Method to Make Project Panels (Warning: Detailed Info!)

[wy6k]

Not too sure how you come up with the $200 figure. If you wanted to you could get the transfer paper and a single color foil for about $25. Another other 10 or 15 for the adhesive. The kit we purchased cost under $100.

Oh that's not bad. I thought I saw a "starter kit" type of thing that was $200 plus an exposure light or something. Do you need a special light? Or any other special gear, or just the material?

Thanks,
Mike

 

[3v0]

Oh that's not bad. I thought I saw a "starter kit" type of thing that was $200 plus an exposure light or something. Do you need a special light? Or any other special gear, or just the material?

Thanks,
Mike

You need a laser printer. A laminator is very nice to have. A heat gun to dry things, but I plan to try a hair type blow drier.

There are no photographic steps. It is done by having things stick and release one from another. Quite clever actualy. The owner/inventor has put
a lot of R&D time in figuring out how to make it work.

There are two methods in the one kit. One is for working with foils which only requires a b/w laser printer. The full color decals require a color laser printer. But they are very impressive.

I do not have a color laser but may be gettin one in the next year.

 

[wy6k]

You need a laser printer. A laminator is very nice to have. A heat gun to dry things, but I plan to try a hair type blow drier.

There are no photographic steps. It is done by having things stick and release one from another. Quite clever actualy. The owner/inventor has put
a lot of R&D time in figuring out how to make it work.

There are two methods in the one kit. One is for working with foils which only requires a b/w laser printer. The full color decals require a color laser printer. But they are very impressive.

I do not have a color laser but may be gettin one in the next year.

Well that sounds pretty good. I have a color laser printer and several heat guns too. I don't have a laminator, but have wanted one often enough to think about getting one.

Thanks for the info!

Mike

 

[crashsite]

More thoughts

I posted about this earlier but I do not think you looked at it. Frank at PulsarProFx/DecalProFx has put many hours into creating good looking transfers.

I had seen this. But, there is a "philosophy" to something like this as well as the mechanics of it. Part of my basic makeup is cheapness (I've been accused of squeezing snot out of Lincoln's nose while pinching a penny) so I like the near-zero cost approach of using a sheet of regular printer paper, a pass through the printer, some clear tape and a spritz of spray adhesive. Oh, yeah, and the extragagance of the corner of a sheet of 600 grit sandpaper to bust the glaze off the tape.

I must also confess to my inherent laziness so, the procedure also suits that vice as well.

I do like the idea of using the adhesive, clear lable material for larger area panels (and other similar projects). But, one problem with the sheet material is finding a width that will span a 19" panel in a single swath. For 1.75" H (one unit) panels, the 2" wide packing tape still works. For 2 or 3 unit heights I can do it with the 6" wide roll of tape I have. I do have a roll of laminating film that I've used for doing wallet cards but, haven't tried for panels.

I do like the durabiltiy of the plastic over the panel layout.

I haven't tried, but am intrigued by the map guy's idea of soaking off most of the paper to make the "sandwich" thinner.

I've also been kind of keeping my eye out for some thin, very white, sheet plastic (maybe vinyl or mylar) that will take the toner to give an even whiter, more paint-like look to the white lettering and trim.

Hmmmm...I think I may be getting that mad-scientist look in my eye again. I just remembered that my Breyer's lo-carb ice cream bars come wrapped in a thin, white, almost pearlescent plastic. Maybe I'll spend the afternoon attaching this material to a carrier sheet and try printing on it with old laser printer. I wonder if Igor is busy (I'll need someone to throw the switch at the height of the storm).

 

[3v0]

Everyone is free to do as they please.

I used to be a horrid penny pincher but have slacked on of that somewhat. Too much time getting the cheapo methods to work.

Once I switched to the pulsar starched paper my PCB were so much better that I have never looked back. At a buck and a half a sheet it is worth every cent and more.

But, when I make PCBs I sometimes save the transfer paper I used for the foil pattern. When it has dried I reuse it to transfer the "silkscreen" layer for parts placement.

EDIT:

 

[wy6k]

Everyone is free to do as they please.

I used to be a horrid penny pincher but have slacked on of that somewhat. Too much time getting the cheapo methods to work.

Once I switched to the pulsar starched paper my PCB were so much better that I have never looked back. At a buck and a half a sheet it is worth every cent and more.

But, when I make PCBs I sometimes save the transfer paper I used for the foil pattern. When it has dried I reuse it to transfer the "silkscreen" layer for parts placement.

