[Rick Ruskin]

I am using a Frontier Design Dakota/Montana PCI card duo for 32
channels of LightPipe in/outs so my computer (New Windows 7 Pro 64
bit) and Alesis HD-24 or TASCAM MX2424 can communicate. The 1st 16
channels of in/outs are from the HD24 to the Dakota. Channels 17-24
on the HD24 are to the ""C" ADAT ports of the Montana. Channels 1-8
of the MX24 are going to the "D" ADAT ports of the Montana.

Everything is fine until I connect the MX2424 ADAT out to the
Montana's "D" ADAT in. At that point all audio from the computer to
the Montana is terribly distorted regardless of whether it's going to
the Alesis or the TASCAM. Removing the Tascam's lightpipe out cable
from the TASCAM or the Montana eliminates the problem. The problem
comes back no matter which TASCAM or Montana ADAT port is connected.
What am I looking at here? Bad cable or should I have some sort of
sync between the Alesis and TASCAM? Any insight/advice much
appreciated.

[Scott Dorsey]

Rick Ruskin wrote:
I am using a Frontier Design Dakota/Montana PCI card duo for 32
channels of LightPipe in/outs so my computer (New Windows 7 Pro 64
bit) and Alesis HD-24 or TASCAM MX2424 can communicate. The 1st 16
channels of in/outs are from the HD24 to the Dakota. Channels 17-24
on the HD24 are to the ""C" ADAT ports of the Montana. Channels 1-8
of the MX24 are going to the "D" ADAT ports of the Montana.

Who is the clock master?

Everything is fine until I connect the MX2424 ADAT out to the
Montana's "D" ADAT in. At that point all audio from the computer to
the Montana is terribly distorted regardless of whether it's going to
the Alesis or the TASCAM. Removing the Tascam's lightpipe out cable
from the TASCAM or the Montana eliminates the problem. The problem
comes back no matter which TASCAM or Montana ADAT port is connected.
What am I looking at here? Bad cable or should I have some sort of
sync between the Alesis and TASCAM? Any insight/advice much
appreciated.

It would be good to make sure one device is the clock master and have
that device provide word clock for everything there. If you don't synch
to external word clock, you're _probably_ syncing to some input but maybe
you're not. Each device should have sync settings.

You can't synch to two different clocks at the same time, that just fails.

The most foolproof way of doing this is to have an external clock device
feeding everything, which makes diagnosis easy since as long as the
external clock is happy, any synch issue immediately can be tracked down
to not being locked to the clock. That's by no means the only way of doing
it but in a rapidly changing environment it's the easiest.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[david gourley[_2_]]

Rick Ruskin
said...news:hgir5c1j9610l62orp19i8n5ihu1ue13tb@4ax .com:

I am using a Frontier Design Dakota/Montana PCI card duo for 32
channels of LightPipe in/outs so my computer (New Windows 7 Pro 64
bit) and Alesis HD-24 or TASCAM MX2424 can communicate. The 1st 16
channels of in/outs are from the HD24 to the Dakota. Channels 17-24
on the HD24 are to the ""C" ADAT ports of the Montana. Channels 1-8
of the MX24 are going to the "D" ADAT ports of the Montana.

Everything is fine until I connect the MX2424 ADAT out to the
Montana's "D" ADAT in. At that point all audio from the computer to
the Montana is terribly distorted regardless of whether it's going to
the Alesis or the TASCAM. Removing the Tascam's lightpipe out cable
from the TASCAM or the Montana eliminates the problem. The problem
comes back no matter which TASCAM or Montana ADAT port is connected.
What am I looking at here? Bad cable or should I have some sort of
sync between the Alesis and TASCAM? Any insight/advice much
appreciated.


I'd vote for simple and easy - swap the cable first. Otherwise, how does
the Tascam perform with the Alesis completely disconnected? I thought
Lightpipe carried its own sync. I'll go out on a limb and ask if a UPS is
involved.

david

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

[Mike Rivers[_2_]]

On 12/23/2016 8:52 PM, david gourley wrote:
I'd vote for simple and easy - swap the cable first. Otherwise, how does
the Tascam perform with the Alesis completely disconnected? I thought
Lightpipe carried its own sync. I'll go out on a limb and ask if a UPS is
involved.

Actually, clock sync can be derived from an ADAT stream, which is sort
of weasel-wording to mean that the device you're connecting the ADAT
output to has to be able to derive its clock from the incoming data
stream. Some devices can do it, some can't.

But Scott is on the right track - when you have multiple devices, all of
the word clocks must be synchronized. If both the HD24 amd MX2424 can
both get a word clock from their ADAT input, then set them up that way
and you should be OK.

The HD24 has a word clock input but no output. The MX2424 has both word
clock in and out. So you could connect the word clock output of the
MX2424 to the word clock input of the HD24, set the MX2424 to internal
word clock and the HD24 to external (BNC) word clock. The Montana card
has a word clock input, and the Dakota card (apparently) can sync to an
ADAT stream. One might work better than the other. If you're going to
use ADAT sync for the computer, you'll have to specify one input port to
be the sync source.

If you're going to use the coax word clock (more robust, usually, than
ADAT sync), put a coax T on the sync connector on the Montana, and put
the card in between the MX2424 and HD24, and remove the 75 ohm
termination jumper P6 on the Montana card. The reason for this is that
the HD24 has a fixed 75 ohm termination on the word clock input. You
want that at the end of the chain. If you left the terminator connected
on the Montana, you'd be putting a 37.5 ohm load on the MX2424's word
clock output, which isn't very polite.


