Question Pertains to the ELMP.

In order to make the ELMP, you need to create a Uc component by etching 
and soldering and what not, they kinda give a bit of information on how 
to do this but I was wondering if anyone in this conference would be 
able to go into elaboration.  Thanks...

36 responses total.

Um.  That's a tall order...

Well, I kinda know the process of etching, I was just looking for some 
tips from someone who has experience.  I don't know much about the 
soldering process and any information on that process would be help 
too.  I appreciate any knowledge handed down...thanx.

Soldering is (usually) easy.  It's just a question of using a low temp
melting metal to glue other bits of metal together.  For electronics,
the most important thing is to get the right kind of solder.  What you
want is an alloy of tin and lead called "60/40" (for the ratio of the 2
kinds of metal), and you want "rosin core".  You do *NOT* want "acid
core".  The rosin or acid are called a flux, and their purpose is to
clean off impurities from the metals being soldered together.  Both do
this, but the acid leaves a corrosive deposit behind that will eat
through thin metal pretty quickly, with obvious bad consequences for
electronics.  In any event, the right stuff is readily available at any
good hardware store, meijers, radio shack, purchase radio, etc.  The
solder you buy will come in the form of a thick gauge wire, and the
rosin will be in a tiny hole run through the middle of the wire.  It
won't be visible unless you cut the solder, which there's no reason to
do.

In order to melt the solder, you need a soldering iron.  Originally,
these were irons that you heated up in a fire, but this has long gone
out of fashion, being replaced by electrically heated irons that are
considerably more convenient.  The typical modern soldering iron looks
something like an overly fat pen with a long metal tip and an electric
cord.  Like anything else, having high quality tools makes the job go a
whole bunch easier.  In a soldering iron, you want to look for something
with the highest wattage (the more power it can burn, the faster it can
heat things up, and if you're soldering small things that can be easily
damaged by excess heat, heating things up faster has the paradoxical
effect that you need less heat to solder.)  The other thing to look for
is a thermostat; it costs more, but it's a lot easier to take care of.
The fancier irons also come with stands, little sponges, and other
conveniences.  These aren't necessary but they are nice.  You can get a
relatively cheap iron for as little as $10-20.  The expensive ones with
the thermostat can run $80 or more.  Different soldering irons (and
their larger cousin, soldering guns), come with different sized tips.
Larger tips heat up better but are not so good for really small stuff.

If you don't have the fancy stand, then you can very easily make due
with a pie plate; the disposable "tin" ones that frozen pies used to
come in work fine (do they still make these?).  You also want something
to rub the tip against to clean it; a sponge will work, or ordinary
paper towels will work (the coarse ones that you find in public
restrooms work best).  The sponge or paper towel should be wet; since
the paper towel will turn into mush, it's best to wet it last.  It
should be thoroughly wet, but not have any standing pools of water on
it.  You want to assemble all your stuff before you proceed further.
You don't want to leave the soldering iron unattended as long as it's
plugged in or still hot.  The first step is to plug the iron in and let
it warm up.  If it's a new iron, the tip will probably not be tinned,
but will most likely be a coppery color.  It will need to be tinned
before you can use it.  To do this, simply wait until the iron gets hot
enough to melt solder, then apply solder to the tip and let it flow all
over the tip.  So far as handling the solder goes -- it's usually
easiest to leave most of it coiled up on the spool, and to unroll it as
you use it.  Generally, you'll want to unroll a foot or so, and you'll
want to hold it at least 3-4 inches away from the end.  To apply it to
something, simply press the tip against the hot object, it will melt,
and you can apply a lot very fast by pressing the tip through the blob
and against the hot object.  It's a very interesting sensation.  When
tinning the iron, you'll want to rotate the tip as you apply the solder,
otherwise, the blob will drip off instead of coating the iron.  You
don't need a very thick layer, ideally, the thinnest layer that is shiny
bright will do fine.  Once you have the iron coated, you'll probably
want to remove the excess solder; to do this, simply run the iron over
the sponge.  It helps to press down fairly hard and move it fast
(otherwise, you'll melt the sponge or burn the paper toweling.) If there
are any black bits on the iron, you'll want to remove those, and and if
there are any bits that somehow aren't coated, you'll want to get those
coated.  You may need to use a file or something in stubborn cases to
solve any of these patchy problems.  The black stuff may be carbonized
solder, but it also tends to be corrosive, so you want to get rid of it
before it eats away at the tip and leaves it a funny shape.  The reason
why the solder might not want to stick to all of the tip is that the tip
will have some corrosion (almost all metals corrode in open air, and
quickly form an invisible layer of corrosion), and solder won't "wet"
corrosion.  This is, in fact, the reason for the rosin.  The corrosion
will also form a thin, but effective insulator, so it won't heat things
up as fast.  Indeed, you will probably find that you need to "scrape"
the iron with the solder to get it to actually start melting, even if
the iron is hot enough.

