Behringer ADA8000 & 002R: Clock Settings

[Catoogie]

I have a Behringer ADA8000 that I sometimes but rarely use with my 002r, usually for a friend to transfer tracks from a friends DA88 into Pro Tools for him.

Everytime we get together to do a transfer I always forget what settings I should use. Can anyone remind me?

I'm using the ADA8000 as a slave right?

When the 002's clock is set to internal and I have the ADA8000's switch set all the way to the left I get nothng. When I set the switch to the right, either at 44.1 or 48 I hear audio but I can hear the clock clicking.....

Someone help.

[mikeyc]

Try slaving the 002r to the lightpipe, then just make sure the session sample rate matches the converter.

[protoolsman]

Slaving the 002 to the lightpipe would be the way to go BUT..
And this is a major but. The clock in the ADA8000 is very very weak. So best way would be to buy a decent (maybe even cheap) clock to make the results way better. There are some pretty decent clocks on the market for reasonable prices. I think Black Lion makes a clock that's unbeatable if it comes to value for money...

[Catoogie]

Thanks everyone, I've done this before I just can't remember how I did it in the past. I'm just curious why when I set the 002's clock to internal and try to use the ADA8000 as a slave why I get no signal. Anybody?

[mikeyc]

When you say you have the switch on the ADA8000 set all the way to the left, do you mean to "WC IN?" If so, then the ADA8000 is looking for word clock from the BNC connector. It's not going to find it, so it likely won't be converting anything to output. It should work on ADAT IN, as long as you have 2 Toslink cables connected.

When the switch is in the "Master" mode (44.1/48K) the two units are each running on their own clocks, not sync'd, thus the clicking.

[A.David.MacKinnon]

You can set it to slave from the ADAT in. That's how mine is running.

[newfuturevintage]

Thanks everyone, I've done this before I just can't remember how I did it in the past. I'm just curious why when I set the 002's clock to internal and try to use the ADA8000 as a slave why I get no signal. Anybody?

you have to run a 2nd lightpipe cable from the 002r adat output into the adat input on the 8000, and set the 8000's rear switch to slave to ADAT. You should then see the "Locked" LED on the front light up.

alternately, in the hardware setup in PT, select clock source to "optical" and optical format to "adat". Then set the 8000's rear switch to either 44.1 or 48khz, whichever your session requires.

[Andy Peters]

Slaving the 002 to the lightpipe would be the way to go BUT..
And this is a major but. The clock in the ADA8000 is very very weak. So best way would be to buy a decent (maybe even cheap) clock to make the results way better. There are some pretty decent clocks on the market for reasonable prices. I think Black Lion makes a clock that's unbeatable if it comes to value for money...

A high-falutin' external clock won't make a bit of difference with the ADA8000 because the Wavefront Semi converters generate the 256*fs modulator clock (the only clock that matters) from incoming sample-rate clock using an internal PLL, and that PLL is pretty jittery.

Having said that, running the ADA8000 slaved to external word-clock and using the 002 as master makes sense, because the recovered clock from the TOSLINK connection isn't very good either.

-a

[protoolsman]

Hey Andy; THANX! That was something I actually didn't now. I always thought that a master clock bypassed the internal clock...or something. I am looking into some other preamps without ADAT anyway as my Mackie SDR can record 24 tracks @ 96 through the analog ins. I used a millenia to test the diffenrence between 44.1 and 88.2 and it was pretty clear. Didn't like the millenia though.

[Andy Peters]

Hey Andy; THANX! That was something I actually didn't now. I always thought that a master clock bypassed the internal clock...or something.

Most people really don't understand how converters are clocked, and there's a lot of misinformation and salesman snake-oil out there, most of which is intended to make users think they need a high-dollar external clock.

Put simply: word clock runs at the sample frequency. But the vast majority of gear (not counting some boutique converters) have oversampling converters. These converters require a high-frequency (256, 384, 512 times the sample frequency; 512 times 48 kHz is 24.576 MHz) clock which drives the sigma-delta modulator and the FIR antialiasing filters. It is this clock which must have very low jitter. But the good news is that it's really pretty easy to get low jitter -- just clock the converters off of a high-quality crystal oscillator. You can buy, off-the-shelf, oscillators in the standard frequencies that spec at 5 ppm, if you're willing to spend more than a buck on an oscillator.

You need an external clock if you need to synchronize more than one device. So if you have three 8-channel converters that you'd like to run synchronously, you declare one as the master, which sources the word clock, and the other two slave off of it. Or you could use an external word-clock generator of your choice and slave all three to it.

So if you are slaving your converter to external word clock, your converter must have a high-quality phase-locked loop circuit that can generate that high-frequency modulator clock. That PLL must be low jitter while at the same time able to sync to the input clock which will have jitter of its own, and this is not a trivial design.

Sum: you can never do better, clock-wise, than using a device's internal crystal oscillator as the clock source.

-a