EDIT:

A lot of the people here seem to use iron on PCB techniques.

I'm curious why you do this rather than just print the pattern on a transparency and then expose resist coated boards?

Thanks,
Mike

 

[crashsite]

Memory Lane

A lot of the people here seem to use iron on PCB techniques.

I'm curious why you do this rather than just print the pattern on a transparency and then expose resist coated boards?

I can give at least one answer to that. Years ago, I used to do the photo resist thing (KPR and Xylene and the positive resist and lye solution developer). It worked but required a good exposure source (I've used many, including the sun, tanning lamps, commercial arc lamps, Diazo exposure boxes (of which I have a few around), etc.) and the procedure is rather time consuming. If you coated your own boards (as I did) it was tricky. I had built a spinner that applied the resist by centrifugal force (It's still around here somewhere). Then there was the mask. Some here claim good results from a toner-on-clear backing but, I always found the toner to be too transparent and the exposure pretty iffy. I used to use my darkroom enlarger as a process camera to expose the films for the masks (Kodalith and Kokak's, D-8 chemestry). The taped pattern didn't work with a negative resist, like KPR and required the photographic step, although the positive resists did (if you were careful about not letting light leak in under trace and pad junction points)...and on it went.

I've also played around with silk screens but, you have all the same tedium as with the photo resist on boards, plus the painting.

The toner resist presents its own problems and quirks but, it's a lot easier and, with the advent of the laser printer, very condusive to computer generated layouts (rather than the old taped patterns).

For experimental and onesy-twosy boards, the toner method is a good choice. For small-scale production work, probably the best method is to generate silk screens and screen the boards. For larger scale production, I don't know what the usual methods are anymore.

 

[wy6k]

I can give at least one answer to that. Years ago, I used to do the photo resist thing (KPR and Xylene and the positive resist and lye solution developer). It worked but required a good exposure source (I've used many, including the sun, tanning lamps, commercial arc lamps, Diazo exposure boxes (of which I have a few around), etc.) and the procedure is rather time consuming. If you coated your own boards (as I did) it was tricky. I had built a spinner that applied the resist by centrifugal force (It's still around here somewhere). Then there was the mask. Some here claim good results from a toner-on-clear backing but, I always found the toner to be too transparent and the exposure pretty iffy. I used to use my darkroom enlarger as a process camera to expose the films for the masks (Kodalith and Kokak's, D-8 chemestry). The taped pattern didn't work with a negative resist, like KPR and required the photographic step, although the positive resists did (if you were careful about not letting light leak in under trace and pad junction points)...and on it went.

I've also played around with silk screens but, you have all the same tedium as with the photo resist on boards, plus the painting.

The toner resist presents its own problems and quirks but, it's a lot easier and, with the advent of the laser printer, very condusive to computer generated layouts (rather than the old taped patterns).

For experimental and onesy-twosy boards, the toner method is a good choice. For small-scale production work, probably the best method is to generate silk screens and screen the boards. For larger scale production, I don't know what the usual methods are anymore.

Thanks for the info, Crashsite.

I only make onesy twosy stuff to suit myself so I don't care about higher production rates.

I use a laser printer to print on a transparency (computer can do positive or negative) and use precoated boards from MG Chemicals along with the developer they supply (have forgotten what it is). I get great results - however, I've been using 20 mil traces and spaces. I really don't know if I could get 10 mil traces and spaces to work - I haven't tried it (all the stuff I've been doing is analog). What kind of resolution can you get with the iron on stuff? Can you do 10 mil lines? Or better?

Thanks,
Mike

 

[crashsite]

Laser Traces

What kind of resolution can you get with the iron on stuff? Can you do 10 mil lines? Or better?

That's a difficult one to answer. I've had pretty good luck with 20 mil traces on the stuff I usually do but, have found that the edges of the traces are not real smooth. Because of that, I tend to avoid finer traces. Othere here have probably had different results (different printer, transfer medium, ironing methods, heat and pressure, board material, etc.).

Also, I have a fair amount of PCB stock (older stuff) that I think tends to have heavier copper. That's great for higher power stuff but, has "undercutting" problems duiring etching (that is exacerbated by narrower traces).

That's one of the great things about a forum like this. It lets people try stuff and then share their results (much as you have done with your panel techniques).

Of course, the fineness factor is not a consideration with the panel work. There, if it looks good it is good. With the PCB stuff it's only good when there's good trace continuity and no bridging.