If you were to leave the iron just sitting for a long time, the hot
solder on the iron will start to corrode, and it will form a yellowish
deposit on its surface as it does this -- it can even look almost
golden, in some cases.  While it may look pretty, again, it's bad for
soldering, and it means you need to re-tin your iron.

Now, for actually soldering stuff.  The "approved" technique is this:
 (1) make the wires (or whatever) mechanically secure, so that they are
        wrapped against each other, and can't wiggle.  This isn't
        absolutely necesary, but it does make the job go easier.
 (2) heat the wires with the soldering iron.  There needs to only
        be a small amount of solder on the iron for this, basically
        just enough to enable the iron to "wet" the wires so that
        the heat travels into the wire better.
 (3) when the wires are good and hot, apply the solder to the wires.
        If the wires are not hot enough to melt the solder, they are not
        hot enough to make a good connection.  Ideally, you would
        use the heat from the soldering iron to travel through the metal
        and into the solder, but you may find instead that you need
        to apply the solder to the crevice between the tip and the wire.
        With practice, however, you can tell how many seconds it takes
        from between when you apply the iron to the wire, and when you
        should apply the solder.
 (4) you want to put enough solder on so that there is a smooth layer
        of melted solder smoothly coating the entire outside of the joint,
        and so it's good and solid.  Any solder past this point is
        at best wasteful, and at worst, could bridge a gap and touch
        other wires that you didn't mean to join.  (This is an
        especially bad thing when soldering IC dip packages which have lots of
        wires right next to each other that aren't met to be joined.)
 (5) remove the iron, and *without wiggling any wires*, blow on them
gently.
        The solder will have a smooth shiny appearance when melted,
        and will become slightly less smooth and look "whiter" when
        it freezes.  If you jiggle it while it's freezing, it may
        crystalize out into a much coarser form, and this is bad,
        because it may not make good electrical contact or may fail early (this
        is one form of what is called a "cold solder joint").
 (6) when the joint is good and cold, (and you can touch it with your
bare fingers),
        *then* it's safe to tug on it.  If you aren't sure if it's cold enough,
        try touching the wires a few inches from the joint, and running your
        fingers towards the joint -- if it gets too hot then stop and blow on
        it more.  The objects should be firmly attached to each other and there
        should be absolutely no wiggle.  Also, inspect the joint. The solder
        should not be beaded up on the joint, but should instead form a smooth
        continuous joint with the metal under it.  If it looks like you could
        pry it apart with a knife, the solder is probably not really attached
        to the metal under it, and the joint is defective.  Typically, with
        rosin core solder, some of the solder will congeal and form a
        transparent "drip" near or sometimes on the joint; this is completely
        harmless and can safely be ignored.

If you've never soldered before, you should practice on some scrap wires
first.  If you've never soldered electronics, ditto there, practice on
some "extra" stuff.  With the wires, try pulling it apart, you can
afford to be rough with your scraps.  Don't move onto the real stuff
until you feel comfortable that you can get dependable results.

Be careful.  Never leave a hot soldering iron unattended.  If you can't
pick it up by the tip, don't go away.  If it *is* hot, be extremely
careful around it.  You can get a *really* bad burn from the iron with
less than a second of contact, and even before the pain impulses make it
to your brain.  In fact, your first warning *may* be the smell of
"something cooking".  Trust me, though, you don't want to do this, and
there's no reason why you should do this if you are careful.

All good advice.  A thermostatically controlled iron is nice,  but not
essential.  *Don't* get carried away and buy a soldering *gun*, though;
they're not really suitable for PC board work.  You want a "pencil iron."  I
bought a generic 30 watt one from Radio Shack about eight years ago, and
except for replacing the tip every so often it's worked fine with no
attention whatsoever.  You don't have to spend a lot of money.  I don't
recommend a 15 watt iron because it just doesn't heat fast enough.  You have
to hold it on the joint too long, which risks overheating components.

