Article View: comp.sys.mac.hardware.storage
Article #73Re: Powered-down IBM drive hangs bus
From: Dave Platt
Date: Mon, 27 Nov 1995 18:51
Date: Mon, 27 Nov 1995 18:51
26 lines
1619 bytes
1619 bytes
>It is recommended that SCSI devices with power off also meet the
>bus loading characteristics required of devices which are powered.
>
>However, I know from my own experience that some early drives did
>not conform to this.
My recollection is that the original SCSI standard specifically forbade
having powered-down devices connected to an active bus. Devices were
not required (or recommended) to be able to present a benign load on the
bus if powered down.
Quite a few early SCSI devices implement a form of bus overvoltage
protection for their transceivers - they tie the incoming bus signal
lines to their +5 supply via a clamp diode. If devices of this nature
are powered off, their +5 supply sinks down to ground, and the clamp
diodes clamp the signal lines to ground. The bus hangs.
The SCSI-2 specification does permit powered-off devices to be connected
to a bus, and devices which meet SCSI-2 must not clamp the bus if
powered off.
--
Dave Platt dplatt@3do.com
USNAIL: The 3DO Company, Systems Software group
600 Galveston Drive
Redwood City, CA 94063
Technical Details
Message-ID:
<1995Nov27.185140.23272@3do.com>Path:
arc1-ca.pugleaf.net!amdahl.com!ntg!dplattReferences:
<48vi2f$k3b@hppadbk.waterloo.hp.com> <492tej$bfu@news.htp.com> <499rqi$2bm@cucumber.demon.co.uk>Show Full Headers
Complete Article Headers
Newsgroups: comp.sys.mac.hardware.storage,comp.sys.mac.system,comp.periphs.scsi Path: arc1-ca.pugleaf.net!amdahl.com!ntg!dplatt From: dplatt@ntg.com (Dave Platt) Subject: Re: Powered-down IBM drive hangs bus Message-ID: <1995Nov27.185140.23272@3do.com> Sender: news@3do.com (USENET news account) Organization: The 3DO Company Inc., Redwood City CA References: <48vi2f$k3b@hppadbk.waterloo.hp.com> <492tej$bfu@news.htp.com> <499rqi$2bm@cucumber.demon.co.uk> Date: Mon, 27 Nov 1995 18:51:40 GMT