Maker Pro
Maker Pro

PCB Layout -- A question of style

  • Thread starter RST Engineering \(jw\)
  • Start date
R

RST Engineering \(jw\)

Jan 1, 1970
0
I know how I was taught back in the '60s, but just for grins (since I've
seen a couple of reversed examples the past few days) how do you lay out the
pads on a polarized component?

For example, most pads are round. However, I was taught to make the cathode
of diodes and the (-) negative lead of polarized capacitors square. Same
for pin 1 of an IC and the collector/drain of a transistor.

However, in the last few days I've seen the ANODE of the diode square. Is
this a new convention or just a goofup in the mind of the layout artist?


Jim
 
J

Joel Kolstad

Jan 1, 1970
0
RST Engineering (jw) said:
I know how I was taught back in the '60s, but just for grins (since I've
seen a couple of reversed examples the past few days) how do you lay out the
pads on a polarized component?

For example, most pads are round. However, I was taught to make the cathode
of diodes and the (-) negative lead of polarized capacitors square. Same
for pin 1 of an IC and the collector/drain of a transistor.

However, in the last few days I've seen the ANODE of the diode square. Is
this a new convention or just a goofup in the mind of the layout artist?

I believe it comes from the "convention" of always making pin 1 square and the
other pins rounded... and anodes are usually made pin 1. Likewise, with
polarized caps, I've seen the anodes square and the cathodes rounded, since
again the anode was made pin 1. This does seem a little more consistent.

Were you taught this in school or at some company?
 
R

RST Engineering \(jw\)

Jan 1, 1970
0
I did PCB layout for four companies going through school and they all did it
the same way. Again, this was 40 years ago. I also remember it from a text
on PCB layout that we used in school.

Jim
 
J

Joerg

Jan 1, 1970
0
Hello Jim,

On SMT that isn't done much anymore. At least I haven't seen it often.
Mostly the identifier will be on the silk screen and it would be
wherever the datasheet says is pin 1.

Regards, Joerg
 
D

Dr. Anton T. Squeegee

Jan 1, 1970
0
I know how I was taught back in the '60s, but just for grins (since I've
seen a couple of reversed examples the past few days) how do you lay out the
pads on a polarized component?

Squares for pin 1 on IC's and connector headers, round for the
others. If I've got a polarized active or passive, like a diode or
tantalum, I'll usually make the positive pad (for caps) square, and the
cathode end of diodes square as well.
However, in the last few days I've seen the ANODE of the diode square. Is
this a new convention or just a goofup in the mind of the layout artist?

Sounds to me like a design hiccup.

Keep the peace(es).


--
Dr. Anton T. Squeegee, Director, Dutch Surrealist Plumbing Institute.
(Known to some as Bruce Lane, ARS KC7GR,
kyrrin (a/t) bluefeathertech[d=o=t]calm -- www.bluefeathertech.com
"If Salvador Dali had owned a computer, would it have been equipped
with surreal ports?"
 
J

JeffM

Jan 1, 1970
0
I was taught to make the cathode of diodes
and the (-) negative lead of polarized capacitors square.
Same for pin 1 of an IC
Jim (RST Engineering (jw))
Early-on I saw it done it that way and have continued in that vein.
The terminal that is marked on the part gets marked on the PCB.
..
..
and the collector/drain of a transistor.
Since the tang on early transistors was the emitter (TO-5)
it was the emitter that got marked.
 
B

Ban

Jan 1, 1970
0
RST said:
I know how I was taught back in the '60s, but just for grins (since
I've seen a couple of reversed examples the past few days) how do you
lay out the pads on a polarized component?

For example, most pads are round. However, I was taught to make the
cathode of diodes and the (-) negative lead of polarized capacitors
square. Same for pin 1 of an IC and the collector/drain of a
transistor.
However, in the last few days I've seen the ANODE of the diode
square. Is this a new convention or just a goofup in the mind of the
layout artist?

The layout technician uses his program and only when he defines his own
parts he will think about. It has become pretty meaningless as well with
pick and place machines, who do not care how the pad looks. And through-hole
parts are hardly used any more, seems just on the hobbyist level where it
matters. Each one has his preferences, so even if you learned this once, now
nobody knows about it any more.
I have worked with both technologies in the 80s, making 2:1 sized layouts on
those big blue grid sheets and I had a big collection of rub-on eyes and
lines, but lately I have thrown all in the garbage, hadn't used it any more
for ages. And I am happy about that!
 
F

Frank Bemelman

Jan 1, 1970
0
RST Engineering (jw) said:
I know how I was taught back in the '60s, but just for grins (since I've
seen a couple of reversed examples the past few days) how do you lay out the
pads on a polarized component?