I highly recommend getting a good stand, of the type that looks like a
spring on a base that you stick the iron into.  It shields the tip a bit and
prevents a lot of burns, at least for clumsy people like me.  The cheap
little metal wire stands that come with some irons leave the tip in a
position where it's easy to brush it by accident.  If you use one of these,
do remember that the *stand* will be quite hot after the iron's been sitting
on it.  Don't just pick it up with your fingers to move it.  (Ask me how I
discovered this.  Don't laugh, it's easy to forget.)

Where is the information on this ELMP?  Sounds neat, though a bit complex
for a first soldering project.

All good advice except the part about the wattage of the soldering iron.
Small integrated circuits a nd transistors will not stand much heat.  A 25
or 30 Watt soldering is ok if you are careful to not overheat the compnents.
Otherwise, stick to a 15 Watt iron.  It will take longer, but you will have
less damage.

It's also a good idea to clean off the extra flux with isopropyl alcohol.
You can get that at any grug store.  91% is best as it removes all the water
as it evaporates.  70% will do, but you will have to wait till it is
thoroughly dry before trying the circuit.

My theory on the irons is that a 30 watt for a brief period of time will
give heat less time to transfer up the lead to the component.  I've never
accedintally heat damaged anything with mine.

You can also get a 40 watt iron and put it on a lamp dimmer for more precise
control.

(thanks for the comprehensive response, Marcus)

I'd also agree with gull that hotter and faster is probably safer for the
component than cooler and longer.

For components that are particularly heat sensitive, if you have room you
can clip a pair of locking foreceps onto the lead you're soldering, between
the joint and the component.  That'll conduct the heat away.

I have a little clamping tweezer that does that and is smaller than the
hemostats you can get. 

Speaking of soldering...surface mount things do require better equipment
and skill. I haven't made any SM circuits, but I once had the SM fuse in a
modem blow, so I got a taste of replacing an SM component. I thought it
was pretty sneaky of them to install a SM fuse - planned obselescence
(actually, planned failure) in action. 

You see SM fuses a lot in computer keyboard ports.  I'm not sure about the
logic of putting in a fuse that's nearly impossible to find or replace. 
What's the point of protecting the rest of the components if it's going to
be dead anyway?

Good question. It does tend to get more good equipment thrown away
(especially after the warranty period), and hence increases sales.
Fuses are sort of timed-failure devices - they do age and become more
sensitive over time, and will therefore terminate the life of a
device that without a fuse would last much longer. Could anyone be so
diabolical?

Protects against fire or overcurrent on the power supply?

I once fixed the logic control part of a fancy turntable; it had what looked
to be little signal transistors with one lead missing.  I think they were
fuses, since the good one had no resistance and the bad one was open.

Fuses become more sensitive due to mechanical fatigue of the wire in them, I
think...just from vibration and such.  Not sure if that'd apply to surface
mount fuses as readily as to other types, though.

The only reason I know to clean off rosin flux is because you don't like
the looks.  If you were stupid enough to use acid flux (which you
shouldn't!) then, yes, you really really do want to clean it off.  In
commercial PCB manufacture, other stuff is used for making boards, at
least some of which is best washed off.  One common technique for this
is to use an ordinary dish washer with no detergent, just ordinary hot
water.  I once toured Ann Arbor Terminals and was quite amazed to see
them load up a dish washer with boards.  Water makes an excellent
cleaning agent, but there is one caveat, you want to dry boards quickly.
The dish washer probably got water up to 140F-180F, so the boards would
have dried quite rapidly once removed.

Fuses in modems may be a requirement of the telco - they don't like it
when you short out their lines.  I suspect a lot of computer parts are
made with the theory that there's no point in it "lasting forever",
because it's going to be functionally obselete in short order.  I still
have 3 external 2400 baud modems at home which probably work fine.  I
have to admit, I haven't been tempted to find out if they actually do
work.

I use to get all bent out of shape about getting transistors and such
too hot till I went to work for a guy who built test equipment for
the auto industry in his basement.  He told me that all electronic
parts were designed to be wave soldered and are not as sensitive to
heat as many make them out to be.  I've found that to be true.  I also
prefer about 25 watts for 95% of the PC board work that I do, especially
if it's only a single sided or double sided board.  Six and more sides
takes more heat simply because the board becomes more and more like a
sheet of copper as far as thermal conductivity.  Thermal control is
a non-issue with a 25 watt iron.  I also prefer iron clad soldering
tips because they outlast simple copper ones about 20 to 1.

