SPI-2/Fast-40 pins

Lohmeyer, John JLOHMEYE at cosmpdaero.ftcollinsco.ncr.com
Tue May 23 15:41:00 PDT 1995


Jim McGrath wrote:
>Actually, the original 6 m single ended SCSI limit was probably a
>REQ/ACK type limit as well, so maybe we can go longer.  Does anyone
>remember (John?) and/or have done empirical tests?

SASI and SCSI-1 specified 6 m for single-ended because of signal quality 
issues -- not timing problems.  So if we use LPD on all the signals, we 
should be able to go as far as we want up to the signal quality limits we 
design into LPD or the arbitration limits (around 125 feet), whichever is 
shorter.

If we elect to save some pads and have a mixture of LPD and single-ended, 
then we will have to determine which driver/receiver technology becomes the 
limiting factor: LPD on the fast signals or single-ended on the slower 
signals.

I recall that Bill Spence once ran single-ended SCSI on very long cables at 
T.I. (over 60 feet, I think).  He had a carefully controlled environment. 
 So the 6 m limit is not etched in stone for single ended.  It really 
depends on a lot of factors.

I believe that the data and parity signals must use the same technology 
drivers/receivers as for REQ/ACK.  Otherwise, it will be extremely difficult 
to control the setup and hold times.

I tend to favor a total LPD solution because it gives us better common-mode 
noise immunity.  And, yes, DIFFSENSE, which is an SE signal, enters into 
this equation, but we can filter it extensively -- it is extremely low 
bandwidth.

John
 --
John Lohmeyer             E-Mail:  john.lohmeyer at symbios.com
Symbios Logic Inc.         Voice:  719-573-3362
1635 Aeroplaza Dr.           Fax:  719-573-3037
Colo Spgs, CO 80916     SCSI BBS:  719-574-0424 300--14400 baud




More information about the T10 mailing list