Homebrew_PCBs

Hi guys,
I realise most people in this group are using toner-transfer, but I did notice
a few folk talking about the more traditional photo-etch based technologies.
Hopefully a few of the more experienced folk might be able to answer my
questions...

I've pretty much given up on TT due to the poor yields (and having wrecked the
non-stick coating on more clothes irons than I care to count!), especially on
double-sided boards (usually the two layers won't align properly).

I managed to find a used UV exposure unit (a Mega AZ210, the smallest of the
double-sided vacuum units) on greed-bay for "not very much money". It weighs
an absolute ton, but appears to work pretty decently. That is to say, it
beeps, counts down, and the "Lamp Failure" light is off.

At this point, AIUI, I need a few more things:
- Photoresist PCBs
- Developer chemistry
- Another developer tray or two (not hard to find)
- Some OHP transfers (I have some Stabilo inkjet ones, but they've been on
my shelf for about five years and are in pretty poor condition)

I'm within a 15-minute drive of the Farnell trade counter, so ideally I'd like
to get as much from there as possible (with the possible exception of the dev
trays, which may well be cheaper at Maplin). What I am stuck with is a "too
many choices" problem.

Farnell appear to stock most of Mega's Microtrak range, which is
(positive-acting and uses a metasilicate developer), and a couple of other
more standard pre-coated boards (from Kelan and CIF) which develop using the
old standby of diluted sodium hydroxide. Other than that, the only difference
I can see is the shelf life -- six months for the Kelan boards, a year for the
Mega ones, or five years "guaranteed" for the CIF boards.

At the moment, I'm leaning towards using Microtrak (on account of the
less-harmful developer), but the long shelf life on the CIF boards makes them
a bit more interesting. Has anyone tried any of the above products? I did
notice Mike Harrison (of Mike's Electric Stuff fame) singing the praises of
Microtrak on his "how to make really good homemade PCBs" page.

As far as OHP transfers go, I've got both a laser (Panasonic KX-P7110) and an
inkjet (Canon iP4600) printer. Which one is likely to produce better results
for PCB work? On the one hand I'd expect the laser to produce more opaque
blacks (at a cost of worse linear accuracy -- e.g. roller slip and paper
shrinking in the hot fuser), but the inkjet to be more accurate (at a cost of
slightly more transparent ink)...

Lastly, I'd like to know if anyone has any hints on aligning the transfers for
double-sided work. I've tried the "align on a lightbox and staple three edges"
method with toner-transfer, and somehow the top layer always ends up at least
20 or 30 mils out of line with the bottom. Would I be better off adding
alignment holes to the master, drilling these to start, then aligning the
transparencies on a lightbox and Scotch-taping them down?

Apologies if you're all getting tired of my questions, it's just that I'd
rather like to get this right, and waste as little FR4 as possible!

(And yes, I'm planning to get a metal shear of some description to cut the
PCBs. It's just that I've spent about 80% of my budget on the UV box!)

Thanks,
--
Phil.
ygroups@...
http://www.philpem.me.uk/

I'm going to mention a few things that I have mentioned before:

I used translucent Polyester paper. I don't have a source right now. Toner adheres very well and is opaque.

I did use targets and alignment holes at the edge. I used thumbtacks to align the masks on the board.

I did spin coat the resist. The alignment holes do create a problem with the spin coating because they were made before the spin-coat not after.

I just used the surface tension of water to hold the polyester paper in place with the alignment Thumb tacks for registration.

I would image that there is room for improvement: For instance using a sheet of acrylic and a PC drill to create the alignment holes and drilling under a safelight and doing the making of the alignment holes after coating the boards. You could probably find some machined pins for alignment too.





















[Non-text portions of this message have been removed]

Philip Pemberton <ygroups@...> writes:

> As far as OHP transfers go, I've got both a laser (Panasonic
> KX-P7110) and an inkjet (Canon iP4600) printer. Which one is likely
> to produce better results for PCB work?

