PDA

View Full Version : LaserJet 6L Printer - Positive Films - A New Idea That Needs Guidance



idbruce
02-05-2012, 04:19 PM
Hello Everyone

As I am sure that many of you already know, I am performing some experiments with my new exposure box design and trying to obtain some very nice exposures. During some of these experiments, I have been using a LaserJet 6L and transparent overhead film to make my positive masks. After a lot of experimenting, I have come to the conclusion that I need a much better positive film to obtain better results, so I have spent an extensive amount of time researching the different methods that are available.

Creating inadequate positives for PCB photo-imaging is a common problem amongst electronics enthuiasts. I am sure there are several printers available that will print a nice transparency, as well as other solutions, but I want to attempt something new to try and resolve this problem. I want to create a photomask interface for my printer. The LaserJet 6L has a maximum density setting of Level 5, but somehow I want to surpass this density level with an electronic interface. I am certain there is a connection somewhere within the printer that can be seperated which controls the density of the toner. So I propose an interface that can be inserted between two connectors to control the density of the toner output. During normal printer operations, the interface is simply bypassed by the selection of a switch position, but by altering the switch position, the interface would allow fine tuning of the toner transfer, capable of surpassing the maximum factory level, for the purpose of creating purely opaque photo masks.

What do you guys think? Is this possible?

Bruce

Phil Pilgrim (PhiPi)
02-05-2012, 04:59 PM
A lot depends on the toner itself. If you use cheap toner, you won't get good results. If you can find an "ultrablack" or "ultrafine black" toner cartridge for your printer, use it. Also, be sure to print a mirror image, so you can do your exposures toner side down.

-Phil

Leon
02-05-2012, 05:14 PM
Many years ago I worked for Rank Xerox and Xerox Research (UK). Increasing toner density tended to cause problems with "background" - unwanted toner deposited on areas that should have been free of toner.

Given that cheap inkjet printers produce excellent transparencies, I don't see the point of messing about with a laser printer, which will probably never achieve the same quality.

idbruce
02-06-2012, 03:43 AM
@Phil and Leon

With all due respect, you guys are not telling me what I want to hear :)

Bruce

Phil Pilgrim (PhiPi)
02-06-2012, 04:00 AM
Bruce,

Maybe you're asking the wrong questions. :)

-Phil

idbruce
02-06-2012, 04:03 AM
Phil

Perhaps, but it is just that endless problem solving quest that I have been on since birth. I simply can't help myself. :)

jrjr.
02-06-2012, 04:46 AM
Bruce,

Phil is right on the reversed image, with direct contact to the pcb film.

I'm assuming this is some type of photosensitive film PCB overlay, correct ?

If so, then removing the cover sheet before exposure will generate finer lines with direct contact.
This will require about double the current exposure time, since the PCB film cover sheets constructively
add at the exposure wavelength. (reduced exposure time required with cover sheet on)

There was a product called 'Laser Buddy', mostly Methylene Chloride etc; that caused toner
to fuse (very slight swelling) better (chemically) for a improved contrast ratio on paper
but I expect that it would eat the transparency films.

That is what I use as a first step to make the first art phototools by camera and laserjet
at a 1.5 to 2X reduction of the laser prints to master films.
Then shoot the negs/pos from LPF films for the final exposure masks.

Most of this stuff is still available as Rapid Access (RA) process, by various makers.

It's about a buck a print for the RA films, and it will eliminate all your problems
on density (except for the pinholes that always creep in).

Most shops then shoot the master phototools (contact print) to Diazo
for the final tools (sacrificial after multiple uses) on production boards.

That brings up another interesting feature, that the Diazo masks are pretty
transparent, and mostly act as FILTERS at the exposure wavelength.

That speaks to Leon's point of making inkjet transparency exposures
that are FILTERS rather than the full spectrum blockage with toner.

Kick around the photo sites on books about inkjet films and inks for making exposure negatives
for making platinum and palladium prints and you will find certain inks that work best at filtering for UV.

