send it to a fab house, its so cheap and you cant beat the quality."

In fact there was 2 in this topic (Nothing against you guys :) )

 

valusoft wrote:

Now back to the topic. I produce a reversed life sized pdf image which I then take to my favourite printing centre for a transparency. They keep their machines in great condition at no extra cost to me. Like the OP, I then place the toner side against the presensitised pcb and cover that with a sheet of glass and place the combination in a frame to ensure a flat even contact. I have a bank of UV leds which is suspended at about 50mm above the glass/film/pcb assembly. The bank of leds is always larger in area than any pcb that I am making. I have two sheets of baking paper in front of the leds to diffuse the uv light so that the pinpoint light source effect is removed. I control the exposure time based upon experience to prevent the undercutting that the OP is witnessing. The finest I have needed/produced is 10mil traces.

 

Yep, that's pretty much exactly what I do, with some minor differences.

Others were saying to use an inkjet printer instead of laser printer, correct me if I'm wrong, but you use their laser printer for the DPI? Do inkjets just not have the resolution needed for small traces?

 

That baking paper is exactly what I'm looking for! But what are your thoughts on using a polarizing film for that purpose?

 

valusoft wrote:

I can have a pcb etched in around an hour from finishing the pdf file, including driving to the printing centre. Assembly of course is then subject to parts availability and drilling, and soldering. I use a chemical dip to tin the copper traces. No solder mask or component overlay at this prototype stage.

 

Hope that helps.

 

Ross

 

I'm guessing that chemical dip is "Liquid Tin" correct?

I've also been using that, but I find that its sometimes a little uneven. But I'd put money on it being caused by forgetting to clean the board after etching.

 

That helps for sure, next board I will try using parchment paper.

Though I'd still like to hear what you think about replacing the parchment paper with a polarizing film.

 

Thanks

 

 

andrewm1973 wrote:

Presensitised PCBs would probably mean positive photoresist (Kinsten).  Positive photoresist although convenient is not very high performance and is easy to scratch.  If you want to break the 8/8 thou line/space barrier go for negative resist.

 

A cheap pigment inkjet will get equal or better performance for this task than the giant clean print centre laser.  Orifice Works in Oz have Epson Workforce pigment printers under $100.  Other parts of world are probably cheaper.  So if you are doing them often or you are not close to a print centre with a well maintained commercial class laser printer, that is an option.  Home/SMB class lasers - the kind that cost $50 to $4000 are not as good as a $100 inkjet for this task.

 

Unless you have deep pockets then LEDs are not the best option for a UV light source.  Most cheap LEDs are 405nm.  UV Resist is optimized for the Mercury "I-Line" of 365nm.  365nm LEDs are much dearer.  Also to get best results you need to have enough power to do the primary polymerization in around 1 minute.  Slower is fuzzier.  Black light bulbs (Disco lights / Bug Zapper lights) are a cheap source of 365nm light.

 

Cheap 405nm LEDs will work.  They are not optimal.

 

Using a sheet of baking paper as a diffuser is the exact opposite of collimation (what the OP mistakenly thought was the polarization he needed).  It will make more light at an higher incident angle to get under the traces.  Again if you are aiming for 10/10, 12/12 or larger you can get away with it no worries.  If you are aiming at under 8/8 thou try to make your light as parallel as possible.  Sunlight is very parallel :)  It is what I used when I made my first UV PCB 1/4 a century ago.  Back then I probably did 20 though traces than necked down to 16 to get between DIP IC legs.  These days I need to make PCBs for things like FPC connectors that have 0.3mm pitch so need 6/6 though trace/space.

 

A sheet of glass for the contact can work fine.  Especially if you are making a small PCB and there is less risk of things being warped causing issues.  Best practice is a sheet of cheap (not UV stable) plastic with a vacuum pump pulling it into contact with the PCB phototool sandwich.

 

The other thing to take into account.  I am saying "If you want reliable 8/8 or under rules" you need all this stuff.  When I say reliable I mean a whole PCB with 8/8 all over it and doing that every time.  If you have a PCB that just has some parts that are fine and the rest of it is 16/16 - then you can just pay attention to the important part to make sure it is OK.  Or if it is a tiny PCB that is <8/8 then just make 16 copies of it on the one blank and one of them will turn out fine even with LEDs and average phototools.

 

 

 

I'm not actually using pre-sensitized boards, I purchase rolls and laminate them. I had no idea there were positive and negative films, but it turns out that I've been using negative the whole time.