I've found that ink blocks UV better than toner (8 steps vs 3 for
toner). I've heard that "photo" inks have an extra UV inhibitor in
them, but I haven't tried mine to see if they work better than black.
Inkjet is the way to go.

The tricky part is getting films that will take enough ink. I ended
up getting some Jetstar and Silkjet films which work pretty well.

--- In Homebrew_PCBs@yahoogroups.com, Philip Pemberton <ygroups@...> wrote:

>

I have used only photo-etch so hopefully I can answer some of your questions.

> I managed to find a used UV exposure unit

I have always made my own. My first and second were wooden boxes, but my last two have been somewhat different.
One of them was in an old flatbed scanner housing, one of the old deep ones. Got four tubes, starters and ballasts from Mega directly (they take phone orders with credit cards). It worked fine but was not perfect.
My most recent one is a dozen UV leds (mounted on veroboard to start with, its first board was what is now its main circuit board; talk about bootstrapping! ) driven by a spare 12V PSU (modified PC PSU, cost three quid from ebuyer.co.uk).

> - Photoresist PCBs

I get mine direct from Mega. I use the economy board (stock code 400-040-1 for SS, 400-054-1 for DS). Yes the microtrak stuff is better for fine work, but to be honest unless I am doing something with MBGAs or TQFPs on a 2mil pitch, the economy is stuff is good enough.

> - Developer chemistry

Best VFM I have found is 4007V00 Seno, again from Mega. Keeps forever in power form and last weeks when made up. Needs to be heated to about 30 deg C for best results. I make 1 litre from each plastic package, and when warm works in seconds.

> - Another developer tray or two (not hard to find)

Maplin, Farnell, gardening centre (yes really) and almost any photographic supply shop.

> - Some OHP transfers (I have some Stabilo inkjet ones, but they've been on

Now here is the bit that is my big secret; I use a different type of printer, specifically a Xerox Phaser 8400 (now the 8550). It uses four solid wax blocks (one each black, cyan, yellow and magenta) to lay down colour output, but I use it in B&W mode. I get 1200 DPI output that is so solid I have had pros ask me what photoplotter I used!
Yes the ink costs a lot, and the basic printer is not cheap (about 500 quid now, but comes with network interface and postscript as standard) but the output beats inkjet,laser and the various dye sub printers I have tried. As a bonus I get great high colour magazine quality gloss photos at the rate of 12 pages per minute!

The film I use is a cheap knock off we had at work. When OHPs went the way of the dodo they threw it all out, so I scooped it up and have been using it for the last ten years with no problems. I think it might be made by HP (the part number is an HP one) but the box in basically unmarked.

> Lastly, I'd like to know if anyone has any hints on aligning the
> transfers for double-sided work.

I have a special jig. A friend spotted an article in Elektor about ten years ago and we developed the jig from that.
Basically it is two panes of glass, each held in an MDF frame. The frame is about four inches around each side of the glass sheets. I got some toughened, non-coated glass from a local glazing firm, he cut the MDE and we glued and bolted the whole things together.
Each of the top corners, and in one place along the bottom edge has a bolt protruding from the bottom frame. On the top frame there are slots cut, and by carefully moving the top frame with respect to the bottom one film attached to the glass can be aligned very accurately. I have done stuff with tracks down to 6mil with no problem.

Is it OK to post photos here?
If so I will post up a few pictures of the frame

> (And yes, I'm planning to get a metal shear of some description to
> cut the PCBs.

A wise investment, and one I made many years ago.
Of course if you can get an old paper guillotine, one with a long blade on one side that will work just as well. I know this as the same friend who made the jigs with me found one for a quid in a second hand shop, sharpened the blade and has been using it ever since!

I hope the above helps.
Please feel free to ask for more details on anything I have might have been vague over.

Now comes my request for info.
I have a number of ideas I want to try and would like to know if others have had any success.