For example, most pads are round. However, I was taught to make the cathode
of diodes and the (-) negative lead of polarized capacitors square. Same
for pin 1 of an IC and the collector/drain of a transistor.

However, in the last few days I've seen the ANODE of the diode square. Is
this a new convention or just a goofup in the mind of the layout artist?

I use square pads for:

pin 1 of an IC
pin 1 of a boxheader/connector etc.
+ side of polarized cap
anode of a LED
all pins of a TO220-78xx regulator.

For pth diodes, I use round pads for both anode/cathode.
 
R

R.Lewis

Jan 1, 1970
0
Ban said:
The layout technician uses his program and only when he defines his own
parts he will think about. It has become pretty meaningless as well with
pick and place machines, who do not care how the pad looks. And through-hole
parts are hardly used any more, seems just on the hobbyist level where it
matters. Each one has his preferences, so even if you learned this once, now
nobody knows about it any more.
I have worked with both technologies in the 80s, making 2:1 sized layouts on
those big blue grid sheets and I had a big collection of rub-on eyes and
lines, but lately I have thrown all in the garbage, hadn't used it any more
for ages. And I am happy about that!
Well said.

I agree it is, and generally always has been, a pretty useless thing to do -
although I admit to having used square pads as pin identifiers in times past
I did realise that it served no real purpose and was likely only to lead to
confusion.
As you say a square pad means nothing to SM or an auto pick and place.
To hand stuffed PCB's the pad is generally on the other side - visible only
with X-ray eyes.
The purpose of the silk is for *exactly* the identifications previously
mentioned although many seem to think the component ident. is there to make
the board look 'professional'.
 
J

Joel Kolstad

Jan 1, 1970
0
R.Lewis said:
I agree it is, and generally always has been, a pretty useless thing to do

For small discrete components, arguably true, but I'd defy you to find me
anyone who'd suggest that one differently shaped pad on a connector footprint
or a many pin package could indicate anything other than "pin 1."
The purpose of the silk is for *exactly* the identifications previously
mentioned although many seem to think the component ident. is there to make
the board look 'professional'.

On very dense boards, much of the silkscreen is often completely obliterated
due to packing components very tightly together. But I agree that silkscreens
should indicate pin 1 as well.
 
R

RST Engineering \(jw\)

Jan 1, 1970
0
The layout technician uses his program and only when he defines his own
parts he will think about.

What the hell does THAT mean?


It has become pretty meaningless as well with
pick and place machines, who do not care how the pad looks.

I didn't ASK you about pick and place machines. There are places in this
world where hand stuffing is still in vogue. Why don't you simply answer
the question instead of telling us how other people do it?


And through-hole
parts are hardly used any more,

The hell you say. Distributors are still selling equal amounts of SMD and
TH parts.

I have worked with both technologies in the 80s, making 2:1 sized layouts
on those big blue grid sheets and I had a big collection of rub-on eyes
and lines, but lately I have thrown all in the garbage, hadn't used it any
more for ages. And I am happy about that!


Well, I used blue and red grid sheets in the '60s and still teach them as
legacy methods to my college students. I also teach them how to lay out a
board using tape and donuts. THEN and ONLY THEN do I introduce them to
computer methods.

Now, I ask the frikkin question again (since there seems to be disagreement
on both sides) what is the STANDARD convention for using square holes to
indicate the lead of a component?

Jim
 
P

Paul Burke

Jan 1, 1970
0
RST said:
Well, I used blue and red grid sheets in the '60s and still teach them as
legacy methods to my college students. I also teach them how to lay out a
board using tape and donuts. THEN and ONLY THEN do I introduce them to
computer methods.

That's about as useful as asking them to design a 32 bit FPU using only
NAND gates. Why not go back to the original method, and silk screen the
etch resist on? If you REALLY wanted to teach them something useful, it
would be to lay out a single sided PCB, black rub on tape or computer/
laser printer, and etch a real PCB from it. Teach them about under/ over
etch, clearances, feature sizes etc.

Paul Burke
 
L

legg

Jan 1, 1970
0
I know how I was taught back in the '60s, but just for grins (since I've
seen a couple of reversed examples the past few days) how do you lay out the
pads on a polarized component?

For example, most pads are round. However, I was taught to make the cathode
of diodes and the (-) negative lead of polarized capacitors square. Same
for pin 1 of an IC and the collector/drain of a transistor.

However, in the last few days I've seen the ANODE of the diode square. Is
this a new convention or just a goofup in the mind of the layout artist?

Pad art used to serve a function in manual assembly, inspection and
test/repair. As all are becoming increasingly rare, the imperative for
standard practice evaporates.