I had said to someone above, but I'll definitely try using an inkjet. The only thing I'm worried about is the DPI on my dads printer. My laser is 2400 dpi so its pretty hard to beat the resolution, but it definitely doesn't have the best blacks on transparency film. I actually print a blank page onto the sheet a few times before I print the actual mask. I do this to get as much binder on the page as possible, it does seem to help.

 

My pockets are definitely shallow. But you can actually get about 100 5054(SMD size) LEDs, that are 365-370nm/395-400nm, and 1 watt for only about $13 CAD.

 

Even though diffusion is the opposite of polarization, the whole point is basically to remove the pinpoint effect, which I still think a polarizing film would achieve.

Next board I'll try the baking paper, but I think its still worth a shot to try the polarizing film. My DLP printer uses one, and its not one of the regular films the LCD would already use.

 

Why not UV stable plastic? What is the difference?

I am using glass, but what is a photo tool?

 

I actually have been doing 8mil traces, and the last board I made has 2 traces going between my DIP pads. ;)

But exactly like you said, the problem is with reliability.

As you can see in the image, in the red areas there is some messy shit.

There was a tiny bit of bleed in the red circles, causing the traces to not etch, and so I had to use my cnc router like a band saw, and hand fed the board to fix the traces.

 

A curing box would probably fix this issue, cause I'm having to move my pinpoint light around constantly, and it causes some areas to bleed if they are exposed too long.

 

The board is for a NES (Nintendo Entertainment System) controller, just need to drill and mill and its done. Just a little side project for a tutorial that I'm currently making.

 

 

Paulvdh wrote:

The polarizers you mention are the same, but just cut in different orientations out of the same base material.

Watch some youtube vid's about polarisers and polarized light.

 

But it won't get you anywhere.

As mentioned beore 2 printed masks won't work well. You need the ink/toner side flat against your PCB to get sharp edges.

 

You can also create more distance between your (point) light source and your PCB to get sharper shadow edges.

 

As an alternative you can use a small CNC machine and an UV laser to write directly into your uncured solder mask.

Marco Reps has been experimenting with this:

https://www.youtube.com/watch?v=gm5P74vcB84

 

Ok, I kind of thought that's what it was.

 

Yeah I will do that. I just thought there must be a reason that my DLP printer uses one (its not one that the LCD always uses).

I know it won't straighten the light, but I thought that if the light was coming in from an angle, it wouldn't be able to pass through.

I'm probably not understanding something though.

 

I always make sure that one of my masks has its toner flat against the PCB, but when I didn't add the second mask it didn't work at all. There wasn't enough time to cure it, before the masked parts started to cure.

 

Next time I make another board, if its before I create a curing box, I'll try increasing the distance like you said.

 

Oh yeah, I've seen that video before, hasn't he also made a board with a DLP printers LCD screen as well? It may have been another similar YouTuber, but I'm pretty sure that was him too. 

I completely forgot about that video, I actually have the same laser too. Its interchangeable with my desktop router's spindle. Only have used it once because I haven't had a use for it, but maybe now I will.

 

mnehpets wrote:

Big clive has a bunch of youtube videos describing his tests with home PCB etching. Toner masks don't give good results. You might get better results with inkjet dye or pigment on transparency:

 

https://www.youtube.com/watch?v=...

 

And his tests for different transparency material:

 

https://www.youtube.com/watch?v=...

 

- S

 

 

Ahh, did not know he made videos like that, I've only seen his tear down videos.

My results are definitely not horrible with toner, its just not as reliable as I'd like it to be. 

If you look at it under a microscope they actually look very nice too. Nice smooth traces, unlike toner transfer.

Its about 90% reliable, but every once in a while I'll get an issue like image in the board above.

 

sparrow2 wrote:

I just remembered one ting. 

 

If you print isn't black enough on a plastic sheet (and there are also many different types) try on paper, and then give the paper some oil (like WD40) then it will work as a film. 

 

I'm not sure I understand what you mean.

Will it allow light to pass through?

 

andrewm1973 wrote:

The second video is the type of tranny material I recommend.  It is far superior to the gelatine based ones.

 

So that's basically Kiwojet?

Do you know if that's PET film?

Last Edited: Mon. Dec 24, 2018 - 10:22 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Does not matter if the laser printer is 2400 dpi if it can't make a solid black to block the light.  Your mask/photo-tool is going to let light through where it should block it and you will have hard areas on the resist that need touching up.

 

Epson inkjets are going to be 1440 dpi as a base resolution.  That is high enough to get you 8/8 without a problem.

 

Really polarizing film _IS_NOT_ what you are looking for.  Polarization is not going to help with the photo resist.  Save your time and money by either moving your light source further away or building a collimation grid.

 

The spray etcher you saw on instructable was probably mine :)  The spray etcher makes etching more consistent and allows higher aspect ratios.  Something needed if you are going to try do boards with very mixed density and sub 8/8 features.

 

"photo tool" is what you are calling "mask"  they both mean the same thing.

 

A pin point (or many pin points) for the light are fine as long as they are a long way away.  Photo resist is fairly high contrast.  Parallel is more important than absolute uniformity.

 

UV stable plastic has UV blocking agents added to it.  This is to make it survive in the sun.  It also makes it opaque to UV light.  If you go to a caravan/boat canopy maker and ask for some clear PVC they use for awnings/windows it will be UV safe and block UV light.  If you go to a homewares store and ask for some clear PVC used as table cloths it will be cheap plastic and not have UV inhibitors.  It will be clear to UV and will work much better.

 

Make the time and effort to build a box.  It will help with consitancy. 

 

Don't bother to chase polarizing film.  It will not help at all.

 

The Kiwojet film I use and the stuff BigClive showed (don't know if it was Kiwo brand or not) is not a single type of plastic.  It make well be PET as the substrate, but it is several layers of different plastic laminated together.  There are actually two types.  One that is better on dye printers and one that is better for pigment printers.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes that is true. Turns out my dads printer is 600dpi for black and white, and then 2400dpi for color.

I'll try making the curing box before I invest in a new printer, because the laser is doing pretty good, apart from needing 2 masks.

 

Very cool, maybe some day I'll follow your tutorial. At the moment I'm able to do letter sized board so there isn't any rush.

 

Photo tool, gotchya.

 

Oh that's interesting, I'm guessing that old plastics go yellow because the lack of UV stabilizing agents?

 

I will for sure, was going to make a timer or something along those lines.

 

Ok I'll skip the film, I was looking at a 30x20cm piece that would only cost $20 CAD with shipping, but I guess for even that its not worth it.

 

I noticed that ink jet transparency film is pretty darn expensive. I managed to find a huge roll of unbranded PET film (advertised for inkjets) that was only like $30, I was hoping it would be the same. Still may give it a try if I get an inkjet.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Shop around.  You don't need Kiwo or Ulano brand stuff.  Search the usual online shopping places for "inkjet Screen printing film"  the ones that say "Waterproof" or that have a grey hazy look to them are the better style.  You should be able to find them sub $1. I know I can see some in Oz that are under $1 per A3 sheet right now.

 

In the 50 pack of A4 brand name ones I paid less than $1 per sheet from memory.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

On a side note:

"transparent" materials are only so at some limited wavelengths.

It may be that the difference in transparency of your black toner and plastic sheet are not so far apart for UV light.

 

UV is on the other side of the spectum from IR, but IR remote controls do not work through a plane of glass (Just stand outside and try to change channels on your TV).

I also saw some thermal camera photo's of a sewing machine which had one of those black timing belts.

The black rubber timing belt was almost invisible on that picuture, you could see the background through it quite clearly.

The reinforcement threads in the timing belt were pitch black though.

 

There may be additives to your plastic sheets that absorb UV light, to increase the UV resistance / longevity of those sheets.

Doing magic with a USD 7 Logic Analyser: https://www.avrfreaks.net/comment/2421756#comment-2421756

Bunch of old projects with AVR's: http://www.hoevendesign.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Perhaps a stupid question but have you tried to print on the same paper twice ? (fiddle with the paper feeder a so i align as good you can)

 

If only 1 of 10 prints align it's worth it :)

 

add:

And have you tried normal paper and spray it with wd40 ? (after you have  printed)

 

 

 

 

Last Edited: Wed. Dec 26, 2018 - 08:15 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

andrewm1973 wrote:

Shop around.  You don't need Kiwo or Ulano brand stuff.  Search the usual online shopping places for "inkjet Screen printing film"  the ones that say "Waterproof" or that have a grey hazy look to them are the better style.  You should be able to find them sub $1. I know I can see some in Oz that are under $1 per A3 sheet right now.

 

In the 50 pack of A4 brand name ones I paid less than $1 per sheet from memory.

 

What do you think of this?

It would have to be cut to size, but man, if it works it would be a great value.

 

Paulvdh wrote:

On a side note:

 

Ah, thats right, I totally knew that, but just was not able to put two and two together.

 

I actually have stood outside a friends house before and turned his TV off with a remote as a prank.

Won't be able to tell you how it worked though, I have also seen similar IR videos of viewing a human through a panel of glass and it was like they weren't even there.