Your next problem will be dimensional stability during the process use,
so use the thicker 7mil films for phototools and keep the process boards
and films <at room temperature> during all the process steps.

Or, you could bypass <all> of this and go to a focused dot UV direct write
and get single board exposures in about 5 minutes off a re-purposed flatbed plotter.

jr

idbruce
02-06-2012, 05:27 AM
@jrjr

I don't mean any offense by this, but this is not my first day at the rodeo :) However, you presented a fairly interesting discussion.


Phil is right on the reversed image, with direct contact to the pcb film.

Yes, he is right, but I knew that :)


I'm assuming this is some type of photosensitive film PCB overlay, correct ?

Yes, it is a positive acting photosensitive film. However, this film does not require UV lighting for exposure. It can be exposed with a standard incandescent bulb, and that is what I am using.

The internet has a lot of information pertaining to the subject of creating PCBs, especially with using laser printed transparencies as a positive film or mask. However, what these articles fail to mention is that if you want very fine detail with the photographic imaging process, you must have an opaque mask, otherwise you will never achieve fine detail. I can achieve 0.010" trace widths with no problem, but I am going after consistent 0.004" trace widths.

As mentioned in the original post, I am not the only person in the world who is having trouble creating a nice opaque positive, and I would like to attempt to achieve an economical solution for a lot of people. I am sure I can still find a print shop with a stat, floor, or vertical camera that can make a nice positive for me, but that does not help the poor guy that is fifty miles from the nearest print shop. Whereas slightly modifying the maximum density of the toner output on a cheap laser printer may be just the ticket for a lot of folks, instead of just me.

Also as I previously mentioned, the LaserJet 6L has a maximum density level of 5, but the levels range from 1 to 5. Now this printer already comes close to an opaque image, but it is not completely light free. At this point, I can only imagine what it would look like if I could increase it to 6.

Bruce

jrjr.
02-06-2012, 06:46 AM
Bruce,

They are fairly interesting <facts>.

The film may be broad in ability but if it has a cover sheet then it reacts
at a <specific> wavelength. (uv shorter wavelengths , means sharper images)
Physics isn't going to change to accommodate your budget.

I've always used florescent UV GERM lights, (short wavelength, intense output, cheap and available, lower heat transfer)
I don't see any real substitute if you are going for 0.004 l/s.

You're bombing the films with the broad spectrum lighting thinking that the opacity will solve
the problems, but you are <creating> the problem with the light source choice.

I didn't see any response to the FILTERS portion of the post and that's the key to your density problems
and a cheap and effective way of making the films by specific inkjet inks.

You can also pick up a thou or two in the post processing by proper dwell times
(a heat stabilization water soak, w/cover film removed) after exposure and before using soda ash
for development with direct impingement and mechanical agitation instead of diluted KOH developers.
KOH is only used as the final stripper.

But that's if you haven't already polymerized the films too much in exposure by the wrong type of light source.

good luck with that wheel, Buckaroo.

jr

the internet is filled with PCB voodoo, here's the facts.
Printed Circuits Handbook, 6th Ed. // Clyde F. Coombs, Jr.
( it's been torrented )

Gadgetman
02-06-2012, 08:39 AM
If I remember right...

The Laserjet 6 series printers have a small wheel you use to set density, hidden somewhere under a cover.
(We mostly used the 6mp at the office, not the 6L)
I assume that this was connected to a variable resistor or something of that nature.
I would try to measure the resistance at both the high and low setting, ti find which way it 'dials'.
If its at 'high resistance' while at the '5' setting, it might be possible to increase the density by adding another resistor in series?
If it's at low resistance, you may have a problem...

Of course, figuring out how the resistor is wired into the system can be the real challenge.