Did you know that you can get SM fuse holders?  I think Digikey sells
both the SM fuse holders and fuses.  I generally just use a strand of
fine wire :-)

#15 slipped in.  The PC board vendors use a special flux that is water
soluble.  Rosin core flux is not.  (As I recall, one of the washable
variants is made from lemon.)

Yes, I've been known to wash PC boards right along with the dishes.
You just have to watch out for water sensitive stuff like speakers,
however.

Often, mass produced PC board products have the flux cleaned off and then a
"conformance coating" applied.  The coating prevents moisture from getting
on the board/components and changing resistances.  This never seemed like it
would be a big problem to me until I met a microphone manufacturer who uses
1 gigaohm resistors in one of his products.

Yeah...flux can cause problems in high-impedance circuits.  It's a little
sticky and can collect dirt, which can then cause leakage paths of a few
megohms or so between traces.

Rosin flux is not nearly as corrosive as acid flux on metal parts, but it
is still corrosive.  It's best to get it off if you can.

 I don't mean to sound ungratefull for you wonderfull responses, but
 that's a bit more information than I need...  I really appreciate you
 handing down some of your own experiences to me.  I was just wondering
 about the etching process, since that is before the component
 installation in the PCB layout.  How to I make etch a board based on
 that strange little 6cm X 4cm layout given to me by the ELMP people? 
 They talk about chemicals...where do I get those....I need to learn a
 lot before I even begin soldering the components.  Any more help
 appreciated greatly :) oh yes, btw in case your wondering the
 information on the ELMP can be found at -
 http://www.algonet.se/~cyrano/elmp/index.html
. lytez .

Board etching is a bit of an art.  The first step is to make a pattern of
'resist' on the board in the same pattern as the circuit traces.  There are
several processes for doing this, most of which I'm not familiar with.  I've
only done it using photoresist, but a friend of mine has had good luck using
toner transfer systems.  Anyway, once you have the resist pattern on the
copper of the board, you immerse it in a chemical that dissolves all the
copper not protected by the resist.  Ferric Chloride is one of a few
chemicals that are often used for this.  Then you clean the board, to remove
the remaining resist, and drill holes for the component leads.

The photoresist method of transfering a board layout works like this:  You
clean the board thoroughly, then apply a coating of a chemical that hardens
when exposed to UV light.  You let the chemical dry, then tape a reversed
transparency of your board layout to the board.  (i.e., the places where you
want copper are clear on the transparency, everywhere else is dark.) You
expose the board to ultraviolet light, then take off the transparency and
immerse the board in a developing chemical for a couple minutes.  After
that, you rinse off the board, and you have a resist pattern ready for
etching.  Questionable spots can be touched up with a resist ink pen before
putting the board in the etching tank.

Obviously, photoresist etching requires a few chemicals and a UV light
source.  You can buy special paper that you print in a laser printer or
photocopier, then transfer the toner to the circuit board with an iron; the
toner acts as the resist.  This is a better method for making a few boards
at home, probably.

As for the actual etching, I think Radio Shack sells a kit with a tank and
some etching solution.

In the case of the ELMP, you could probably wire the interface circuit they
show on a piece of perfboard, if you don't want to get into circuit board
etching.

I've seen drips of rosin on metal on decades old electronics.  The metal
is still nice and shiny and bright under the rosin--often brighter than
the metal that wasn't protected by the rosin.  Rosin may well be
corrosive, but it's probably less corrosive than ordinary air can be.

Wave solder machines are definitely pretty impressive things.  The basic
idea is that there is a long thin nozzle facing upwards.  The solder
comes pouring out of this and spills back into the machine.  The boards
to be soldered are attached to a conveyor belt that pulls the boards
over this flat fountain of solder, one by one.  Each board, which can
easily have hundreds or even thousands of connections, gets soldered in
less than a second.  I've hand-soldered PCBs -- it takes days and days
of painstaking labour to accomplish the same result, and it's
considerably more error-prone.

Commercial PCBs, especially ones to be wave-soldered, are generally
coated with a "solder resist".  This works just the same way that water
& oil does, in that the solder doesn't like to "wet" the resist and just
beads up or rolls off.

