what do you think about "pcb laser exposer" ?

[legacy]

see this (uncompleted) paper
my expression on that page:

what do you think guys ?

[jeremy]

Yes, I have had this idea before  . But I was not the first, google Laser Direct Imaging (LDI).

The basic trouble I had was that in order to beat photolithography masks, you need 4mil/4mil or better. All of a sudden,

There is one problem with this: as the beam length varies with the angle of the deflected laser beam, the focus of the laser lens would also have to be adjusted for each beam length. As this isn't done, the laser gets blurry at the ends and the middle of the scanline. But this is not a problem in practice, as the image gets sharp enough.

actually matters a lot. Also, in this particular case, getting a second axis of movement is not solved.

Doing it with the classic mechanical XY stage *does* work, but as far as I know it is neither cheaper nor faster than plain old UV exposure lamps. Particularly if you try to get stages which can go down to anywhere near 4mil precision and accuracy.

Actually, it is quite common to do this for the mask, using a red laser. It is much easier to mount and move plastic sheets than FR4, because they can roll on a drum and be tensioned. See http://en.wikipedia.org/wiki/Photoplotter

[jeremy]

PS: here is an interesting paper with some results: http://bulletin.pan.pl/(56-1)17.pdf

[Alex Eisenhut]

see this (uncompleted) paper
my expression on that page:

what do you think guys ?

 

Why bother? I don't understand this obsession with home-made PCBs. Shops do it better. End of story.

Hardware ain't like software, you can't just recompile. Let someone else build PCBs for you.

[Jeroen3]

Both have advantages and disadvantages.
Where sitting on your couch waiting for the mailman is preferred by some people. Other like to work with nasty chemicals and dusty cnc's.
Don't reinvent the wheel yourself. Buy a wheel.

In a school lab this would be a nice toy. Better than the cheap laser printers with toner stains all over the paper.
And the stained glass plates in the UV lithographer.

[westfw]

Meh.  At this point in time, the "etching" phase is no longer so much an issue.  You want more than one layer, plated through holes, and soldermask.

[legacy]

Actually, it is quite common to do this for the mask, using a red laser. It is much easier to mount and move plastic sheets than FR4, because they can roll on a drum and be tensioned. See http://en.wikipedia.org/wiki/Photoplotter

interesting things 

[legacy]

video (unfortunately in german) is summarizing the paper i posted

[legacy]

could a machine like this be interesting ?

[atferrari]

see this (uncompleted) paper
my expression on that page:

what do you think guys ?

More or less related information collected in one place. Not to the point. What point? I do not know.

As is, it is just     

[legacy]

More or less related information collected in one place. Not to the point. What point? I do not know.

the point is: home made LDI, could it be cool ?

in the wiki page the author is collecting ideas and informations, that page looks like a proof of concept in where he/she is talking about the "idea" of using a laser printer optics + an UV laser + a scanner plane, in order to realize an UV laser exposer, something that goes into the direction of Laser Direct Imaging

the real point is:
- could it work ?
- could it work better than traditional methods ?
- could it cost the less ?

plus technical information/considerations about that

[atferrari]

Last update, April 2012. He lost momentum I guess.

[jeremy]

- could it work ?
- could it work better than traditional methods ?
- could it cost the less ?

my guesses:
- yes, it already does
- yes, it already does (for PCBs)
- no

[legacy]

that's cool 

have seen good home-made LDI ?
i'd like to see how they are made, which laser they use, and something about the photo film.

p.s.
have you seen the home-mace LDI of the german guy i have posted ? is it a toy or something that could become "cool" ?
this technology comes from recycling old stuff that are flatting around, e.g. an old laser print, a scanner, etc

p.s.2
thank you for the paper, i am going reading it 

[jeremy]

I hadn't seen the home made guy's stuff, it's an interesting take on the problem. But I think he realised that he was a bit over his head. Trouble is that the ideas are really simple, but usually the problem with these optical ideas are a) you have to machine things at ridiculous tolerances and b) when you have a spot of 1um, vibration, humidity and temperature actually can change things a lot (your metal frames expand and contract, diode changes its wavelength with temp, etc). And yes, I have literally watched a diode undergo a wavelength shift as it warmed up (using an optical spectrum analyser).

I don't know about "good" home made stuff (although you can see a few ok-ish videos on youtube), but you can definitely do it well in a commercial setting. I work with lasers, so I find it interesting as an engineering problem, but I don't really see it as a real competitor to fab houses. I think it might be popular in MEMS research for prototyping.

