[slinkelist] Master-Slave / Track Restart / MD Manager

[Hoen, Dave]

For those of you who have implemented the Master-Slave modification to
your Slink-e, you may be interested in the following problem.  For those
of you who haven't, no need to read on, this doesn't apply to you.

The last couple of versions of CDJ have not worked correctly with my
Master-Slave circuitry.  This is the circuitry recommended on the Nirvis
website for those of us who want to be able to use CDJ to control our
sound system or let our system run standalone in a Sony Master-Slave
environment.  Up until the last few versions of CDJ, this circuitry
worked very well and more or less solved most of my mysterious CDJ hangs.

Here are the symptoms with the latest software (08/07):
After running CDJ for an hour or so, I will start getting track restarts
or track aborts.  From that point on, about 1 in 5 tracks will have
restarts or aborts.  (A restart is starting a track over after it has
already played for several seconds.  An abort is stopping a track after
it has played for a few seconds, and going on to the next track.)  The
problem seems to be exacerbated if I run MD Manager.  Using MD Manager
the problem usually happens within two or three tracks and usually halts
playing after the current track.  I have seen it occur in as short as one
track and as long as 10 tracks, however.

What I have discovered:
The DIO 0 signal from the parallel port of the Slink-e is used to switch
from Master-Slave environment to CDJ environment.  When you start up CDJ,
this signal is set to logic high and is used to energize the relay.  When
you shut down CDJ, this signal is cleared to logic low and de-energizes
the relay.  At least that is how it is supposed to work.  What is really
happening, is while CDJ is running, this signal is being cleared to logic
low.  This reconnects the s-link busses, and things go amok.

What I have done about it:
Although it may be something peculiar to my setup, I don't know how
anything in my setup could possibly be changing the state of the DIO 0
signal.  I have been in e-mail conversation with Colby, so he is aware of
the problem, but may be low-priority.  Which I can understand
completely.  In the meantime, I have installed a switch across the
collector and emitter of the transistor used to energize the relay.  If I
want to run CDJ, I close the switch.  If I want to run standalone, I open
the switch.  It is a little more work, but everything is working great
again, especially MD Manager.  Hopefully the cause can soon be found, and
I can just leave the switch open.

Dave Hoen

p.s. In addition to the normal slinkelist, I am copying this to several
people (on the slinkelist) who have reported similar problems.  Thanks
for your suggestions, and hope this might solve some of yours.