 

[3v0]

With the good transfer paper the line size seems to be limited by the ability of the laser printer. I have an inexpensive HP1020 and often use .012 lines. I can do .010 but have to take a bit more care.

As crashsite suggested thinner copper undercuts less. But this and maybe a few other parameters are common to all etching regardless of how you apply the resist.

The site shows examples of .006 lines, I think this was done on 1/2 oz copper.
**broken link removed**

I use a cnc mill to drill holes. Up to now I have used surplus PCB stock and do not know exactly what the copper thickness is/was. Last this year I plan to be doing more surface mount work and may order new 1/2 oz PCB stock.

Toner transfer can be quick to do. If I have a pre drilled PCB blank and a hot laminator I can go from computer artwork to an etched board in a few minutes.

For etching pulsar has a green foil that you can optionally laminate to the toner. It makes a PCB-toner-foil sandwich. The foil bridges pinhole sized gaps in the toner. For me this is a step I can often skip unless my toner cartridge is over about 1/2 used or if I am doing lines under .012.

For me cost is a factor. If I make anything for school I need 10 or 12 copies. With the surplus board and the pulsar paper the PCB cost is much cheaper then pre sensitized boards.

EDIT:
On factor that has made PCB etching easier is that higher levels of circuit integration and smaller package sizes have reduced the size of the boards we make. A 10 by 5 inch board is about 50 times more likely to have a process error then a 1 by 1 inch PCB. Back in the days of the Z80 I played with photo resist and silkscreen as did crashsite.

IMHO if the average person has to make more then a few identical PCBs, a board house is the way to go. Then you also have solder masks and plated through holes. Drilling holes by hand is tedious, if I did not have the CNC to do that for me I would be making fewer boards at home.

 

[philba]

Can you get UV stable tape or transparency material? Every piece of tape I've ever seen last a couple of years tends to yellow (from UV, I presume). I'd hate to put work into a project and have it slowly get ugly. I know there are spray-on UV blockers for art but they aren't that durable.

 

[crashsite]

Ancient History

Can you get UV stable tape or transparency material? Every piece of tape I've ever seen last a couple of years tends to yellow (from UV, I presume). I'd hate to put work into a project and have it slowly get ugly. I know there are spray-on UV blockers for art but they aren't that durable.

Here's a snapshot of some of the early stuff that I did at least 13 or 14 years ago. They were printed at work on HP LaserJets. As you can see, the toner density was not as good as today's cheap printers (Yes, the density is as it was).

**broken link removed**

I keep these around for just this purpose. To see how they hold up to time. They have never seen prolonged exposure to direct sun but have been in windowed rooms, under flourescent lights and in storage in non-temp controlled conditions. The photo was taken outdoors, in the shade but, under a clear sky so you see a pretty accurate color rendition.

I suppose, if I was doing that job today, on the MIDI fan-out panels, I would take a cotton swab and run a layer of enamel or maybe latex paint around the cutouts for the MIDI connectors (they were to be the 5-pin DIN connecotrs, on a PCB mounted behind the panels). That would add a decorative effect and also protect the tape (which was from my wide roll) where it interfaces with the edge of the holes.

I was doing the MIDI thing on spec to try to get a pawn shop interested in buying the setup as a way to demonstrate the keyboards they were selling...but, they turned out to be "tool", "gun" and "jewelry" thinkers so the task languished.

You can see where the adhesive has dried up in the corners of the smaller panels. Also note that the tape was left shiny on the smaller panels so some of what's seen is reflections. The other panels have the glaze sanded off.

I hope this gives you a better feel for what you could expect. I would think that the lable material would do even better than the tape regarding the concerns you have.

 

[wy6k]

With the good transfer paper the line size seems to be limited by the ability of the laser printer. I have an inexpensive HP1020 and often use .012 lines. I can do .010 but have to take a bit more care.

As crashsite suggested thinner copper undercuts less. But this and maybe a few other parameters are common to all etching regardless of how you apply the resist.

The site shows examples of .006 lines, I think this was done on 1/2 oz copper.
**broken link removed**

I use a cnc mill to drill holes. Up to now I have used surplus PCB stock and do not know exactly what the copper thickness is/was. Last this year I plan to be doing more surface mount work and may order new 1/2 oz PCB stock.

Toner transfer can be quick to do. If I have a pre drilled PCB blank and a hot laminator I can go from computer artwork to an etched board in a few minutes.

For etching pulsar has a green foil that you can optionally laminate to the toner. It makes a PCB-toner-foil sandwich. The foil bridges pinhole sized gaps in the toner. For me this is a step I can often skip unless my toner cartridge is over about 1/2 used or if I am doing lines under .012.