I think the best option nowadays is to get a 405nm diode designed for a blu-ray drive. I believe DuPont make a 405nm sensitive photoresist that you can use with it. The key is to get good focussing to reduce the spot size and decrease the required exposure time. Assuming you had the right mounting gear in a lab setting, getting a spot size of a few micron is definitely achievable. If you want to do the stage mechanics at this sort of level, you will probably need something with decent (aka precision THK) ballscrews with not just a rotary-encoder style servo, but with linear encoders to counteract any backlash. Or possibly linear motors from a pick-n-place. You could do it with galvanometers too, but you would make the focussing problem even more sensitive.

The true UV lasers I have seen in the past (<400nm) have cost about as much as a car. So I wouldn't go down that road unless you have a lot of disposable income  Even just the properly certified safety glasses for this stuff are a few hundred dollars.

[Alex Eisenhut]

could a machine like this be interesting ?

Most machines can be "interesting". Can they be useful and affordable?

Why this fixation on PCBs? I would understand if it's 1989.

 

[free_electron]

LDI is obsolete. the PCB industry has used LDI for at least 15 years. it's useless to pursue that stuff anymore. the mechanics are too complex and the beam deforms depending on deflection.

Maskless imaging is the future. basically an overhead projector using a DLP module that projects the rasterised data directly. using exposure bursts and dithered information they can do structures down to 20 micron ... no problem.

the large maskless imagers use an array of DLP modules and a deep-uv emitter. a 18 by 24 inch panel is exposed in mere seconds at a zero consumables cost.

[tonyarkles]

For me, the "fixation with PCBs" comes from the fact that they are currently the bottleneck in my design process. I have an idea, sketch it out on paper, do some math, draw it out in eagle, and have a first cut design in a day. The parts show up from digikey or newark the next day. And then for the board, I can either do it cheaply and wait a few weeks for it to show up, or pay through the nose to have it show up around the same time as the other parts.

[legacy]

thank you jeremy, excellent exhaustive answer

[SirNick]

Meh.  At this point in time, the "etching" phase is no longer so much an issue.  You want more than one layer, plated through holes, and soldermask.

Exactly.  I started to give up on home-brew PCBs about a year ago.  Even if I could put bare copper-clad board into a machine and get an etched PCB out, any board I could finish at home using DIY methods is still only marginally more useful than a breadboard.

I need stencils, 6-mil resolution, PTH / vias, and at this point even 2 layers is a limiting factor for some designs.  If a single-sided PCB with realistic tolerances will do the job, you may as well use protoboard and TH parts, because the design is obviously not all that demanding.  Even then, a recent board I made had a few hundred holes.  I couldn't have beaten the fab house round-trip time doing it myself anyway.

Waiting sucks, but at $5 / sq. in. in two weeks or so, I just suck it up and use the lead time to round-robin projects, or work on code.  (The latter is usually holding up my project far longer than the PCB fab.)

[legacy]

Maskless imaging is the future. basically an overhead projector using a DLP module that projects the rasterised data directly. using exposure bursts and dithered information they can do structures down to 20 micron ... no problem.

great information, never seen before, have you got a video about this kind of tools ?
and when do you think we (hobbits) could buy/try-to-realize something like that ?

[legacy]

I need stencils, 6-mil resolution, PTH / vias, and at this point even 2 layers is a limiting factor for some designs.

i think it depends on kind of project-target you aim for. In my case i have been asking PCB_services to realize everything is more than 1 layer and has VIAs, they do the job better than i can try to do, but i usually realize PCB by myself if they are easy enough, e.g. 1 layer without any VIAs.

for example i do not want to wait the mailman if i want to try a sensor, in this case i realize a sort of home-made "breakout", just routing it on 1 layer pcb plus headers and footer to accommodate the sensor.

it's better than using 1000hole, better than using proto-board, and faster then waiting for the mail-man, and for digital stuff i am pretty happy with my FPGA, because it saves me from wiring chips


btw i am asking about "PCB exposer" not because i want to build one of them, but because i am curious about them: i recently got a degree in eng. electronic at my university and i did a lot of exams about LDI, pure theory, never seen in reality

[legacy]

Most machines can be "interesting". Can they be useful and affordable?

because LDI are used in optoelectronics's courses in my university, unfortunately with ideal context, so i'd like to understand real problems about such a machines, also i was shocked about the guy who tried to realize an home-made LDI because i was thinking (and i am still thinking) that it could not be useful and affordable

i would like to know the next step, what will take the place of the LDI technology

[atferrari]

I hadn't seen the home made guy's stuff, it's an interesting take on the problem. But I think he realised that he was a bit over his head. Trouble is that the ideas are really simple, but usually the problem with these optical ideas are a) you have to machine things at ridiculous tolerances and b) when you have a spot of 1um, vibration, humidity and temperature actually can change things a lot (your metal frames expand and contract, diode changes its wavelength with temp, etc). And yes, I have literally watched a diode undergo a wavelength shift as it warmed up (using an optical spectrum analyser).

