JaMi makes a good point here. I would guess the original question arose from
the use of ODB++ database usage or a net list that was translated from the
original CAD design database for their assembly house usage. I expect we
will see more similar issues with assembly as the front to back process
becomes more integrated. CAD Designers who make databases and library
symbols and parts assign the pin number (which in Protel can be
Alpha-numeric) arbitrarily. The expectation is for the assembler to LOOK at
the silkscreen and orient the part based on the graphic. They used to not
get the pin number information unless someone offered to give them the
net-list. They were still able to get the polarity correct.

As we get more integrated with assembly there are bound to be
inconsistencies between the assembly house practice and the PCB Designers
component symbol creation practice.  I was not suggesting that anyone
'ass-u-me' anything regarding the pin one practice. However, I do make my
own discrete diode library symbols with the Cathode at Pin 1 along with the
graphic polarity indication in the silk screen legend when I am given the
choice of where to put them. Some diode land patterns I have are not
numbered with pin numbers and have A/K designations instead for the pin
numbers.

 Here's also an exception - Some diode packages are multi-leaded SOT-23's
and the cathode may not be on pin one. The diode data sheet description will
take precedence in determining how I number my parts. Especially if I want
the symbol to reflect the configuration of the part accurately.

Axial leaded diodes that have only 2 leads are typically done with pin 1 on
the cathode side when I do them. It just feels more correct and less
ambiguous to me. The manufacturers Data sheet does not assign a pin number
to either lead of the diode. Therefore, since my net list requires a pin
number to be able to identify the connection, I can assign these pin numbers
arbitrarily.

I believe I have not seen any official spec for this sort of thing and if
there was, it would probably have to come from JEDEC or IPC.

Sorry if I offended anyone with the comment about setting a new precedent...
it would be odd though to me if anyone chose to make an axial diode part and
numbered the cathode pin 2.



Bill Brooks, CID


-----Original Message-----
From: JaMi Smith [mailto:[log in to unmask]]
Sent: Monday, June 23, 2003 12:31 PM
To: [log in to unmask]
Subject: Re: [TN] Diode Pin Assignment

Steve,

There is no "industry standard" for pin numbers on diodes.

A designer / engineer will usually assign pin numbers to an individual part
based on the way that the pin numbers are specified on the manufacturers
datasheet for that specific individual part. This may in fact even change
from datasheet to datasheet and manufacturer to manufacturer for the same
part type in the same package.

A designer /engineer may also arbitrarily change the pin numbers to make an
attempt to standardize a particular part with other parts in his design or
his component libraries, simply because he wants to try and keep all of his
parts the same.

More often than not, a designer / engineer will simply use the "component
part" which is already pre-defined in the component library of his cad
system. Pin numbers for diodes and capacitors may differ from part to part
with a different package style within a given component library on a given
cad system, and even from one compinent library to another within that same
cad system. Pin numbers may also differ for the identical part from the
component library of one cad system to the same part in another component
library of another cad system, which is why you may see a difference in pin
numbers for the identical part from one of your customers to another
customer. You may even see different pin numbers on the same part from the
same customer, simply because the design was worked on by a different
designer / engineer.

Pin nunbers can even vary due to the method of packaging in order to conform
to EIA Standards for Tape and Reel

This is why you cannot rely on pin numbers to determine the polarity of any
part.

Every Diode, just like every polarized Capacitor, should have it's polarity
clearly marked on at least the Assembly Drawing, and preferrably also on the
Board itself. If not, then consult the designer / engineer.

Never Never Never Never Never Never Never Never Never Never Never Never
Never Never Never Never Never Never Never Never Never Never Never Never
Never Never Never Never Never Never Never Never Never Never Never Never
Never Never Never Never Never Never Never Never NeverNever Never Never Never
Never Never Never Never Never under any circumstances whatsoever take it
upon yourself to "manually rotate the part within [your] CAM system to stay
consistent with what [your] operators typically see on their visual aids".

YOU MUST INSTALL THE PART WITH THE CORRECT POLARITY IRRESPECTIVE OF WHAT THE
PIN NUMBERS ARE.

IF YOU CANNOT DETERMINE THE CORRECT POLARITY, STOP EVERYTHING UNTIL YOU CAN
CONTACT THE DESIGNER / ENGINEER AND DETERMINE THE CORRECT POLARITY.

In an ideal world, Pin 1 would always conform to EIA Standards for packaging
the part on Tape and Reel, but even this is not always the case, Even here,
sometimes the size or shape of the part determines that it be packaged
differently. Sometimes the manufacturer just doesn't follow the EIA
Standards. This is why it is imparative that you contact the designer /
engineer whenever the polarity of a diode is not absolutely clear.

When it comes to polarized parts, never assume anything based on pin
numbers.

JaMi Smith

* * * * *

----- Original Message -----
From: "Steve Vargas" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Friday, June 20, 2003 7:18 AM
Subject: [TN] Diode Pin Assignment


> All:
>
>     As a SMT subcontractor, we see CAD files from many different sources.
> An item of frustration for us is how some customers assign a Pin 1
> designator to diodes. Primarily, we see Pin 1 being assigned to Cathode
and
> Pin 2 to Anode. For those customers who use Pin 1 to anode, we manually
> rotate the part within our CAM system to stay consistent with what our
> operators typically see on their visual aids. This is both cumbersome and
> inefficient.
>    I don't believe (or should I?) that there is an industry standard that
> addresses this, but is there some type of defacto standard that exists
> which I can use as evidence to some of our customers that they should
> consider using the 'Pin 1 to cathode concept'? Thanks.

---------------------------------------------------
Technet Mail List provided as a free 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/html/forum.htm for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700
ext.5315
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a free 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/html/forum.htm for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
-----------------------------------------------------