On 11/30/2010 9:54 AM, Jim Brain wrote: > On 11/30/2010 5:17 AM, Marko Mäkelä wrote: >> On Tue, Nov 30, 2010 at 11:24:30AM +0100, Niklas Ramsberg wrote: >>> Is there a single line that can be switched out to make the drive >>> invisible to the computer? >> >> Cutting ATN should do the trick. All communications require ATN to be >> pulled down by the controller. If the drive always sees ATN=1, it will >> remain passive. >> > Because cbm-hackers is used as a reference resource, I thought I'd add a > bit, in case someone sees this later and tries to apply it to a > difference situation. > > Cutting ATN does exactly as Marko notes, the drive is "passive". But, > it is not truly invisible, at least from an electrical perspective. > > The other lines (DATA/CLK) are still connected to the bus and loading it > up (or down, depending on whether the drive is turned on or not). As > such, if you use such an "ATN-killer" switch on a set of drives, the bus > could become problematic even though the user thinks many of the devices > are "invisible". > > The cuplrit is the pullup resistors on each drive. When powered, the > resistors are dragging the lines high. When powered down, they tend to > drag the line to ground. > > For the topic of today (SX-64 drive), it's not a concern, but I thought > I'd put it out there in case someone gets the impression they can > connect every drive they own and just mark them "off the bus" by > switching off ATN. It won't work beyond a certain number of drives (I > think 8 is probably the limit, at least somewhere around there). Jim's analysis is good. I've found that with one drive that is off but connected to the IEC bus, it doesn't drag the idle level too low (3.3v high versus 5v). But it's better to always operate an IEC bus with all devices powered on and counting each attached device as a load (even if the device isn't talking). Don't try more than 4 drives + 1 printer as a max load on a single bus, and if using more than 2 devices, try to keep the cable length as short as possible. -- Nate Message was sent through the cbm-hackers mailing listReceived on 2010-11-30 22:00:08
Archive generated by hypermail 2.2.0.