Leon
02-06-2012, 10:55 AM
Bruce,

I can't see you getting anything like 4 mil tracks with a laser printer and standard resist. I can just about manage 6 mil with my process and the etched tracks are very non-uniform. If you can create suitable transparencies you will need special high-resolution resist. Few commercial PCB suppliers offer 4 mil tracks, it needs special equipment and processing; 6 mil is the usual lower limit for standard processing.

jr,

Your comment about inkjet ink acting as a filter reminded me that some people on the Homebrew PCB group have tried different coloured inks. IIRC, yellow was best. I get good results with black ink, so I haven't bothered trying other colours.

idbruce
02-06-2012, 12:24 PM
@Gadgetman



I assume that this was connected to a variable resistor or something of that nature.
I would try to measure the resistance at both the high and low setting, ti find which way it 'dials'.
If its at 'high resistance' while at the '5' setting, it might be possible to increase the density by adding another resistor in series?
If it's at low resistance, you may have a problem...


Yea I was thinking almost exactly along those lines

@Leon



I can't see you getting anything like 4 mil tracks with a laser printer and standard resist.


I am already getting 4mil tracks. I am just not getting them consistently.

Just to prove my theory, I will find a local print shop that can make me an opaque positive, but that will still not solve the problem for folks in rural areas.

Another reason for having the positive completely opaque, is that it also takes the positive image out of the equation for possible exposure problems. I won't have to think, "Well it could be because the positive is letting light pass through".

@jrjr and Leon

Unless you are creating half-tone etchings, I cannot see any possible reason for allowing any light to pass through wire trace areas, and wire traces were meant to be solid lines.

Leon
02-06-2012, 12:37 PM
I don't believe that you can get consistent 4 mil tracks, especially on a real PCB, without special materials and processing. I'd aim for a reliable low-cost 8/8 mil process, it's all that most people want.

idbruce
02-06-2012, 01:08 PM
Leon

Besides the 4 mile tracks, I also want the periods (dots) to consistently show up in my web address. I would imagine they are 4 - 6 mil in diameter.



I'd aim for a reliable low-cost 8/8 mil process, it's all that most people want.


I would have to disagree. That is what most people settle for because of inadequate positives or negatives, incapable of achieving consistent 4 mil tracks.

Bruce

idbruce
02-06-2012, 01:19 PM
Leon

Okay Leon, I will retract some of my last comment. It is not very often that you will actually need a 4 mil track, but just in case it is necessary, I want a process that can achieve it.

Bruce

EDIT: Plus that will permit bragging rights :), as implied by your comment:



I don't believe that you can get consistent 4 mil tracks, especially on a real PCB, without special materials and processing.

Leon
02-06-2012, 01:32 PM
What have you achieved reliably, so far, with actual working PCBs?

idbruce
02-06-2012, 01:41 PM
Leon

With the old exposure box approx. 50 mil, but with the new exposure box, I have not built any working circuts yet.

Bruce

Leon
02-06-2012, 01:53 PM
I'd see what you can do, to establish a baseline for future improvements.

idbruce
02-06-2012, 01:59 PM
Leon

Instead of wasting anymore valuable PCB material, my very next step is to make that opaque positive. That will eliminate all doubts pertaining to the positive mask. I truly believe that is my biggest problem. From there, at least I know the positive image will not cause me any unwanted grief.

Bruce

Leon
02-06-2012, 02:07 PM
I don't understand why you don't just get an inkjet printer and some JetStar film. That's what I and DJ Delorie on the Homebrew PCB group are using. I think that he can manage 6 mil reliably, but he's spent a lot more than me. Here's his PCB stuff:

http://www.delorie.com/pcb/inkjet/

He's actually tried 4/4 mil, but it doesn't seem to have been successful.

I think he's mentioned that the film he is using is similar to the JetStar material I use.

idbruce
02-06-2012, 02:17 PM
Leon

To be perfectly honest, I have never liked inkjet printers or their quality of print, at least with the printers that I have seen and/or experimented with. I have a HP DeskJet 932C that I may have used to print perhaps 10 documents, which is practically brand new. I would not even consider attempting using it for my current experiments. Additonally, JetStar film is a UK remedy and the shipping could not be justifiable for a viable solution.

