Joe,
I could probably find hundreds of additional examples, and that is why I said that
there are always exceptions to the "rule" (which is the industry standard practice
of marking the marking the Cathode).
This is especially true in the area of LEDs where the size, shape, and physical
design of the package varies so much.
That is why it is so important that whatever the type of package, and whatever the
method of identifying the orientation of the diode, it should be very carefully and
very clearly spelled out in the related Documentation.
This conversation began in another forum (Designers Council) with the question of
which terminal of the diode (which in this case evolved during the discussion into
an LED) should be "Pin 1", which is one of the "root problems" of the "orientation"
issue in Manufacturing, and the conversation evolved again into why IPC-7351 chose
the Cathode as the "Polarity Indicator".
I won't rehash the old ground already covered below, except to say that you have
pointed out yet again one more example of the need for the standardization of
components when it comes to what we could call "Polarized Orientation", where the
component, whatever it may be, from diode, to capacitor, to transistor, to SOIC and
BGA packages, or any other component package type which could be installed in more
that one "orientation", needs to be clearly identified in the PCB Design when it
comes to the issue of the "orientation" of that component during installation.
This is why from a PCB Design perspective, when it comes to design of the
"footprint" in the Component Library (and hence it's usage in the PCB Design)
IPC-7351 has dictated that there shall be a "basic orientation" for all components
within the Comport Library, where the "footprint", as designed, in its original
unrotated position (which is called "zero rotation"), will have "Pin 1" in the upper
left for all components having 3 or more terminals, and that 2 terminal components
shall be oriented horizontally with "Pin 1" to the left, with the orientation of
components requiring specific "orientation" as previously described. (Please note
that there is one exception to even this rule, which is for "legacy" packages, such
as the "PLCC" (which are "obsolete" per JEDEC JEP95 and not to be used for "new
design") where "Pin 1" is located in the middle of the top side of the component
based on the pin numbering of the "die" for such devices, which is the same as the
"die" used for DIP packages.)
The ultimate goal here is to end the current confusion, of which you have provided
yet another example, such that ultimately, if the PCB Design is done using
components "oriented" in accordance with IPC-7351, and the Component Manufacturer
packages their components in accordance with the applicable "packaging
specifications", automated manufacturing of the PCB could be accomplished without
any physical marking of the component at all, which is not to say that the
Manufacturing Engineer doesn't still have to double check component "orientation",
or that the Assembly Drawing doesn't still have to convey whatever information is
required to properly install each and every Component in the PCB Design, whether it
be a diode (including LEDs), capacitor, transistor, SOIC, BGA, or whatever.
My position in this conversation has simply been to point out that whether or not
everybody is happy with the "orientation" of Components as specified by IPC-7351
respecting "Component Libraries" used in automated PCB Design systems, that if the
"orientation" specified by IPC-7351 is followed in the Design of the PCB, then it
will remove any ambiguity or questions when it comes to the subsequent Assembly of
the PCB as to what the actual "orientation" of the component should be, especially
when it comes to fully utilizing the automation capabilities of the PCB Design
"software", such as "Pick and Place" data.
Again, we could probably find hundreds if not thousands of examples of the problem,
and the whole point is that there needs to be some solution to the problem.
I believe that that solution is the complete implementation of IPC-7351 in the
Design or the PCB, coupled with clear and unambiguous Documentation.
If properly implemented and adhered to, IPC-7351 will ignore the exceptions such as
those that you have brought up by properly compensating for them, and eliminate all
but a very few of the current problems associated with the "orientation" of a
Component, especially when it comes to Automated Manufacturing.
Please remember that a "Pick and Place" machine, for all of its speed and
manufacturing capability, cannot "look" at a part and "see" what it's "orientation"
should be, based on whatever kind of "identification" there may be identifying
whatever kind of electrical termination there may be (at least not yet, maybe in the
future, but not now).
