Multiple interface clocking question - Avid + Mytek + Apogee

[dtonthept]

Hey all,

I've been running my PT11 HDX system with a Mytek 8x192, an Avid HDIO 16x16 analogue and an Omni for some time now. They talk happily and follow session sample rates when I have my Mytek set as the master with internal sync, then a word clock out going to the HDIO Loop Sync in, then HDIO Loop Sync Out going to Omni Loop Sync In.

Strangely enough I'm not looping the Loop Sync, but this seemed to be the only way to get all the interfaces to talk happily. The MYtek has 8 word clock outs, but running a direct link from two of these to the word clock inputs of the two Avid interfaces didn't work.

I've got an old Apogee Rosetta 800 lying around, just updated the X-HD card to the version 3 firmware.

Since then I've been unable to figure out a way to get the Apogee integrated into the system. I've tried tagging it on to the end of the Loop Sync, and running it from another Mytek Word Clock Out, and also with and without 75 Ohm termination.

Pro Tools will see the Apogee sometimes when I fire it up with all the interfaces on, but it always sees it if I have solely the Apogee connected.

My HDX card has one port going to the Mytek then chaining to the Apogee, the second port goes to the Avid interfaces. The system runs fine with just the Mytek and Avid interfaces. I can just switch off the Apogee and it all runs as per usual.

From a clocking point of view is this insanity to run interfaces from three differenct manufacturers?

I'm also unsure about exactly how to set the Apogee's front panel for this setup - I'd assume it should be set to external sync, WC in, with the source to analog out being Option.

Does anyone have any suggestions?

Any advice, be it just word clock, or Apogee settings for HDX, most gratefully received!

Thanks so much,

Damian

[Bill Denton]

Several possible issues...

Loop Sync should be used only with Avid interfaces, which means Loop Sync is out of the picture.

If you want to use your Mytek as the master, it would be set to internal, which you've already done.

You would then run a 75 ohm cable from a Mytek word clock out to the word clock in on each of the other devices. Make sure and engage the termination on the device (if it has one), or use a tee and a 75 ohm terminator.

If you then set all of your devices to use external you should be good to go...

[JFreak]

Originally Posted by dtonthept

I've got an old Apogee Rosetta 800 lying around, just updated the X-HD card to the version 3 firmware.

Since then I've been unable to figure out a way to get the Apogee integrated into the system. I've tried tagging it on to the end of the Loop Sync, and running it from another Mytek Word Clock Out, and also with and without 75 Ohm termination.

Apogee's is not compatible with Loop Sync. I'm pretty sure you don't need termination, but have it on regular wordclock chain and press the buttons on the front panel as long as you get that lock (sometimes, strangely enough, you need to cycle the settings even though it seems you have proper settings to begin with. it just won't lock to it when you power it up, but when you cycle the settings then it locks).

[dtonthept]

Thanks for the replies so far - just a quick question - should the Avid interfaces all be happy running off word clock from individual WC outs from the Mytek (or indeed any other master clock) as opposed to running Loop Sync?

From memory they weren't behaving right when I did this, but I may well have had something wrong.

I'll hope to have another fiddle with this later tonight...

D

[JFreak]

As long as there is only one master clock in the chain, everyone should be happy. Loop sync is for Digidesign/Avid interfaces.

[Bill Denton]

One thing I didn't mention in my previous post (I didn't have enough time that day)...

Very Important...You can not use multiple devices' (interfaces') WC-Outs* to WC-Ins* to "daisy chain"! *WC is Word Clock.

Here's why...

The WC-Out is not a "pass through" of the device's WC-In. There is no physical connection between a single device's WC-In and it's WC-Out.

Both the "timing" signal used in the device's circuits, and the one sent to the device's WC-Out, are generated by the device's internal clock.

The device's internal clock uses a "time base" to generate the time used by the device and sent out the device's WC-Out. If the device is configured to use "external" or something like "WC-In" as its "clocking", it will use the signal from the WC-In as its "time base". Otherwise, it will use the device's internal "time base" signal.

Given the vagaries of digital audio systems, the signal sent out the WC-Out will typically not be 100% "in sync" with the signal received from the WC-In.

If you have several devices "daisy chained" by using WC-Out to WC-In, you will be picking up errors at every device in the chain.

The connection protocol will then be: "master" clock's WC-Out to a properly terminated WC-In on the "slave" device with nothing connected to the "slave" device's WC-Out.

Further, you can only have one connection to each of a device's WC-Out connectors. If you have three devices, you will need two WC-Outs. Many interfaces now offer multiple WC-Outs, or you can use a discrete master clocking device.

Hope this is useful...

[JFreak]

Originally Posted by Bill Denton

you can only have one connection to each of a device's WC-Out connectors

That is not exactly true. You can split the WC signal, but each passive split drops signal level so depending on the tolerances of any given box, there is a limit to how many splits you can do before the signal level is too low.

After all; WC signal is a digital signal which is either on or off, and that is interpreted by voltage levels. If the signal level is high enough to be considered "on", then it is "on" no matter if it's just barely high enough or 100 times more powerful than needed. Binary logic. 0 or 1.