First I want to try the heat transfer to do silk screens. This site says they work (http://www.pcbfx.com/) but has anybody here tried them? I do happen to have the laminator they talk about so I was thinking of investing a set of the green ones (the top of the board I use is white) as they do not do black despite the photos on the site showing black results!

Second I am interested in through hole plating. I know there are various mechanical work arounds (panel pins, favorits, etc) but I would like to try and make a real through hole plating machine.
I have some heated pressure rollers (they goto about 200 C) and have got a working control board for them. What I need to do is figure out how to melt solder and flow it over the board. My best idea is to get some of those ceramic heating elements (from hair tongs) and line a hopper with them, then use gravity to get the molten solder to flow onto the joint between the rollers, feed the board into that and catch the drip (and the board) on the other side. Has anybody tried this before?

Lastly (for now) I want to build an SMT oven. Again inspired by a design in Elektor I think one would be possible, but I intent to cheat a bit as I will be using an very different oven, one made for pizza!
It is thin and flat with massive heating elements both top and bottom. Do you think it will be better than the usual toaster oven?

Sorry for the long post, but I know I have a lot to learn, so all pointers most welcome.

DJ Delorie wrote:

> I've found that ink blocks UV better than toner (8 steps vs 3 for
> toner). I've heard that "photo" inks have an extra UV inhibitor in
> them, but I haven't tried mine to see if they work better than black.
> Inkjet is the way to go.

What do you mean by "steps"?

> I have a special target footprint I use on my boards - four square
> pads with a 1 mil gap between them. Since I use negative prints, this
> results in a big sqare hole in the ink with a thin crosshair in it. I
> pre-drill registration holes in the pcb, and line up both sides' films
> using the targets.

I've pretty much decided on Microtrak after spending most of the evening
digging through MSDSes and datasheets. That's a positive-acting resist (i.e.
on the transparency, black = copper, clear = blank FR4) which is developed in
metasilicate.

I did get a reply from Mega, basically the difference is mainly down to
Microtrak being an "inherited" product. Fotoboard is roller-coated, Microtrak
is dip-coated. Other than that, the FR4 laminate is basically identical. The
shelf life is a guaranteed minimum, not an absolute, so theoretically
pre-coated MT should last a while, but will get harder to expose and develop
as it ages.

This should be a fun week. :)

--
Phil.
ygroups@...
http://www.philpem.me.uk/

Philip Pemberton <ygroups@...> writes:

> I've found that ink blocks UV better than toner (8 steps vs 3 for
> toner). I've heard that "photo" inks have an extra UV inhibitor in
> them, but I haven't tried mine to see if they work better than black.
> Inkjet is the way to go.

I use Yellow ink, Works better than Black ink. Also if you have a printer that lets you increase the amount of ink is good too.


Ben

"Ben" <bhleavi@...> writes:

> I use Yellow ink, Works better than Black ink. Also if you have a
> printer that lets you increase the amount of ink is good too.

Just a few days ago I posted a link to a program that lets me put down
as much ink as I want, it's more a matter of how much can the film
take without smearing.

Yeah, I need to find time to try all four inks to see how they rate.

Greetings to all, I am a new member and I must say I am very impressed with the wide range of knowledge you all have. My experience with homemade PCB's is limited to single sided boards made using the photo-etch technique....there are many other techniques that I plan to try, but if I can help in any way, I will. I've recently moved from Chattanooga to Colorado Springs, so I haven't set up my circuit workshop, but I am thinking about a whole new approach. With that said:

>
> Second I am interested in through hole plating. I know there are various mechanical work arounds (panel pins, favorits, etc) but I would like to try and make a real through hole plating machine.
> I have some heated pressure rollers (they goto about 200 C) and have got a working control board for them. What I need to do is figure out how to melt solder and flow it over the board. My best idea is to get some of those ceramic heating elements (from hair tongs) and line a hopper with them, then use gravity to get the molten solder to flow onto the joint between the rollers, feed the board into that and catch the drip (and the board) on the other side. Has anybody tried this before?