JaMi
* * * * * * * * * *
----- Original Message -----
From: "Schleman, Joe A." <[log in to unmask]>
To: <[log in to unmask]>
Sent: Monday, November 01, 2004 7:17 AM
Subject: Re: [TN] Diode Pin Numbering Question...
Steve,
This doesn't concern through hole components but, I just want to add that not all
surface mount LEDs have the cathode marked. High intensity, high efficiency red LEDs
tend to have the anode marked.
Joe
-----Original Message-----
From: TechNet [mailto:[log in to unmask]]On Behalf Of JaMi Smith
Sent: Thursday, October 28, 2004 5:26 PM
To: [log in to unmask]
Subject: Re: [TN] Diode Pin Numbering Question...
Steve,
First, its always "fun" to tackle some of the questions that you throw at us when
you are checking to see whether or not we are awake. So my "fun" answer follows, but
should not be taken to seriously.
I'm actually going to pull a quick one on you here and say that "none of the above"
(or in this case "below") are correct.
First and foremost, the schematic symbol in both cases is that of a regular signal
diode, as opposed to the correct symbol for a LED (Light Emitting Diode), which
universally would have some kind of little "squiggly" ray, or at least a little
arrow or two, beside the diode symbol.
Secondly, while it is almost universal that the Anode is the longer of the two
leads, I have personally experienced LEDs with both leads cut to the same length,
and additionally LEDs that were in "tape" where you cannot tell anything by the
length because the ends are obscured.
The problem here is that by industry standard, almost universally, diodes of every
kind, including LEDs, are actually identified by a band or some other marking on the
Cathode lead of the diode. Specifically, when it comes to LEDs, I think that you
will find that in spite of the Anode lead generally being longer, that the actual
preferred industry standard is to have a "flat" on the side of the LED which
identifies the Cathode lead, and that it is this "flat" that is actually what has
been standardized as the orientation identifier.
There are of course, just as with everything else in the world of PCB Design,
exceptions to the rule. One example would be a LED which did not have a flanged base
which could have the "flat", but it is hard to tell from your picture whether or not
the LED in your case is round or square, and whether or not this would apply in your
case. Another would be in the case of different SMT packages, which is obviously not
the case in your instance. Respecting the case where an LED did not have a flange, I
have actually seen LEDs that didn't have the flange, but which still had an indented
"flat" on the circular body.
Actually, the real answer to your question is dependant upon the actual physical
part that you specify in the Design, and whether or not it has any specific
distinguishing orientation marks or characteristics, and also importantly whether or
not there any approved substitutions which may differ from the primary source part.
In any event, your drawing does not look at all like your part, which again could
lead to ambiguity, and more importantly does not indicate the industry standard
"flat" (which again may not be present in all cases), but more importantly I would
be afraid that with all of the information that you have given them in your
pictorial explanation, that some would except it all as the "gospel truth", and save
a copy of your little pictorial explanation for future reference, and possibly make
improper references somewhere down the line, such as using the shape of the internal
part of the leads for identification, when they should be looking first for a
"flat", and secondly I would give you the lead length for most cases, but I would
not bet my life on all LEDs having the "die" attached to the Cathode (although I
would probably give you five 9's on that one). Point here being that you can give
someone too much information, such as the picturing the internal structure, when
they should really be looking at other things. Remember who we are dealing with
here, and that it is your responsibility to make it clear on the drawing so that
there isn't any ambiguity. Sometimes the problems arise when there is not enough
information to work with, but conversly, sometimes the problems really can arise
because of too much information.
Respecting an interesting example that I just grabbed at random from Digikey, look
at page two of the following PDF:
http://rocky.digikey.com/WebLib/Lite-on/Web%20Data/LTL-2F3VxKNT,H3VxKNT,P3VxKNT,R3Vx
KNT.pdf
. . . and then answer this question for me: Which is the proper interpretation
respecting a LED which has a long lead on the Anode, and a short lead on the Cathode
(please select only one answer):
O The longer lead identifies the Anode.
O The shorter lead identifies the Cathode
Six of one, half a dozen of the other.