 

 

sparrow2 wrote:

Perhaps a stupid question but have you tried to print on the same paper twice ? (fiddle with the paper feeder a so i align as good you can)

 

If only 1 of 10 prints align it's worth it :)

 

add:

And have you tried normal paper and spray it with wd40 ? (after you have  printed)

 

Not stupid at all, yes I have tried it. I could never get it to align properly. That was on my old printer though, and this new one is a ton better. I just might give it another go.

I have actually been printing a blank page a few times (on the same sheet) before I actually print the mask. It seems to help, I guess it just gets a bit of extra toner that was left on the rollers. Really not sure why it helps though.

 

I saw you mention the WD40 thing, I haven't yet. I just don't really know what you mean by that. I asked in one of the last huge replies, I don't blame you for missing it.

Is that a toner transfer method? I tried to do that with alcohol and acetone with no avail.

Last Edited: Fri. Dec 28, 2018 - 12:41 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I've never tried it myself but I've heared of the "WD 40 thing"  before. The goal apparently is to first print your layout on regular paper, and then soak the paper in WD 40 (or other oil). The oil makes the paper partly transparent and you can stick the oily paper directly to your PCB and under the UV light.

Doing magic with a USD 7 Logic Analyser: https://www.avrfreaks.net/comment/2421756#comment-2421756

Bunch of old projects with AVR's: http://www.hoevendesign.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

That looks like the cheaper gelatin coated ones from the lower pictures

 

This one (that says both "waterproof" and "pigment") looks more like the better type

 

http://www.aliexpress.com/item/2...

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

We did a lot of try and errors in the 80's.

And one thing that worked was to print on normal paper (good quality meaning no big fibres) because the printer liked to print on that.

 

Then there was two ways to get light though the white part,

 

1 just give the hole thing some oil. All paper get more or less transparent

 

2. a bit better but more work, we had some transparent film that was meant to seal photos, this version needed heat from an iron, when that was done put the hole thing in warm water and let the paper desolve (for this cheap paper is best).

the ink will stay in the glue. (we also made overheads this way as well)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Also:

Because you already have a CNC or 3D printer like thing you can put some kind of pen in it and draw on any material you have.

With this you can also write multiple layers of ink on the same piece of paper / film / PCB.

With the right kind of etch resistant pen you can write directly on a PCB & etch it.

Another maybe novel way is to write with your pen on the UV film, and then develop it.

Doing magic with a USD 7 Logic Analyser: https://www.avrfreaks.net/comment/2421756#comment-2421756

Bunch of old projects with AVR's: http://www.hoevendesign.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Paulvdh wrote:

I've never tried it myself but I've heared of the "WD 40 thing"  before. The goal apparently is to first print your layout on regular paper, and then soak the paper in WD 40 (or other oil). The oil makes the paper partly transparent and you can stick the oily paper directly to your PCB and under the UV light.

 

Oh very interesting, I will definitely try that out.

andrewm1973 wrote:

That looks like the cheaper gelatin coated ones from the lower pictures

 

This one (that says both "waterproof" and "pigment") looks more like the better type

 

http://www.aliexpress.com/item/2...

 

 

Yeah I saw that a couple days after I had last posted, not the link you sent but the details about it being non-waterproof.

I almost bought 1 that's the same from your link before I read the post, its almost the price as buying 2! 

 

sparrow2 wrote:

We did a lot of try and errors in the 80's.

And one thing that worked was to print on normal paper (good quality meaning no big fibres) because the printer liked to print on that.

 

Then there was two ways to get light though the white part,

 

1 just give the hole thing some oil. All paper get more or less transparent

 

2. a bit better but more work, we had some transparent film that was meant to seal photos, this version needed heat from an iron, when that was done put the hole thing in warm water and let the paper desolve (for this cheap paper is best).

the ink will stay in the glue. (we also made overheads this way as well)

 

I just said to someone else, but I will try this out next time.

Someone else said they used parchment paper on top of the regular films to defeat the pinpoint light, I'm guessing using regular paper with oil would void the need for that though.

 

I think my printer might jam if I tried to print on a laminate, but that is truly interesting.

 

Paulvdh wrote:

Also:

Because you already have a CNC or 3D printer like thing you can put some kind of pen in it and draw on any material you have.

With this you can also write multiple layers of ink on the same piece of paper / film / PCB.

With the right kind of etch resistant pen you can write directly on a PCB & etch it.

Another maybe novel way is to write with your pen on the UV film, and then develop it.

 

Funny I was thinking of doing something similar, but I came to the conclusion it would be too much work.

There must be a way to get the same reliability and quality as a fab house using their methods.

Guess it all comes down to equipment though, which is hard to match without piles of cash.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I think my printer might jam if I tried to print on a laminate, but that is truly interesting.

No you print on the paper, then laminate, then desolve the paper.