On the other hand, I have always been impressed with the output of laser printers.

Bruce

idbruce
02-06-2012, 02:36 PM
Leon

That was an interesting link. Just give me a bit more time, which will result in either glory or shame. I sincerely doubt that I could accomplish 4/4, but I believe 4/8 is possible.

Bruce

idbruce
02-06-2012, 02:41 PM
@For Those That Might Be Interested

These excerpts are taken from the HP LaserJet 5L and 6L Printer Service Manual.



High Voltage Power Distribution

The High Voltage Power Supply PCA applies an overlap of DC and AC voltage to
the primary charging roller and the developing roller. This circuit also applies a
positive or negative DC voltage to the transfer charging roller according to the
instructions from the CPU on the DC Controller PCA.

This circuit also controls the image density by changing the primary AC voltage and
the developing AC bias according to the setting of the print density dial.




Step 2: Drum Conditioning

After the drum is physically cleaned, it must be conditioned. This process consists
of applying a uniform negative charge on the surface of the drum with the primary
charging roller, located in the toner cartridge. The primary charging roller is coated
with conductive rubber. An AC bias is applied to the roller to erase any residual
charges from any previous image. In addition, a negative DC bias is applied by the
charging roller to create a uniform negative potential on the drum surface. The
amount of DC voltage is modified by the print density setting.

jrjr.
02-06-2012, 03:53 PM
Bruce,

It helps to talk to the correct end of the horse when at the rodeo.

google : Stouffer Step Tab.

www2.dupont.com/Imaging_Materials/en_US/assets/.../tb9526.pdf

at 4/4 you will be making ADDITIVE circuits.

4/8 is what you make when the process is out of control.

spaces are equally important as traces.

read the Coombs book, 10 times.

there are bias tabs on the toner
cartridge that allow you to push
the HV for varying drum bias, It acts
as a range override to the density setting
for differing qualities of photosensitive drums.

You don't need a propeller to control it, a paper clip will do.

but none of this will help your problem.

jr
73

idbruce
02-06-2012, 04:07 PM
It helps to talk to the correct end of the horse when at the rodeo.

I could not agree more :)


google : Stouffer Step Tab.

I believe a Stouffer Step Wedge is for UV exposures. As I mentioned previously, I am using incandescent lighting for my exposures.

Leon
02-06-2012, 04:18 PM
The same technique will work with an incandescent light source.

idbruce
02-06-2012, 04:54 PM
Leon

Here is something interesting, http://www.sandcarver.org/laserfilm.htm

Lucky for me, I have a graphics art supply house here in town. He just gave me a bunch of high resolution film and various other stuff for free. :) All I can say is WOW. Time for more experimenting.

idbruce
02-06-2012, 06:27 PM
I must say that one of those films has produced a much better positive, especially with a clear coat on top. It may not be perfectly opaque, but it is worth another experiment.

Leon
02-06-2012, 06:58 PM
Leon

Here is something interesting, http://www.sandcarver.org/laserfilm.htm


Inkjets vary a lot, and must be used with the correct film. I get much better results than he gets with both printers.

idbruce
02-06-2012, 07:00 PM
This is an inkjet print on Photobrasive’s Accuart film. The blacks are very dense, and have printed well. The printer is an Epson 1160, set on film (resolution at 1440 by 720 dpi).

Leon
02-06-2012, 07:04 PM
My HP printer delivers 1200 x 1200 dpi. I don't get the problems he mentioned. The JetStar Premium is essential, results are poor on everything else I have tried.

deadwhale
02-06-2012, 08:58 PM
I used to make rubber stamps years ago using a similar method, printing with a laser printer and exposing a substrate to ultraviolet lights to harden it. I could get very fine lines, having nothing left of that venture, I can't say it was 4 mil or 5, but it was quite fine.

