Another connector gripe... (fwd)

lengfeld at austin.ibm.com lengfeld at austin.ibm.com
Wed Sep 28 13:12:18 PDT 1994


We have used a 68 pin "ribbon" style connector on a few adapters and
provided 1st device cables to convert to the SCSI standard 68 pin
connector with very good results on the "ribbon" connector. The "ribbon"
type has not experienced any bent pins since it has no pins. We have had
numerous reports and experiences with bent pins in the SCSI standard 68
and 50 pin high density connectors.

I was not involved in SCSI when the 68 pin connector was adopted, but I
am told that IBM made an effort to get the "ribbon" style adopted as
standard by showing improved reliability and mating longivity, but this
was not accepted by the committee. We are now moving to the SCSI standard
connector set on future adapters with mixed feelings, since our experience
has demonstrated more problems with the "pinned" connector.

     Thanks,
         Wendell Lengefeld
         SCSI System Development
         IBM, Austin
         lengfeld at austin.ibm.com





Forwarded message:
> From root Mon Sep 26 01:40:51 1994
> Subject: Re: Another connector gripe...
> To: SCSI Reflector List <scsi at WichitaKS.NCR.COM>
> Date: Sun, 25 Sep 94 23:52:30 MDT
> From: John Lohmeyer <jlohmeye at ncr-mpd.ftcollinsco.NCR.COM>
> In-Reply-To: <9409231534.AA17102 at adphdw20>; from "Tom Beach" at Sep 23, 94 8:34 am
> X-Mailer: ELM [version 2.3 PL11]
> Message-Id:  <9409260210.ab24054 at ncrhub1.NCR.COM>
>
> "Tom Beach wrote:"
> >
> > Agreed, also in evaluating SCSI devices with the 68 pin connector I'm
> > finding that there are two cable retention schemes. One uses a latch and
> > the other uses screws. As you might guess, these are totally and
> > completely incompatible. If you have a cable with the latch it won't
> > even mate with a connector with the screw retention. You can't even
> > TEST it!!!
>
> And only one is following the draft SPI standard, which only documents
> the jackscrew retention method.
>
> > Maybe it would be prudent for the committee to deselect this inferior
> > connection scheme and quicky (did I really use quickly and committee
> > in the same sentance?) move to approve a mini ribbon....
> >
> > There is precedent! Anybody remember the ORIGINAL SCSI shielded connector?
>
> Yes, I remember.  I'm the person that brought in the proposal to add the
> miniature ribbon connector to SCSI-1 (because of the bent pin problems we
> had with the AMPmodu connector).  To AMP's credit, they did come up with
> a solution to the AMPmodu problem, but it cost more than the miniature
> ribbon connector.  And the committee adopted the change in one meeting.
> It can be done.
>
> The last time the issue of bent pins came up on the Amplimite connector,
> I was assured by the AMP folks that the problem had not been reported by
> anyone else.  Apparently there are several more examples now.  I urge those
> people with bent pins to send samples of the damaged parts to Bob Whiteman
> at AMP (whiteman at cup.portal.com) along with any documentation of how
> the damage occurred, if available.  It would also be extremely interesting
> to find out whether the damage occurs with same vendor connectors or does
> it only occur with heterogeneous vendors?
>
> Jumping to one of the high-denity ribbon connectors would, of course,
> avoid the bent pin problems, but these connectors are not necessarily
> perfect either.  Ribbon contacts do not handle vibration as well as the
> tab-and-receptical style contact.  I think it is prudent to do a little
> failure analysis first.
>
> John
>
> --
> John Lohmeyer                      E-Mail:  John.Lohmeyer at FtCollinsCO.NCR.COM
> NCR Microelectronics                Voice:  719-573-3362
> 1635 Aeroplaza Dr.                    Fax:  719-597-8225
> Colo Spgs, CO 80916              SCSI BBS:  719-574-0424 300--14400 baud
>





More information about the T10 mailing list