Ferric chloride is popular because it's safe (no fumes), easy to handle,
and produces dependable results.  But virtually any strong acid will
work; you could use hydrochloric acid and get similar results.  One
catch is that most acids will release small amounts of hydrogen as they
attack the metal--not enough to be dangerous, but the bubbles can cause
more uneven results, especially at high concentrations.

The results of the etching process will be some sort of copper salt,
probably blue or green, and definitely rather toxic.  In theory, you
really shouldn't pour it down the drain, but ought to dispose of it as a
hazardous waste; I don't know of anyplace you could just drop a small
amount off at however.

Right.  Being at a college, I have the benefit of an etching machine...it's
basically a tank of heated ferric chloride, with a nozzle arrangement to
spray the stuff on your board, hopefully evenly.  The board is never
actually immersed, so bubbles aren't a problem.

Small amounts of acid can be neutralized with baking soda or washing soda.
Ferric chloride is tougher as it is toxic at moderate concentrations.  Low
concentrations are toxic to aquatic life.  Same thing for the copper salts,
only not quite as toxic.  I suppose you could let the water evaporate and
handle it as a solid.

Ferric chloride is easily "neutrealized" with sodium bicarbonate (baking
soda - but let's use the right names for chemicals). The result is ferric
(basic) oxide (rust) and sodium chloride (salt). Carbon dioxide will be
given off. Add enough of the bicarbonate until it no longer bubbles,
though an excess does not hurt.  You can dispose of the "rust" slurry down
the drain, except that after use it contains copper, and that is the
problem. 

Copper is much more toxic to both plants and animals than iron. It would
also be precipitated by sodium bicarbonate and form the basic carbonate,
but this is still slightly soluble. You could let it settle, pour off the
liquid, and absorb the precipitate into (say) kitty litter, and disposed
of as toxic waste. 

Bicarbonate neutralizes the caustic properties of Ferric Chloride, but it is
still toxic.  Dumping down the drain is not recommended,  it is on the EPA
list of water contaminants.

What's toxic about "rust"? There is lots lining the sewer pipes already.
What. exactly, are you saying is on the EPA list of water contaminants?

I'm guessing it's the copper dissolved in the used ferric chloride that's
toxic.

Well, I said that - and how to deal with it.

Well, I see we need a chemistry lesson here.  To begin with, rust is not
ferric oxide.  It is a mixture of various iron oxides, whose exact composition
is unknown.  When you add bicarbonate to a solution of ferric chloride, you
do not get ferric oxide.  You get something called hydrated ferric oxide
(Fe2O3.nH2O), sometimes called ferric hydroxide.  The important point is that
the ferric ions are still soluble.  Ferric ions are considered a water
pollutant.
The only thing that appears to be possible is to add sodium phosphate to your
ferric chloride solution.  Ferric phosphate should precipitate out.  This
compound is insoluble and proably can be disposed of in the trash.  Copper
also forms a cupric phosphate which is only slightly soluble, and then only
in warm water.  The problem is getting sodium phosphate.

If you had a solution of hydrated ferric oxide, turning it into a solid
("rust") is no big deal, and doesn't require any other chemicals.  Just
let it dry out.

The real problem here, though, is that you want to get rid of it.
Getting rid of ferric oxide is no big deal; while it's true "too much"
iron oxide is actually a bad thing, the amount that's bad is orders of
magnitude larger than for copper, and there is certainly a lot of rust
out there already.  If nothing else, you could put your "rust" in a
paperbag in the trash.  Copper oxide, on the other hand, is a bad thing,
and even trace amounts (the amount you might get off a penny dropped
into a goldfish bowl) turns out to be not a good thing (for the
goldfish).

Hydrated ferric oxide is not just ferric oxide with water added.  It is an
entirely different compound with different properties.  If it could be
converted to ferric oxide it would be safe to throw in the trash. 
Unfortunately, that process is not very easy.

I agree that copper in any form is not good for aquatic life, but copper is
thrown in the trash every day.  

What's hard about roasting hydrated ferric acid?

(er, ferric _oxide_, that is...)

If all this chemistry makes your head spin, you can also do mechanical
etching.  I've seen carbide end mills as small as .010 diameter.  With one
of those and a mill with a turbine head (10,000 to 70,000 RPM) is is
pretty simple to cut away unwanted copper on a board.  Add CNC and you can
go from a CAD layout directly to a copper board.

Response not possible - You must register and login before posting.

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss