I have a mobile studio setup featuring an ADAT (blackface), an EMU
Darwin 8-track harddisk recorder with ADAT extension and a Behringer
3216 DDX mixer with ADT1616 dual lightpipe interface.

Currently this studio setup is in beta-testing and I'm facing some
problems for which I currently have no explanation. When the band (who
are fortunately very aware that they're beta-testing the setup) starts
a certain song, the blackface ADAT always loses sync at the first beat
all band members play together. That is, tape is put in record, all
tracks armed, some talking and stuff is going on and at a certain
moment the drummer beats his sticks together-- no problems until then
but as soon as all play the first 'real' beat of the song the blackface
loses sync (Err 8) Also, frequent digital peaks appear on the meters of
both ADAT and EMU darwin. These peaks are not present on the level
meters of the mixer.

The ADAT blackface is configured as master. The EMU darwin harddisk
recorder acts as slave (this same EMU recorder served me well for years
connected as slave to an ADAT XT but with an analog mixing console).
During playback, things sync fine- the problem only shows in record
mode.

Both blackface and EMU are connected 2-way to the behringer mixer via
toslink lightpipe interface (in total 4 optical cables in "star"
configuration). Blackface has ADAT sync out to the EMU darwin's ADAT
sync in.

Equipment is switched on in the following order: Blackface, Darwin,
Mixer.
Blackface is master, Darwin detects it needs to slave, Behringer syncs
with blackface clock via lightpipe. The Behringer ADT1616 interface
module of the mixer does not have 9-pin ADAT sync out.

What surprises me is that the blackface ADAT gives an error 8- being
the
master it should not need synchronizing to anything than to the mixer
whose clock is being fed by the clock of the blackface itself.

Does ERR 8 on the blackface mean that it lost sync to the Behringer,
or did it detect the EMU didn't sync right to it? Why does this appear
to be related to the audio input signal? Have I connected something
wrongly?

I do suspect the quality of the sync cable is not very good, but it
doesn't seem to make sense that this would cause an ERR 8 on the master
ADAT- it would be logical for the slave to lose sync. Also, it looks
like the EMU darwin syncs fine to the blackface ADAT and sync is only
lost while recording, not during playback.

Any ideas what could be causing this strange behaviour and what would
solve it? Is a 9-pin ADAT sync output needed on the Behringer ADT1616
interface? (it doesn't have one, is this a design flaw?)

Best regards,

Marc Brevoort

wrote:

I have a mobile studio setup featuring an ADAT (blackface), an EMU
Darwin 8-track harddisk recorder with ADAT extension and a Behringer
3216 DDX mixer with ADT1616 dual lightpipe interface.



What happens if you make the mixer the master, and slave both recorders
to it?

--
"It CAN'T be too loud... some of the red lights aren't even on yet!"
- Lorin David Schultz
in the control room
making even bad news sound good

(Remove spamblock to reply)

Lorin David Schultz wrote:

What happens if you make the mixer the master, and slave both recorders
to it?
That would be ideal, but the mixer cannot be set as master.
What can be done is to set the mixer to a fixed (internal) clock of
either
44.1 or 48 khz, however setting the mixer to internal clock doesn't
turn it into an ADAT master. (Note: the DDX3216 does however have
separate wordclock out/MIDI out, MIDI out stream contains both MMC and
MTC. Neither ADAT nor EMU have wordclock in; the EMU has MMC and MTC
support though). If the mixer is running at internal clock, the
blackface ADAT won't recognize it as
master, decide to be master itself, and also use its own internal
clock.

I have an explanation for this: It is common in an ADAT setup to to
have lightpipes connected as ring (e.g. the lightpipe of the last slave
connects back to the master) which allows routing from any ADAT to any
ADAT over an optical 8 channel bus. Because of this, it makes sense
that master/slave status is detected by ADAT machines by looking at
sync in; when something is providing sync in, it is the master. As
mentioned, the mixer does not have ADAT sync out, which is why I wonder
if this is a design flaw. If so, it'd be nice to find a workaround,
perhaps using the word clock of the mixer? I seem to have read
somewhere that ADAT sync is something like a combi of
wordclock+MIDI+MMC.

Regards,
Marc Brevoort

Looks like I finally found the cause of the problems. As it took long
enough to isolate, let me answer my own question, so that the next
might not have the same trouble that I had.

Pressing 'digital in' on the ADAT did not (only) refer to the origin of
the audio signal, but to the clock source as well.

As a result, with the mixer set to 'use external clock' and 'digital
in' pressed on the ADAT, the clock would drift further and further
off-sync: The ADAT used the clock of the mixer which used the clock of
the ADAT which used the clock of the mixer...... not a very good way to
sync.

Setting the mixer set to 'internal clock' (rather than external, as
expected) and the master ADAT on 'digital in', recording to the first
ADAT went OK; it simply followed the clock of the mixer.

However, there was a second ADAT (compatible device). With the second
ADAT now receiving digital audio directly from the mixer but the
(probably slightly lagged) clock signal from the first ADAT, this had
the effect that things didn't end up in their correct tracks. I needed
to look further.

As it turns out (after once again reading the manual), when 'digital
in' is selected, the ADAT can choose to either use lightpipe as clock
source or its internal clock. Needless to say, it was set to lightpipe.
Pressing "Set locate/digital in" switched it to internal clock source
which was the way I had assumed things had to operate from the start
(it seemed logical to me that an ADAT master would provide clock to all
other machines, lacking ADAT sync out from the mixer).

This meant that the 2 ADAT (compatible) devices now synchronized
properly to one another, but the mixer was still happily using its own
clock.

Of course this was solved after setting it back to use the external
clock source.

In article .com writes:

Setting the mixer set to 'internal clock' (rather than external, as
expected) and the master ADAT on 'digital in', recording to the first
ADAT went OK; it simply followed the clock of the mixer.

As it turns out (after once again reading the manual), when 'digital
in' is selected, the ADAT can choose to either use lightpipe as clock
source or its internal clock. Needless to say, it was set to lightpipe.

Aha! So you finally read the manual, or finally understood what
choosing Digital In as the sync source means.

However, there was a second ADAT (compatible device). With the second
ADAT now receiving digital audio directly from the mixer but the
(probably slightly lagged) clock signal from the first ADAT, this had
the effect that things didn't end up in their correct tracks. I needed
to look further.

You need to set the second ADAT so that it gets its word clock through
the sync cable. Then it will be in sync with the first ADAT, which is
in sync with the data stream coming out of the mixer.


--
I'm really Mike Rivers )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me he double-m-eleven-double-zero at yahoo

As it turns out it was a compound problem -- not *just* a matter of
synchronization but also a defective lightpipe in port in the master
ADAT. Currently this is being taken care of.