I don't know about "good" home made stuff (although you can see a few ok-ish videos on youtube), but you can definitely do it well in a commercial setting. I work with lasers, so I find it interesting as an engineering problem, but I don't really see it as a real competitor to fab houses. I think it might be popular in MEMS research for prototyping.

I think the best option nowadays is to get a 405nm diode designed for a blu-ray drive. I believe DuPont make a 405nm sensitive photoresist that you can use with it. The key is to get good focussing to reduce the spot size and decrease the required exposure time. Assuming you had the right mounting gear in a lab setting, getting a spot size of a few micron is definitely achievable. If you want to do the stage mechanics at this sort of level, you will probably need something with decent (aka precision THK) ballscrews with not just a rotary-encoder style servo, but with linear encoders to counteract any backlash. Or possibly linear motors from a pick-n-place. You could do it with galvanometers too, but you would make the focussing problem even more sensitive.

The true UV lasers I have seen in the past (<400nm) have cost about as much as a car. So I wouldn't go down that road unless you have a lot of disposable income  Even just the properly certified safety glasses for this stuff are a few hundred dollars.

By no means at your level of expertise but I would say that is easy to understand how hard  is to focus a laser beam properly just by looking at what refined optics they had to resort just to read a CD. And then the DVD...

I spent quite some time these days in reading, thinking of building a laser PCB engraver but after your post I started to actually think!    Gracias 

[Alex Eisenhut]

For me, the "fixation with PCBs" comes from the fact that they are currently the bottleneck in my design process. I have an idea, sketch it out on paper, do some math, draw it out in eagle, and have a first cut design in a day. The parts show up from digikey or newark the next day. And then for the board, I can either do it cheaply and wait a few weeks for it to show up, or pay through the nose to have it show up around the same time as the other parts.

Well that's certainly a valid argument. However, since modern electronics seems to be mostly one IC with some support parts and a digital bus, what's wrong with stocking up on Surfboards?

http://www.digikey.ca/product-search/en?x=0&y=0&lang=en&site=ca&KeyWords=surfboards

[legacy]

breakouts are good for chips, in my case i use to integrate OA and a bit of analog part into an home made PCB which is the most problematic stuff if done on 1000-holes boards; PCB toner transfer is exactly good for that, small circuit adapter/breakout, or things related to "carriers" boards, typically related to power and ground, plus their connectors.

[Alex Eisenhut]

breakouts are good for chips, in my case i use to integrate OA and a bit of analog part into an home made PCB which is the most problematic stuff if done on 1000-holes boards; PCB toner transfer is exactly good for that, small circuit adapter/breakout, or things related to "carriers" boards, typically related to power and ground, plus their connectors.

My projects aren't that ambitious. I use surfboards to create the "local systems", then I solder the whole mess down to a 3x4" single-side copper clad which are still quite cheap, then all the little messes talk to each other.

[jeremy]

I spent quite some time these days in reading, thinking of building a laser PCB engraver but after your post I started to actually think!    Gracias 

Don't let me stop you! It's not going to be competitive with lithography, but if it's for fun then I say why not  even if you don't finish, you will learn a lot. Nothing wrong with thinking or reading.

Just some notes:
- you can't use a co2 laser to get rid of copper (10um wavelength). You need to vapourise it at a wavelength that absorbs a lot more energy. More like 1um or below
- you need an insane amount of power to vapourise metal. A laser outputting this much power continuously would melt itself straight away. So you need to use a pulsing/duty cycle trick like "q switching" or "mode locking". The other problem is that if your pulse takes too long to deliver all of the power, the metal will redistribute the heat like a heat sink (particularly a problem with copper!). When I say too long, I'm talking something like more than 100ps.
- you also need to generate a pulse which will not remove the fibreglass.
- you will almost certainly need to use a diode pumped solid state laser (DPSS)

Here is a commercial product:
- $150k each iirc

Or Google "laser ablation pcb" for lots more interesting information.

[legacy]

thanks for the key-search (laser ablation pcb), it's exactly the WOW technology i was looking for

[westfw]

Maskless imaging is the future.

So... Can I make a registration frame for my "XXX with Retina Display",  shove in a pre-sensitized PCB, run some display software to make sure the scaling is right and I only use blue pixels, and expose a PCB that way?   I'm not clear on the issues of pixel continuity on high-res color displays or screen intensity and wavelengths vs photoresist sensitivity, but not nominal "pixel density" of 300+ppi should be plenty for a lot of the same designs where people expect to be able to use mechanical etching or direct toner transfer techniques...  Obviously I could put up with "10 minute" scale exposures rather than "mere seconds."