The markings in the silk screen were more important in the first two
functions - teckies dealing with solder-side were supposed to be able
to figure out what they were looking at, regardless.

In manual assembly the longer or marked lead/end aligned with the
marked receptacle/hole.

Producing designs that enhance manufacturability will become
increasingly difficult, as the distance between design and mfring
sites increases. This is one reason why design capability tends
inevitably to follow mfring physical relocations.

The new designs you've examined may have been produced by individuals
with little knowledge of or concern for mfring processes, having had
no experience of them.
RL
 
L

legg

Jan 1, 1970
0
I know how I was taught back in the '60s, but just for grins (since I've
seen a couple of reversed examples the past few days) how do you lay out the
pads on a polarized component?

In more current designs, using higher volume assembly methods,
locations of tooling holes and optical locating fiducials are probably
more relevent than pad art. Copper patterns are more likely altered to
improve solderability and test point access, than as visible guides
for manual operations.

It's never really been 'style', unless it followed 'function'.

RL
 
R

RST Engineering \(jw\)

Jan 1, 1970
0
And you teach where, sir? And for how long? And how many of your students
are working in the industry today?

You evidently have no clue as to how to introduce an absolutely green
student with no prior knowledge of the art how it is done. Teaching button
pushing on a computer is what got us to our sorry state of affairs in the
industry today.

The first lesson is actually to use a resist pen on single sided copper and
etch it themselves in a tray with ferric chloride. Everything else follows
from that.

Jim
 
B

Boris Mohar

Jan 1, 1970
0
And you teach where, sir? And for how long? And how many of your students
are working in the industry today?

You evidently have no clue as to how to introduce an absolutely green
student with no prior knowledge of the art how it is done. Teaching button
pushing on a computer is what got us to our sorry state of affairs in the
industry today.

The first lesson is actually to use a resist pen on single sided copper and
etch it themselves in a tray with ferric chloride. Everything else follows
from that.

Jim

Why not make them mine and smelt copper, roll into a foil etc? At this
rate when will they learn how many layers are required to fan out 780 pin
BGA?
 
R

RST Engineering \(jw\)

Jan 1, 1970
0
The principle is that it is impossible to build a cathedral on an outhouse
foundation. THey will learn how many layers are necessary when they have to
know that. It well may take a lifetime. Have you stopped learning?

Jim
 
R

R.Lewis

Jan 1, 1970
0
Joel Kolstad said:
do

For small discrete components, arguably true, but I'd defy you to find me
anyone who'd suggest that one differently shaped pad on a connector footprint
or a many pin package could indicate anything other than "pin 1."


On very dense boards, much of the silkscreen is often completely obliterated
due to packing components very tightly together. But I agree that silkscreens
should indicate pin 1 as well.

'Should indicate pin1 as well'?
As well as what?

This is the point I was making.
Silks are used for also sorts of mumbo jumbo reasons- pretty rectangular
blocks denoting dil IC's for example- whereas showing pin1 for a manually
stuffed board is the utmost priority for the silk.
 
J

Joel Kolstad

Jan 1, 1970
0
R.Lewis said:
'Should indicate pin1 as well'?
As well as what?

As well as a differently shaped pad on pin 1 does.
This is the point I was making.
Silks are used for also sorts of mumbo jumbo reasons- pretty rectangular
blocks denoting dil IC's for example- whereas showing pin1 for a manually
stuffed board is the utmost priority for the silk.

With pick and place machines, I consider silkscreens to mainly serve the
purpose of the design engineer insofar as debugging the prototype goes. :)
Thereafter, it's main purpose would be to help anyone repairing the board,
although these days that often never happens (i.e., "troubleshooting" consists
of swapping entire boards until a system comes back to life).

---Joel
 
Q

qrk

Jan 1, 1970
0
In more current designs, using higher volume assembly methods,
locations of tooling holes and optical locating fiducials are probably
more relevent than pad art. Copper patterns are more likely altered to
improve solderability and test point access, than as visible guides
for manual operations.

It's never really been 'style', unless it followed 'function'.

RL

The different pads styles aren't necessarily for assembly purposes
except for the engineering build that requires some hand assembly and
twiddling. Think of the poor technicians that repair boards either
coming off the production line or in a dimly lit, hot, humid, cramped
compartment of a ship that's rocking back and forth. Sometime
helicoptering in a replacement board isn't feasible, so you need to
fix the problem on the spot. People who design PCBs should try fix
boards in uncomfortable settings so they will understand why distinct
pads and helpful silkscreens are provided, even in this day and age of
SMT.
 
Top