I have not tried this, but I have thought about it a lot. My thinking was to use some sort of temporary solder mask(perhaps peel-able or regular) , mask off the entire PCB except where the via's would be, then dunk the entire board into molten solder, dunk into a solder pot like the 2000W unit sold by jameco.com (part# 149876), using some sort of ceramic tongs to hold the board. If this method worked you would be able to make through holes on PCB's whose dimensions would be less than 12" by 8"(a lot of circuit can fit that area).
My greatest curiosity to the above mentioned idea, is to how well would the joints hold up if one was to then reflow the board with components (I presume the surface tension of the solder would keep the joints sound...but only trials would prove the concept.)

--- In Homebrew_PCBs@yahoogroups.com, "Daniel Howard Bryant" <coloradobryant@...> wrote:

>
>
> I have not tried this, but I have thought about it a lot. My thinking was to use some sort of temporary solder mask(perhaps peel-able or regular) , mask off the entire PCB except where the via's would be, then dunk the entire board into molten solder, dunk into a solder pot like the 2000W unit sold by jameco.com (part# 149876), using some sort of ceramic tongs to hold the board.

OK I had not thought of something like that. How would you attached the masks and what would make them out of?

> If this method worked you would be able to make through holes on PCB's whose dimensions would be less than 12" by 8"(a lot of circuit can fit that area).

The ones my contractor does with his electro-plating machine are all sizes from about 5mm square up to 400mm square.
I would hope my idea would enable me to do the same.

> My greatest curiosity to the above mentioned idea, is to how well would the joints hold up if one was to then reflow the board with components (I presume the surface tension of the solder would keep the joints sound...but only trials would prove the concept.)

Yeah good point. I will try and find out how the pros do this.
The contractor we use at work has a electro-plating line to do PTH, and I know he uses a reflow system, but it is very different to a toaster oven (lots of blets and flow solder baths).
I would love to try to make an electro-plating system but I am not sure how to go about it so, based on what I have seen for sale by Mega I came up with the pressure roller idea.

Hi, don't take me wrong i don't want to kill both your enthusiasm, but
forget the molten solder stuff. You will never do through holes with it.
And even for finishing purposes is a thing of the past. Most machines
used for that are forgotten on some corner or already in the junk for
various reasons. Roller tinning i have seen it foolishly buy new and
putted in the corner after first run (seller didn't accept it back).

PTH, is a chemical process of desmear + activating holes + copper
plating. There are almost 10 chemistries for that. Tin finish is done by
electroless chemistry. There are at least 4 very simple chemistries for
it.

--- In Homebrew_PCBs@yahoogroups.com, Simao Cardoso <simaocardoso@...> wrote:

Some people have a good gut feeling for what will work and what
probably won't work.
Anyone with soldering experience has probably made the observation
that surface tension of solder is incredibly high and it would not
stay in the hole, certainly not for the also proposed reflow.
Filling the holes with solder alone is a dead end idea, but feel free
to prove us wrong.

BTW there used to be copper plated solder wire, the copper layer was
scored in short sections. You'd stick the wire in the hole, break off
a section, and then rivet it both sides. Then you'd desolder the
solder out of the hole, the copper plating would stay leaving a thin
hollow rivet similar to a PTH hole (but much inferior of course).
That's all the way solder is going to go for making VIAs, in my opinion....

ST

On Fri, May 15, 2009 at 1:00 PM, Daniel Howard Bryant
<coloradobryant@...> wrote:

Regarding that, I have a CD-ROM PCB somewhere that has just that. The
vias are filled with something that looks like conductive paste and
covered with some other laquer.
It is the only PCB I've ever seen like that, but it was a production part.
If I stumble across it again I'll make some photos....

ST

On Fri, May 15, 2009 at 1:00 PM, Daniel Howard Bryant
<coloradobryant@...> wrote: