Skip to main content

I have a question about connecting 3 different digital devices together via wordclock bnc connectors.

I am connecting a digital mixer, an Echo Layla, and an PreSonus DigiMax mic preamp. I want to use the PreSonus as the "master", and then slave the Layla and the mixer to the Digimax. All 3 of these devices have both workclock ins and outs.

How do I go about hooking all of these up together? Do I need any "T" connectors or end terminators? Or do I just go out of the master, into #2, and then out of #2 into #3 (daisy chain)?

Topic Tags

Comments

sdevino Mon, 03/03/2003 - 15:59

You should really get a word clock distribution amplifier. Several companies like Lucid, Rosendahl and Aardark make them.

You would connect the Presonus clock out to the input of the dist amp. Then feed the other 2 from its outputs. this will provide each clock line witha clean, electronically buffered signal with proper termination at either end.

Your T's and daisy chains will result in a very unreliable if not unusable clock if the signals are not buffered correctly.

Steve

anonymous Fri, 03/14/2003 - 01:13

My post addresses the larger question of word clock boxes in general. I'd love to hear other peoples thoughts. Has anyone been blown away by the difference, or not blown away?

Well, I'm about to be a hypocrite because i do own and use a word clock generator. I have the Lucid box that will spit out word clock and super clock individually assignable on different outputs. Honestly, I have no idea if its helped the sound of my system, I can only hope that it is, I've never A/B'd it. I'm only running 3 outs from the Lucid though so not alot of gear would be getting daisy chained anyway.
So, as far as daisy chaining goes, I do understand its frowned upon, but i also know that some of the biggest and most reputable studios in New York City still daisy chain 888's like crazy and claim its fine.
Also, I recorded and mixed David Byrne's latest solo record (on an SSL) with a Pro Tools system that had 6 888's all daisy chained together (sometimes even a 7th 888). At the time i didnt even know Word Clock boxes existed. Long story short, I have gotten more compliments on how that record sounds than probably any other record of my career. So, although i have recently succumbed to the Word Clock box craze, I really think you are fine daisy chaining your system together. 3 pieces, probably not a big deal. Maybe it'll sound 2% better with a word clock box, question is, is that 2% worth the expense of the word clock box and will anyone else hear it.

Alécio Costa Fri, 03/14/2003 - 03:47

Hi!
I hav a Pro tools Mix system, with 2 adat bridges and one MTP AV connected via light pipe to an 02R.

sometimes I have to put the 02R as master clock and sometimes ( rarely) I end up with a better sound with PT as master. Couls someone explain that to me?
Hey, after I attach MONSTER CABLES word clock cables ( BNC) I felt a slight improvment in the sound.

anonymous Fri, 03/14/2003 - 04:11

I use an aardsyncII for a HDR, wz96 converter and a motu 308. The aardsync as a clock is much better then any other built in clock that I have.

I've done quite a few mixes trying the different clocks and configurations. With the aardsync a the master clkock, the mixes sounded clearer with more defination.

Daisy chaining well work but does not sound as good as distributed clock. My few pico seconds.

sdevino Sat, 03/15/2003 - 03:25

Originally posted by your ad here:
"You should really get a word clock distribution amplifier"

This is not necessarily true. In many cases daisy chaining is perfectly fine. Based on your description of gear, I would say yours is one of those cases.

Daisy chaining needs to done with the utmost attention to termination impedance. The wordclock is sent over a system witha charactorisitic impedance of 75 ohms. The source impedance of the drive output is 75 ohms. The open circuit signal level from the wordclock source is cut in half when properly terminated into a 75 ohm external load. It is very important to the over all quality of the word clock that the source impedance and the termination impedance be matched as close as possible, as any deviation results in excess VSWR which distorts the leading and trailing edges of the word clock. In order for daisy chaining to work reliably, you have to make sure that the transmission line impedances of the word clock signal chain are well planned.

If you use BNC "T" connectors to daisy chain then you need to make sure that all wordclock inputs in the chain are high impedance (about 1k ohms or greater), and that you put a 75 ohm terminator on the last "T" on the last box in the chain.

If you are daisy chaining wordclock via the individual box's word clock in to word clock out, then you want to make sure that the manufacturer has taken care of providing proper impedance isolation between in and out.

Another concern with wordclock output is the chips used to drive the lines on many pieces of pro gear cannot handle the inductive and capacitive loading of the cable once it gets beyond about 1 meter. If the loading is to great the wordclock signal shape gets greatly distorted causing converter jitter of even loss of word clock (pops and clicks).

I found that my Digidesign ADAT Bridge and my Alesis BRC were unable to drive more than a meter of good quality RG59 into a high impedance oscilloscope input.

Everytime you add a connector or additional cable you effectively increase the "electrical length" of the word clock cable and degrade the quality of the word clock signal.