What I used instead of a transparency was vellum paper, translucent instead of transparent. This seemed to help with the light leaks. Not being familiar with using an incandescent light for this type of work, I can only guess that the biggest difference you would experience other than perhaps quality would be length of exposure time.

ctwardell
02-06-2012, 09:11 PM
I have to ask this question...

Given all of this effort, what is the primary reason for wanting to roll your own boards as opposed to purchasing them?

This isn't just a question for Bruce, it's for Leon and any others that are producing their own boards.

C.W.

Leon
02-06-2012, 09:25 PM
The main reason for me is the turnround time - I can make a board in about 30 minutes, not including drilling. Low-cost services like ITead Studio (10 small boards for under $10) are very good value, of course, but it takes two weeks for me to get the boards, unless I pay a lot ($30 or so) for express delivery. I can get a board made in 24 hours by a company here in the UK, but it's expensive, and still takes three days to get to me.

idbruce
02-07-2012, 12:16 AM
@C.W.

Definitely turnaround time. I will side with Leon and say 30 minutes, and that is being quite generous depending on your exposure setup. Now lets take a simple power supply circuit for example, and lets assume it has pcb mounted transformer, a full wave bridge rectifier, a smoothing capacitor, and a couple junction blocks. Now let's also say that it might take 45 minutes to draw a nice board, and once again that is being generous. Now let's add another 15 minutes to solder it all together. Providing you have all the materials, you can have a fully functional power supply in an hour and a half. Plus it gives you the ability to test different circuit scenarios rather quickly, but I guess that is also turnaround time. Attached is a photo of a power supply board that I would like to start making, the transformer is not PCB mounted, but this board could be made very quickly and it is a very nice concept.

@deadwhale

Thanks for your input. I tried vellum for quite some time, but eventually ended up abandoning it's use. I now have a mask made of matted film and it holds the toner quite well, especially with a clear coat on top, and it is also fairly opaque. I have not tested it yet, but I will very shortly.

Bruce

idbruce
02-07-2012, 04:36 PM
For Those That May Be Following Along

The LaserJet 6L uses a C3906A toner cartridge and apparently this cartridge also has something to do with Canon printers. Here is a link to a document about cleaning, refilling, and the theory of operation pertaining to the C3906A.

http://www.uninetimaging.com/downloads/technical/TecArtWebAdded/Canon_AX_Toner-Summit_Web.pdf

Bruce

idbruce
02-10-2012, 10:44 AM
Since I have come to the conclusion that I will never achieve the detail that I desire with a LaserJet 6L printer, I am abandoning the idea of altering it.

Bruce

Gadgetman
02-10-2012, 11:44 AM
What it has 'to do with Canon' is that it's Canon that made the 'internals' of the printer.

A more modern laser may be able to give you better detail?

idbruce
02-10-2012, 12:06 PM
@Gadgetman

I am really not sure what it has to do with Cannon, perhaps they make the laser cartridge and just package it for HP.



A more modern laser may be able to give you better detail?


It probably would, but if I needed multiple traces less then 0.008", I would probably just have a laser photoplotted mask made. For simple circuits and boards, with traces greater 0.006", the LaserJet 6L will work just fine with a special laser film and a clear coat, and will probably work well for most hobbyists. However, there is also problems associated with very small text and laser printed photomasks.

On the other hand, if you want or need very fine detail, laser photoplotting or the digital to film process is definitely the way to go.

As a result of what I have learned this week, if I need very fine detail, I will order a laser photoplotted mask, and in the meantime, I am also going to attempt to build my own small little laser photoplotter for boards 3" X 4" or less.

Bruce

Phil Pilgrim (PhiPi)
02-10-2012, 05:58 PM
The guts of HP LaserJet printers have been made by Canon from day one, starting with the Canon CX. The LaserJet 5L and 6L use the Canon AX print engine.

-Phil