--

For a good time, call http://mikeriversaudio.wordpress.com

[david gourley[_2_]]

Mike Rivers said...news

Actually, clock sync can be derived from an ADAT stream, which is sort
of weasel-wording to mean that the device you're connecting the ADAT
output to has to be able to derive its clock from the incoming data
stream. Some devices can do it, some can't.

Yes I wondered about that too.


But Scott is on the right track - when you have multiple devices, all of
the word clocks must be synchronized. If both the HD24 amd MX2424 can
both get a word clock from their ADAT input, then set them up that way
and you should be OK.

Right.


The HD24 has a word clock input but no output. The MX2424 has both word
clock in and out. So you could connect the word clock output of the
MX2424 to the word clock input of the HD24, set the MX2424 to internal
word clock and the HD24 to external (BNC) word clock. The Montana card
has a word clock input, and the Dakota card (apparently) can sync to an
ADAT stream. One might work better than the other. If you're going to
use ADAT sync for the computer, you'll have to specify one input port to
be the sync source.

I agree, and also have an HD24. I only use it with an interface but not
with another deck which to sync.



If you're going to use the coax word clock (more robust, usually, than
ADAT sync), put a coax T on the sync connector on the Montana, and put
the card in between the MX2424 and HD24, and remove the 75 ohm
termination jumper P6 on the Montana card. The reason for this is that
the HD24 has a fixed 75 ohm termination on the word clock input. You
want that at the end of the chain. If you left the terminator connected
on the Montana, you'd be putting a 37.5 ohm load on the MX2424's word
clock output, which isn't very polite.

No, very touchy. I've used similar setups in radio labs, too (albeit
different terminations).

david

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

[Mike Rivers[_2_]]

On 12/23/2016 9:42 PM, david gourley wrote:
If you left the terminator connected
on the Montana, you'd be putting a 37.5 ohm load on the MX2424's word
clock output, which isn't very polite.

No, very touchy. I've used similar setups in radio labs, too (albeit
different terminations).

The problem with terminated vs. bridging word clock inputs isn't with
standing waves, it's with the clock input threshold level. This is
something for which there's no established standard, nor is it usually
specified. When you terminate a known source impedance with an equal
load impedance - the "right" was to do it with a transmission line - you
get half the open circuit voltage at the terminated end. This is a fact,
and not even Donald Trump can repeal Ohm's Law (though he might try). If
that voltage is too low for the clock input circuit to reliably detect
an "on" stage, clock synchronization will be unreliable.

When Mackie introduced the HD24/96 hard disk recorder, owners of
Mackie's d8b digital console with the optional word clock I/O card
discovered that word clock sync worked with the switch on the recorder
in the unterminated position, and believed that the switch was marked
incorrectly. The actual problem (I discovered and documented this) was
that half the console's word clock output voltage was too low for the
recorder's clock to synchronize reliably. The solution was to make the
recorder the clock master and sync the console to it.

Mackie owners, however, who had spent extra money to get a word clock
output and input on their console, were a bit disgruntled because that
clock I/O card was designed by Apogee (it was known among Mackie users
as "the Apogee card") and they figured that it was a really, really
great clock and therefore should be the master. But it also really good
as an input and, in reality, it made no difference which was the master.

Another issue with this clocking setup (remember, this was 1998-1999,
when few people had more than one digital device in their audio stream,
so there was no need for clock synchronization) is that both the Mackie
recorder and console used the same audio I/O cards, and the ADAT I/O
card didn't have the ability to provide word clock from the ADAT stream.
While you could connect the console to an Alesis ADAT by making the ADAT
input serve as the word clock slave, you couldn't do this with the
Mackie recorder - you had to add the optional word clock I/O card to the
console.




--

For a good time, call http://mikeriversaudio.wordpress.com

[Les Cargill[_4_]]

Rick Ruskin wrote:
I am using a Frontier Design Dakota/Montana PCI card duo for 32
channels of LightPipe in/outs so my computer (New Windows 7 Pro 64
bit) and Alesis HD-24 or TASCAM MX2424 can communicate. The 1st 16
channels of in/outs are from the HD24 to the Dakota. Channels 17-24
on the HD24 are to the ""C" ADAT ports of the Montana. Channels 1-8
of the MX24 are going to the "D" ADAT ports of the Montana.

Everything is fine until I connect the MX2424 ADAT out to the
Montana's "D" ADAT in. At that point all audio from the computer to
the Montana is terribly distorted regardless of whether it's going to
the Alesis or the TASCAM. Removing the Tascam's lightpipe out cable
from the TASCAM or the Montana eliminates the problem. The problem
comes back no matter which TASCAM or Montana ADAT port is connected.
What am I looking at here? Bad cable or should I have some sort of
sync between the Alesis and TASCAM? Any insight/advice much
appreciated.


I almost hate to ask this, because it sounds stupid, but I've had
it happen to me many times, so ...

Can you completely verify that all the clock hierarchy is correct?
You're familiar with what a buggered clock setup sound like, right?
( roughly like the audio is modulated by a ... pulse train for
lack of a better description).

Is it that distortion?

--
Les Cargill

[Rick Ruskin]

On Sat, 24 Dec 2016 17:24:11 -0600, Les Cargill
wrote:




I almost hate to ask this, because it sounds stupid, but I've had
it happen to me many times, so ...

Can you completely verify that all the clock hierarchy is correct?
You're familiar with what a buggered clock setup sound like, right?
( roughly like the audio is modulated by a ... pulse train for
lack of a better description).

Is it that distortion?


That's as good a description as any.

Having the HD24's word clock in connected to the MX2424's word clock
out did the trick. All devces are now happy and distortion is gone.

Thanks so all that responded.

RR