For me cost is a factor. If I make anything for school I need 10 or 12 copies. With the surplus board and the pulsar paper the PCB cost is much cheaper then pre sensitized boards.

EDIT:
On factor that has made PCB etching easier is that higher levels of circuit integration and smaller package sizes have reduced the size of the boards we make. A 10 by 5 inch board is about 50 times more likely to have a process error then a 1 by 1 inch PCB. Back in the days of the Z80 I played with photo resist and silkscreen as did crashsite.

IMHO if the average person has to make more then a few identical PCBs, a board house is the way to go. Then you also have solder masks and plated through holes. Drilling holes by hand is tedious, if I did not have the CNC to do that for me I would be making fewer boards at home.

I'm very curious about what CNC drill you are using?

Thanks,
Mike

 

[3v0]

The high school where I teach has a MaxNC 10 CNC mill. Inside eagle I run pcb-gcode.ulp to generate gcode that the mill understands. I drill the blank boards, apply the toner to match the holes and etch.

It would be nice if there was a low cost PCB drill.

 

[crashsite]

PCB Drilling Solution?

It would be nice if there was a low cost PCB drill.

Well, color me naive but, it seems like building a CNC PCB drill would be just about the easiest CNC drill you could make.

1. There would be no need for speed control as is needed for milling type operations.

2. The precision is less than needed for milling, especially as relates to backlash. In fact, with even something as simjple as usting threaded rod, as you'd buy at the hardware store, running through a threaded hole on the carriage would certainly give you precision as good or better than drilling by hand with a drill press.

3 The Z axis would requiire no precision as it would only need to plunge the drill through the PC board and then withdraw it.

4. Because the drilling operation is a point-to-point affair, where you simply move to an X and Y coordinate and drill a hole, the programming for the machine could be as simple as starting from a reference point and counting a series of pulses to the X and Y stepper motors to position the erill for each hole. Sure, you could do that with computer programming but...why? If you know your current location (based on where you are in the series), all you'd need to find the next hole postion are two numbers and their signs (for direction). The "computer" part would simply fetch the next set of numbers (probably from a ROM), generate the stepper commands and then fetch the next set of numbers to the end of the set.

5. Multiple drilling heads could be employed with no modification to the X-Y carriage system. You'd simply set the count to positon the board under the right head and then use it to drill. In fact, the Z axis would not even need to be and automated operation. Having the X-Y carriage sitting on a drill press table and positioning the board, while you manually pull the handle to drill and then, maybe press a button to have the table drive to the next hole, would be a huge improvement over manally drilling the boards. You culd also change the drill size as needed.

Mechanically, the thing could be built out of angle iron and threaded rod. Yeah, it's pretty coars, I know but, for this application I think plenty good enough.

 

[3v0]

I would prefer that

• it be CNC in XYZ
• it be much more accurate then hand drilling
• once you set the machine up you can walk away from it

I agree that Z could be a up or down sort of thing but it must be under computer control, better if it was regular CNC. No way am I going to hang around running the Z axis.

As it happens I have a machine to use. I would like to see an inexpensive one that would allow others to have one too. If I did not have other things to do I would work on it.

 

[crashsite]

How Good is Good Enough?

I would prefer that

• it be CNC in XYZ
• it be much more accurate then hand drilling
• once you set the machine up you can walk away from it

I would prefer it to be gold plated.

But, it doesn't need to be gold plated...nor does it need prcision to the nth degree.

I call this the "meter resolution syndrome". Ever watch a tech make a 5 V voltage adjustment with a meter? If it's a cheap analog meter, the voltage will be set to 5 V. If the meter has a mirrored scale, it will be set to exactly 5 V, to the accuracy of the meter scale. If it's a 2 1/2 digit cheap digital meter it will be set to 5.00 V. If it's a digital that has 5 place accuracy, the tech is not satisfied until it reads, 5.0000 V. Of course the spec will be 5 V, +/- 5%.

Crap is not good enough but, good enough is good enough and a couple of thousandths is good enough for hole accuracy in home project PCBs (and much better than hand drilling). And, that's a tolerance that's pretty easy to attain without resorting to NASA spec. precision machinery.

As a guide, the nominal thickness of a sheet of 20# printer paper is about 0.003 in. so I just don't believe that a hole position error that's less than the thickness of a sheet of paper will be noticed during lead bending and assembly. Fortunately, with digital techniques, the position error does not "accumulate", when doing a point-to-point-to-point positioning, as it will with analog methods.

