Updated FC-TAPE & DISK Connector w/8 channels

JoeBre at Exabyte.COM JoeBre at Exabyte.COM
Wed Jul 28 12:43:49 PDT 1999


* From the T10 Reflector (t10 at t10.org), posted by:
* JoeBre at Exabyte.COM
*


-----Original Message-----
From: Bob Snively [mailto:Bob.Snively at EBay.Sun.COM]
Sent: Wednesday, July 28, 1999 11:05 AM
To: Bob.Snively at EBay.Sun.COM; curtr at lsil.com;
Paul_A_Suhler at notes.seagate.com; JoeBre at Exabyte.COM
Cc: sff_reflector at sun.com; fc at nsg0.network.com; t10 at t10.org
Subject: RE: Updated FC-TAPE & DISK Connector w/8 channels



> > > drives.  There are at least four separate connector mounting systems
that
> > > have been worked out to allow the board to be in different locations,
> > while
> > > keeping the connector in a fixed location.
 
> > Are any of these connector mounting systems as space-efficient as
directly
> > mounting the connector on the circuit board (as viewed from the drive's
> > perspective)? We are currently grazing volumetric limits in what can be
> > achieved in a 5.25", Half Height form factor. 

> Joe,  

> 	These schemes actually improve volumetric efficiency.  The systems I
> 	know about include: 
	
> 		pin-through hole above card
> 		smt above card

Actually, both the above schemes fill more volume, unless the connector body
sits directly upon the card, with no vertical displacement. This also does
not allow for horizontal displacement. This also assumes that connector
manufacturers will be ready, willng, and able to manufacture connectors with
arbitrary displacements in two dimensions.

> 		straddle card
> 		right-angle on paddle card (which makes the vertical axis
> 			available for both passive and active parts).

I'm sorry, but I cannot imagine any scheme wherein either of the above
methods is not necessarily less efficient from a volumetric standpoint. In
both cases, there still exists the original connector, and a second
connector has been added to connect the two boards together.
	
> 	I am pretty sure that there are several variations on each of those
> 	allowing significant displacements with respect to the card
surfaces.

 
> > > The reason SCA-2 is a success is precisely that the drive connector is
in
> > > a fixed location relative to key mounting holes.  Admittedly, tape
drives
 
> > So it appears to me that you plan on connecting tape drives directly to
a
> > backplane? Or are there other benefits of fixing the location of the
> > connector that I have not envisioned?

> See Jim McKain's letter and Arlan Stone's letter.  We all want the
> capability of sliding drives straight into a slot with a connector at
> the back.  They want that capability for all tape drives from all
> manufacturers that would be about the size of that slot.  They probably
> want to design common rail, guide and retention  hardware among all
> those vendors, as well, so it is critical that front dimensions and
> mounting hole locations and dimensions be constant for each class of
> tape drive (3.5" any height and 5.25" any height). Whether they choose
> to run a backplane to the back of the slot, a flex circuit to the back
> of the slot, or a cable to the back of the slot is part of their added
> value.  I personally would usually use a backplane, depending on the
> physical requirements.

> Bob

I understand the motiviation for this requirement, and it is a fairly
compelling vision. I just am skeptical as to whether it is truly
implementable. As the market stands today, many (most?) tape drives exceed
the envelope of the form factor they claim to be. OEMs that insist on having
the connector at a certain spot may have to give up either features or
form-factor (or jump from half-height to full height, for example), due to
internal device volume constraints.

Joe Breher
Exabyte Corp.
*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org






More information about the T10 mailing list