Correct Connector Symbols to use IAW IEC 60617

All connectors shown in the KiCad symbol library “conn” need an overhaul to conform to IEC 60617. For instance take a look at the symbol for a D-susb 9 (DB9) connector. The graphic symbol shows unfilled circles and the outline depicts the characteristic D shape. An unfilled circle is the symbol for a terminal but in this case I presume this is supposed to be the symbol for a socket/female contact. Likewise a filled circle is presumed to be for a pin/male contact.

First off the D shape outline needs to be deleted as this is showing a mechanical perspective that goes on the associated assembly drawing. Then delete the terminal symbols (the unfilled circles) for the sockets/female contacts. Here is a snippet from what I found on the Internet.

For those of you who work with BNC, SMA, N and other types of coaxial connectors here is the IEC 60617 symbols:


This is from IEEE 315A, but as noted these are IEC 60617 symbols.

I must admit I don’t understand the point of this standard. If we agree that mechanical detail such as the “D” shape is not relevant for the schematic, then why is the type of pin mating relevant?

Put another way, what possible schematic error does this type of convention aim to prevent?

Personally I put a generic 9 way connector and that is all the information required for the schematic.

2 Likes

For BNC connectors (or specifically an SMA one) I have a symbol with that ring and the shield connection… but that’s the only one.
As for plug and socket I don’t differentiate in the schematic either.

D types have caused a lot of problems. In Racal we used to talk about a female connector with pin contacts. Plug and socket are too ambiguous.

Reverse DIN and reverse sma have also tripped up a pcb design

1 Like

I think it is similar to the way that Analytical Chemists confront Wine Tasters, and Strict Grammarians square off against Creative Writers.

Including a caricature of the “D” shape (or, for that matter, pins clustered in a circle, or the double row of a header strip, or some indication of a shielded conductor) conveys useful information. The schematic is still primarily a way for the design engineer to communicate his ideas and his intentions to other humans. Those “mechanical details” may be superflous if we approach this communicative solely from the perspective of information theory, but if we think of it as a specialized form of human language then adding those details is no different than exploiting the richness, subtlety, or creativity of English, German, Mandarin or Swahili.

Dale

1 Like

Schematics were aids to service technicians long before they were used to generate netlists, so I see nothing wrong with using the D outline as a clue to somebody unfamiliar with the board as to which connector is which.
Many connectors are very badly labelled, with no numbering visible from outside the case, so the schematic is a logical place to illustrate

A schematic diagram is for showing electrical connectivity and has meaning for different people. You have “customers” that need to know what gender a mating connector should be. I extol you and plead with you to indicate gender. Do you show the package outline of a transistor? No, you use a generic NPN BJT symbol no matter that the XSTR is in a TO-92, TO-220, SOT-23, or whatever package. I do agree that you generally do not interface to a transistor like you do with a connector, but the principal is the same. (If you were a test engineer and needed to know where to place pogo pins on a test fixture to interface to a transistor then you certainly would need to know physical placement.)

I “retired” from Honeywell Aerospace and their internal standards called out the use of IPC-D350 (I think it was). Connectors, of any sort, were just rectangles. IPC-2612-1 also depicts any connector as a rectangle. I am on the IPC 2-40 committee and I am trying to get the IPC to acknowledge, and use, international standards that already exist.

On the subject of reference designators for connectors, everyone knows whether to apply the class letter P, J, or X according to ANSI/ASME Y14.44 (used to be IEEE 200), don’t you? P is for the most movable of a mating pair, irregardless of gender; J for the most fixed of a mating pair, again irregardless of gender; and X is for a socket. (There is no such class letters/classification of JP.) This works even when you have a gender less connector or a connector that has both genders Those who go by the ISO/IEC/EN (European) standards will have to refer to IEC 81346-1 and IEC 81346-2.

1 Like