I agree that the Z axis would be nice but, for those of us that don't have a CNC machine handy, just the X-Y would be a big help. The Z axis also presents problems. Unless yoiu have rather sophisticated sensing, it's essentially a brute. If the drill flutes pack up with material and the drill breaks, it will merrily continue to "drill" away. Humans can feel and watch the progress and correct for feed speeds and drill clearing. Adding the Z axis also makes the machine substantially more complex and maybe too much so for the guy who wants the accuracy of the holes, provided by the X-Y positioner but, doesn't mind the tedium of drilling the holes for the few boards he acttually produces.

There is another forum (that uses the same format as this one) dedicated to CNC. The problem is that they have the attitude that if it's not commercial grade, they don't even want to think about it. And, while it's kind of disappointing that the "little guy" is left out in the cold, it's probably better in the long run that they don't try to cover too wide a spectrum.

 

[chuckoo]

What i do :

is fairly the same ..

I print with a laser printer , sometimes color or with an Inkjet Printer on printable labels you can get in even A4 ( letter ) sizes. Sometimes also on clearfilm labels. depending on what i want to achieve.

After printing , especially for the Inkjet prints, i additionally COAT the print with adhesive clearfilm.

 

[crashsite]

Thanks for the post

What i do :

is fairly the same ..

The basic process really is pretty basic and it seems like the variations are in the details and this is a good forum to fine tune techniques by hearing what others have done and their success (or lack thereof).

 

[xanadunow]

Common sense..

First of all I thank you all for excellent contribution to this thread (it means, I have learned a lot ). Secondy (and there after), there are few noticeable trends to follow when making descriptive sense of controls fitted to the front panel as this thread describes.

There is a "commercial" solution (let someone else do it) and the "one and only" (DIY). The concern is the price we need to pay to achieve the result we want.

In the commercial world, we should be looking for commercial solution, i.e. if the "production" goes into hundreds of pieces - look for a bussiness that specialises in just this area, they will always be cheaper.

For a "single" project, what we want is "the look of it" versus the "cost of it" solution.

[Side story here:] Some many years back I was contracted by a "Hi-Fi" shop to make a box that would allow to connect "x" number of stereo car radio units to "y" number of boosters/speakers solutions. I had used the "relays approach" to the circuit (simple), but then I did stumble against the problem of the front panel of that box. It had to be presentable to the shop owner and to his clients - after all, it was something his clients would see when he presents all possible configurations he could offer. It was the dot-matrix printer era and for a computer, I was using Commodore 128 (for bussiness) and Commodore C64 for gaming. For a solution to the front panel, I have chosen a photosensitive material I have been already using to make my PCBs. A coat of acrylic (or two) had sealed the work and some 15 years later, the panel looked as good as when I made it but the original labour was quite intensive. I did not make another panel since but I am loking to make a few now and this is why I find this thread very interesting.

Out of all solutions presented here so far, the Make your own custom DRY TRANSFERS in less than 10 minutes with "DecalPRO" sems the most sensible, cost efficient, repeatable and good looking.

No, I do not have any endorsment from them, I am acually crossed off that they would only distribute within US and do not think to expand their sales to other countries like Oz (I do deserve some free samples for that ).

Until they do and following all lessons of this thread, I will laser print either individual labels or whole front panels on thin transparencies and fit them using spray-on adhesievs.

Another method I am inclined to use it a "label printer". I do use it at work to make descriptive names to identify cable numbers, reference designators or whatever is required. These are relatively cheap portable printers (toys like) with 2" x 3" (or so) LCD screen and a keyboard (all in one) and all you need to do is to turn it ON, type the text of a label and hit the print button. You can choose the font/size and a variety of tapes including the transparent with black or white font. These tapes also do come in a variety of widths (1" or less). The final steps of producing your label is to cut the end of the printed string, peel the backing and attach the label to your panel. A coat of acrylic or two may also be used to finish the job. Office Supplies shops stock both, these printers and tapes (Newark and RS Components probably too).

Another aspect not mentioned here to a great extend is the surface preparation of the front panel. Assuming it is made out of aluminum sheet, the initial treatment with 800 and 1000 grade of sand paper should be followed by a short "bath" in the solution of caustic soda (cheap) in a water (wear protective glasses in the process). The end product is a professionally looking matt finish followed with your labelling. There are also other cheap chemicals to change the tint of the aluminum to your content but it is not exactly the topic of this thread.

Regards,
xanadunow