Langevin AM-16 DIY?

[t-wurst]

hey guys,

i´m working at a printers and the procedure here is like the same as for our printing plates. the layout has to be mirrored because the printed surface of the paper/foil must directly contact the copper. the light doesn´t have to pass the foil before it lights the copper and on the other hand no light from the outside can fall under the traces and cause an unsharp borderline.

for the filetypes try to use vector formats like eps. other filetypes like bmp are pixel-based, so there are a few bits of information about color, lightning and so on for just every pixel in your file even if its blank. and when you need at least 600 dpi (better 1200dpi) to get a sharp picture - you can imagine that this blows up the filesize.
vector format files (like .eps) include just a description of the layout ie. "write a black line from x to y." often it contains a low res image (72dpi) of the layout just for viewing and placing on your screen - so its really smaller

try to export or save as EPS from your PDF or use the GIF format...

hope you understand my little english

cheers,
toby

 

[Gus]

I saw something on the web the other day about mercury electronics making a Clone of this. I wonder what transformers and who makes them if this is true. Chris any ideas?????

 

[CJ]

Good info Toby! Your English is better than mine!

Gus, I saw the Mercury stuff at AES but the lids were on. I did not know that someone besides Ollie was making V76 type stuff.
Here's a shot of their rack at AES:

 

[Kev]

[quote author="t-wurst"] .... the layout has to be mirrored because the printed surface of the paper/foil must directly contact the copper. ... [/quote]


I don't have trouble
I've been doing this a very long time
my light box has a lid so outside light is not an issue

Yes there is thickness in the films but it is not a major stumbling block.
many printer control panels have a mirror function if you need it

I make great circuit boards that have been in continuous 24/7 service for a decade and more.

whatever method you find works for you is fine by me. :thumb:

 

[Winston OBoogie]

...

 

[CJ]

Cool Winston!

Ok, I mirrored the pdf's that Duka has done. This is in case you want to use vellum/tracing paper for your artwork.

One file is for making a single board, and is scaled 1:1

The second set of files is for making 4 boards on a six by six sheet of double sided copper. In order to get 4 boards on a piece of six by six, I rescaled the width only to 96 percent, not enough to cause any problems, but small enough so thatr you can get a 1/16 inch saw blade or mill bit in between the boards to cut them out. Sorry about the file size, but everytime I mess with parameters on Illustrator, I am afraid that all the bugs in that program will corrupt the files, as has happened many times before.


Interesting trace coming off the pot, iit goes nowhere so it must be for heatsinking.
This is the original trim pot version.
Let me know if anyone is interested in a six by six mirrored versions of the Phifer pot.
I have not burned these in real life, so double check my work if you want to go for it.
cj

 

[dukasound]

Guys
I am owner of one non functional AM16 :grin: :?
My best man who buy it for me will send me Langevin around Chrismass.
Duka
Thanks to guy who send link e-bay :thumb:
I saw that was ytrehalf :sam: :sam:

 

[t-wurst]

[quote author="cjenrick"]Good info Toby! Your English is better than mine![/quote]

Thanks cj. Its all DIY :green:

[quote author="Kev"]I don't have trouble
I've been doing this a very long time
my light box has a lid so outside light is not an issue

Yes there is thickness in the films but it is not a major stumbling block.[/quote]

Kev you´r right. For pcb layouts its less important cause of the wide traces. But for printing you handle with very small raster dots, so the light from beside will cause a smaller dot and also a wrong color image.
Its not the light from outside, its the light from your uv-lamp that fells "under" the printing from the sides because of the thickness...

but no matter, not very important for PCB - just for information :sam:

toby

in the example above, i use a common 2 cm thick paper/film :green:

 

[CJ]

Give me 4 AM-16's to go. Hold the mayo:

Duka got a few minor tweaks, otherwise, looks good!

I will point them out on the pdf.
Hold on.
Looks like Illustrator corrupted a few things on the files. Your's is probably alright.


cj

 

[CJ]

OK,

Tweak 1: Those large transistor holes need to be aligned a little better. They should be 0.518 in. up from the bottom edge of the board.
They are located at 0.90 and 1.72 inches from the left edge of the board.

I think that's about it. You could actually use your boards the way they are, this is only a mechanical issue, not electrical.

I accidently mirrored the original trimmer on mine with the phipher on the otherside.

I pulled the pdf's I posted til I fix it. I can still use the boards I just made, just some minor trace scraping.

 

[CJ]

OK, I found something else:

Tweak #2: On the top board, the area in blue rectangle should probably be applied to the area in the red rectangle, as they are symetrical on the original board.

On the bottom board, the area in the rectangle should have six pads, like the red pads shown.

This should be applied to both versions, the original trimmer and phipher trimmer.

cj

 

[dukasound]

[quote author="cjenrick"]OK, I found something else:

Tweak #2: On the top board, the area in blue rectangle should probably be applied to the area in the red rectangle, as they are symetrical on the original board.