OK , I was being deliberately anal and technical. If you did not follow what I just described then in my opinion it is not a great idea to daisy chain word clock unless you can hire a technical engineer who understands this whole concept. Some of this gear is really designed to be set up and wired by "engineer" engineers, not your typical audio engineer. If you want to daisy chain, then make sure your gear has well buffered word clock in and out AND that you use the shortest high quality cable you can get.

I do not use word clock distribution amps but I think they are a really good idea for the technical novice.

Doublehelix Sun, 03/16/2003 - 02:02

Thanks for the responses. Here is how I have decided to wire my setup for now, although I am looking to move to either a clock distribution system or a whole new clock generating system:

Word Clock out from the Digimax into the mixer, then I am syncing the Layla to the Digimax via the ADAT lightpipe. No daisychaining, plus both units are getting sync'ed to the same clock.

I have since heard that daisy chaining Word Clock souces can be problematic due to the impedience loading mentioned here by sdevino, but also that some devices do not *pass* the Word Clock signal, but rather accept the incoming signal, and then use their internal clocks to *regenerate* a new signal to pass along via the Word Clock out. This could add some latency in addition to the fact that I would now no longer be sync'ing to the same clock that I had originally intended.

Apparently, a lot of devices have the ability to disable this function of "clock regeneration", but I could not find it, so for all the reasons mentioned above, I have wired and clocked my system as described. It appears to be working fine so far.

anonymous Sun, 03/16/2003 - 09:22

I am achieving very stable clocking by daisy chaining three devices, however the last device in the chain is a Lexicon Core 32 which is being used as a standalone reverb and is built to be the terminal device in a chain (no BNC clock output). Just a thought......

The lucid Genx6 is a word clock that can send a signal to 6 different devices. These can be had for around $400.00 (48KHz version). A decent distribution amp is around $350.00. When I decide to add another device to my studio, I'm going to spend an extra $50.00 and get this as a dedicated clock rather than distributing a clock signal from one of my other devices. These can be purchased through Mercenary Audio.

Regards,

Doug Joyce
Animatrix Studios
Durango, CO

anonymous Thu, 03/20/2003 - 09:28

Hi,
This is a very interesting issue. I must say I'm a novice about wordclock. I've been thinking about it and maybe I can get it a bit clearer now.

I have a RME Hammerfall without a wordclock connection (available if I need), a Digimax (wordclock) and a Creamware A 16 Ultra (wordclock) (http://).

I've just used the stuff recording without any wordclock at all with nice results I think.

Would wordclock (whatever it is) make the sound better for me, do you think? And how should I connect it if so.

Hans,
http://

Opus2000 Sat, 04/05/2003 - 10:52

Hey guys

Of course most of you know I work for Apogee Electorincs as their technical support rep so I thought I'd chime in here.

Yes, indeed when it comes to only using two or three pieces of gear a distro box is not absolutely necessary.

The fact about latency being added is very true in daisy chaining as the process time for the word clock signal from the input to the output can vary between devices and in fact most of the time there is a slight lag. So, to correct this the proper way is using the BNC T connectors.

The first device acting as the master clock should not have a T connector, sending that out to the nect device with a T connector and then sending the output of that to the last device with a T connector and a termiator cap on the open end.

Of course the main thing to look at here is the clock being used for the master. Is it low jitter? Is it stable? These are very very important questions you need to find out.

Opus

Doublehelix Sat, 04/05/2003 - 14:14

Of course the main thing to look at here is the clock being used for the master. Is it low jitter? Is it stable? These are very very important questions you need to find out.

Excellent question Gary, now I have to ask, how can you determine how good a clock really is? Is there a test for jitter and stability other than to A/B compare a known clock to an unkown one?

I hear that a better clock will give a better stereo image to the music as well...

Opus2000 Sat, 04/05/2003 - 15:05

The only way to measure jitter is through a scope or a jitter measurement device.

Over time you will see the signal start to stray and add noise.

If you were to leave your "master" clock source on for a long period of time providing clock to the other deivces and just listen to "empty" sound, listen to see if you hear any oddity starting to creep into your speakers or headphones. Rarely will you hear this unless the clock source is sooooo bad that it's that noisy.

Yes, the tighter the clock signal the more stereo imaging you get, also depends on the circuitry used and how well it was designed. Also the power supply involved has a great deal to do with that.

Linear power supplies are superb at that but if done right a switching power supply can do the job right as well.

I will say this, the Apogee Big Ben clock is so low in jitter you can't measure it!

Opus

Doublehelix Fri, 05/30/2003 - 07:23

The first device acting as the master clock should not have a T connector, sending that out to the nect device with a T connector and then sending the output of that to the last device with a T connector and a termiator cap on the open end.

OK...with a "real" Word Clock generator or distribution system, is there a need to any "T" connectors and/or 75 ohm terminators?

Let's say with a clock that has 6 BNC Word Clock outs, I am clocking 4 different devices, and have 4 separate BNC cables going from the clock to the Word Clock Ins on 4 separate devices. Is this all I need, or do I need to add terminators???

Thanks!