I would have to fall back on the industry standard convention of identifying the
Cathode with a "flat", and in the lack of a "flat" I would have to fall back on the
manufacturers datasheet, which oddly enough in this example identifies the shorter
lead in addition to the "flat".
Now as to which lead should be "Pin 1".
You will never resolve this issue to everyone's liking, but the fact is that the
industry standard is and always has been to identify the Cathode end of the diode
for orientation.
Getting back to the primary issue, which is how will we resolve this issue in the
real world of PCB Design and Manufacturing: IPC-7351 has resolved the issue by
stating that from a Component Library of the PCB Design perspective, all two
terminal components shall be oriented horizontally, with "Pin 1" on the left (at
"zero degrees of rotation" (before being placed in the Design, which may require
rotation)), and that further, that when two terminal components are "polarized"
(require specific orientation to operate properly), that the Polarity Indicator
shall always be "Pin 1", which again is always to the left, and that for Capacitors
the Polarity Indicator shall be the Positive Terminal, and that for Diodes the
Polarity Indicator shall be the Cathode Terminal (as has always been the industry
standard).
"Pin 1" is always to the left at "zero rotation" in the Component Library, and it is
always the Cathode.
So will this make everybody happy?
Obviously not, but it will resolve the issue, and it's related problems, "IF" it is
properly applied and adhered to.
Now respecting the where the "square" pad should be, if one is used at all, in view
of the above, and in view of the fact that the industry always identifies the
Cathode of a diode (yes I know that someone will show me a conflicting example), I
would say that if it was used at all, that it should be on the Cathode lead, but
preferably, in most cases, a standard T 1 3/4 LED should be identified on the silk
screen with a circle that has a "flat" at the Cathode lead, and beyond that, do
whatever it takes to get your shop to put the LED in your Board correctly.
Now, respecting those who would cry that the leads should be identified as "A" for
Anode and "C" for Cathode, I would throw the original versions of ANSI Y32.2, and
even possibly back as far as MIL-STD-15, at them, and say that they would have to
make that "A" for the Anode and "K" for the Cathode.
Boy, talk about a screw up, the Military really blew it on that one, and messed up
the industry for a number of years on that one, but what do you expect, SNAFU. That
one was almost as bad as the Navy changing the abbreviation for Ground from "GND" to
"GRD", and then back again to "GND", but not quite.
AAAARRRRRRRRRGGGGGGGGGHHHHHHH!!!!!!!!!!!!!!
( As he lets out the occasional Primal Scream ! )
Hope it's been "fun".
JaMi
* * * * * * * * * *
----- Original Message -----
From: "Steve Gregory" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Tuesday, October 26, 2004 7:14 AM
Subject: [TN] Diode Pin Numbering Question...
> Mornin' All!
>
> There's been a pretty interesting thread going on in the Designers Council about
what is supposed to called pin 1 of a diode...the Anode or the Cathode. Believe it
or not, there's been conflicting answers. So I'd like to do a little test.
>
> Look at:
>
> http://www.stevezeva.homestead.com/files/LEDPin1.jpg
>
> and tell me if we installed this LED correctly.
>
> Then look at:
>
> http://www.stevezeva.homestead.com/files/LEDPolarity.jpg
>
> and tell me if this this illustration guide I made for our operators is correct.
>
> (This ought to be good)
>
> -Steve Gregory-
>
> ---------------------------------------------------
> Technet Mail List provided as a service by IPC using LISTSERV 1.8e
> To unsubscribe, send a message to [log in to unmask] with following text in
> the BODY (NOT the subject field): SIGNOFF Technet
> To temporarily halt or (re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL)
> To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]:
SET Technet Digest
> Search the archives of previous posts at: http://listserv.ipc.org/archives
> Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for
additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100
ext.2815
> -----------------------------------------------------
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]:
SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for
additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100
ext.2815
-----------------------------------------------------
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]:
SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for
additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100
ext.2815
-----------------------------------------------------
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------
|