On the bottom board, the area in the rectangle should have six pads, like the red pads shown.

This should be applied to both versions, the original trimmer and phipher trimmer.

cj

[/quote]

Hi Chris
I agree :thumb: You are BOSS :grin:
I tried to put transistors right between their three pads. Also I wanna make same raster for R13, R15 and R14, R16.
I agree with two missed pads. I lost them when I exported from Corel
There is no parts connections on that side.
If I had scanned boards this job would be exelent
Best
Duka

 

[CJ]

I think we are done. Anybvody know how to turn a pdf into a gerber?
:guinness:

 

[CJ]

Who is MCS llama breath?
:razz:
Dang, that thing must be on Red Bull.
When is it gonna run out of steam?
:green:

 

[tommytones]

I saw that wild beast monster on a DIY CNC machine group a few days ago. You're getting popular because of that thing :green:

 

[CJ]

Oh, that MCS!

They ought to come out with a new energy drink called Green Llama.

 

[Consul]

So is the LSD Marshmallow Llama our official mascot, then? :green:

 

[hitchhiker]

maybe we need a sweat lodge!

Lance

 

[PRR]

> the layout has to be mirrored because the printed surface of the paper/foil must directly contact the copper. the light doesn´t have to pass the foil before it lights the copper and on the other hand no light from the outside can fall under the traces and cause an unsharp borderline.

> I don't have trouble

You are both right.

Best (sharpest) results happen when the image-side of the transparency is in direct contact with the photograpic paper. That's how I did photographic contact-printing, and that's how we did newspaper printing-plates. The same is best with PCB mask transparencies (photographic or xerography) and photo-treated copper.

Toby's excellent illustration shows the problem. We usually have a big light source close to the negative. Many of the rays hit at a slant. If the image is on top of the transparency, light leaks under the edges. Even if we had a point-source light at an infinite distance (the Sun was popular in early photographic printing, a small lamp behind a long lens is still used in photographic projection printing), the transparency isn't perfect and bends some rays, also diffraction will cause a "rainbow halo" under the image.

BUT: while in photographic printing we need every skin-wrinkle and tree-branch clear, in PCB printing we can be very sloppy. A trace can be 50% wider or narrower than the mask and still work fine, as long as it is not too narrow to carry current or so wide that it touches another trace.

So Kev working for many years with big parts can project through the back of the transparency, get 0.005 inch blur around a 0.1 inch trace, and get perfectly good results. If Toby's print shop did that for magazines or books, text would look un-sharp, and "halftone" dot-images would "block-up", all the dots run together. On modern very small trace PCBs it might bleed-over enough to cause shorts or opens or simply look bad to a paying customer with a sharp eye.

In this case, we have a very wide-open PCB layout. The original was probably done freehand with pen-and-ink and/or stick-on tape and dots. You could probably tape-up an old piece of wavy window-glass, project the north sky (south sky for Kev) through the wrong side, and still get a functional board.

It is good to put some text on any printing transparency. Printers are used to thinking forward and backward, and coming out right in the final product. When I made newspaper printing plates, I made sure they came out backward. For me, the plate was the "final product" (my job was done), but really the plates were inked and rolled on paper, and that paper had to be readable! There are some many-stage processes where it can become confusing. But in this case the PCB is the final product, and the printer will lay the negative so the text comes out forward.

 

[Kev]

thanks guys
let me say again

Basic Protel and CM printing for the bottom layer.
Print is on top of the transparency and this comes in contact with the pre-sensitised PCB.

When placed into MY light box with the foam to help keep it all together and still ... it is very dark inside until the UV gets turned on by the Micro controlled timer ..

I get nice bottom layers.

When I turn it all around to do top layers, it is as stated above that, the ink is now one transparency thickness away from the pre-sensitised PCB board. Yes
can lead to fuzzsyness.

Most of your guys don't do double-sided PCB's.
More than half my PCB's are double sided and mostly use DIL IDC ribbon style connectors.

Sometimes we have to bring a track between these pins and also between the pins of the typical IC.

Our top layers are often not as complex as the bottom layers but a few have just as much complexity as the bottom and the number of vias is over 100. I don't want to really know how many there are ... all I do know is that it takes me more then one shift to do a single board of vias

For those that are interested this board is a stereo balance 16 X 1 audio switcher with multiple balanced outs ... one set having our favourite VCA for remote Dim / Cut / Mono and volume. It is quite large and needs nearly a full sheet ... A3 sized.
:green:
smug mode

back to the point,
some print drivers have reverse mode at the print stage and I'm sure if you dig deep the PCB layout software will also have this.

last note
my PCB's always have some text and some well spaced registration holes to keep everything the correct way around and right way up.

EVEN THEN I some times stuff it up and once my work-mates let me drill the last hole before telling me so.

arrrgh

I got them all back ... eventually. :twisted: