X3T9.2/92-99 April 13, 1992 To: X3T9.2 Subject: Connector Agenda Item From: Dal Allan A number of things have happened since San Diego on this subject. The big news for connector suppliers is that AMP has agreed to provide an opportunity for everyone to show their connectors. This will make for an event like a mini-show. My apologies for the short notice. Connector vendors who want to be involved should not call me about the arrangements, contact Ken Schreder of AMP at 717-780-7699 or fax at 717- 780-4777 to let him know you want to participate. If no call goes to Ken, don't come complaining there was no provision for you. It is tough to arrange something like this on short notice, and Ken needs all the time he can get. NOTE: This event is sceduled for Monday evening, which is in conflict with the Serial SCSI session. Somehow or other John Lohmeyer and I will have to come up with a way to schedule around this. Attached is an extract of an email from Kurt Chan to me which prompted the idea of a joint show, and contains ideas about what he would like see in St Petersburg. Read closely, and please do as much as possible in line with his items. a) At the April plenary, have AMP, 3M, and whoever else wants to participate bring their high density ribbon cable assembly tool. b) After the meetings, have a connector "face-off": each supplier actually MAKES several assemblies in front of our eyes. Each assembly should have 6-7 connectors pressed onto it. c) After 5 or so assemblies are made, test each for adjacent shorts/opens and note the failure rates. Portable cable testers should be available for this purpose, or as a last resort I can provide a fixture for the AMP connectors. d) Have each supplier provide off-the-shelf pricing for, say, 5000 connectors, with lead times and phone numbers of individuals to contact for ordering purposes. With all the rhetoric surrounding the connector controversy in recent months, real data such as this would be a breath of fresh air. Item d) is not suitable for a standards meeting but is information that one company representative can provide another on a private basis. This question will be asked, so if you all are prepared to quote delivery and price on 5,000 pieces, it would be very valuable. Now that you have Kurt's list of desires, here are Dal's. Although we only specify the mating interface in the document, the backend is a decision factor. Mike Eneboe made the point in San Diego that second sourcing is not practical unless the backends are as compatible as the mating interface. Looking at connector parts is not a turn-on because how the various backends are mounted on PCBs is not intuitively obvious, and neither is the actual amount of board real estate needed for mounting. Could all those who plan to display please bring mounted parts? I would especially like to see multiple connectors mounted on the same card, such as a 6U VME card and a 3 1/2" disk drive card. The drives need dual port and there are an increasing number of integrators with two ports on a card. In fact, Sequent has a 4-port SCSI adapter (of which up to 8 may be combined in a single processor system). In the life of SCSI-2 and SCSI-3 you can expect such configurations to become increasingly common, so how close these connectors can be fitted together has to be considered. To all out there with unanswered or unasked questions on connectors, I encourage to use the reflector as your way to add what you want to the list that Kurt and I came up with. I do not want a connector war in which vendors pour their hearts out, especially since the purpose of this agenda item is to get the opinion of system integrators. System integrators, where are you? The only two we heard from in San Diego were the two who prepared the proposal for an alternative connector. To all system integrators out there. Come and tell us about your connector experiences, what you think about the choices in the standard, and what you think about the proposal that has been made. The message at San Diego was bring test results. We are only interested in data on the SCSI-2 high density connector and the proposed high density ribbon alternatives. I do not want to hear any of the old horror stories about pin and socket vs ribbon brought up in St Petersburg, there is enough emotional content in what we are doing as it is. Also in San Diego, the EIA project on the Burndy ribbon connector was mentioned. I checked for progress on the other choices and David Rundle, the Secretary of IEC/SC48B sent me a fax which states that the AMP tab and receptacle connector is in draft form and about to go out for review, and the 3M ribbon style is a proposed New Work Item. A note to companies making the connectors being proposed. To vote in June, we need a patent statement (if applicable) plus drawings and documentation on mating dimensions in a form ready to go into the SPI draft. This must be available to Larry Lamers in time to make the mailing which goes out after the April plenary. It would be helpful to have it in St Petersburg, but it absolutely has to be in Larry's hands by May 14 (which is before he leaves for the May working group). See you all in St Petersburg.