clocking my converters... (ada8000)

Hello,

I have a setup with 2x rme cards and 4x ada8000 converters.

my rme multiface 2 is set as clock master and then I distribute the wordclock signal with some y-adapters to my ada8000 converters...

I read some posts from mr. lavry in another forum, he claims that it is not so easy and you have to deal with termination, signal reflection, and so on....

What I'd like to know is, if it's not better to clock the ada8000 converters over the optical-adat-cable...
or are there any other disadvantages?

does anyone here know what to prefer and why??


thank you very much.

mat

You treat it like a video signal. Use the proper 75-ohm coax and connectors, and make sure it's only terminated in 75 ohms at the end. Do your converters have switchable terminations?

[quote author="matthias"]What I'd like to know is, if it's not better to clock the ada8000 converters over the optical-adat-cable...
or are there any other disadvantages?[/quote]
That's a tough one.

As far as I can tell, there are three ways to clock an ADA8000: internal, word clock and ADAT. As with any converter, the internal clock will give best jitter performance, but that's not an option if you need to have multiple units running in sync. The ADA8000 uses a simple 4046 CMOS PLL with what amounts to an RC VCO to sync to WC, or the AL1402 for ADAT clock. Neither look particularly good from a jitter POV.

So which is best ? Hard to say. The optical link is inherently more jittery than a 75-Ohm coax cable, and the Q of the external capacitor on the 4046 is probably higher than the VCO cap or ring oscillator that's integrated in the AL1402. Then again, the ADAT signal has way more transitions than the Word Clock, which could conceivably yield lower jitter.

If you want you can test which setup gives lowest jitter. Hook up one of your ADA8000 channels in loopback (say, ADA8000 output #1 directly to input #1). Create a 16kHz sine wave in your DAW with an amplitude that's just below clipping (~0.5dB is fine) when transmitted over this loopback setup. Now do three full-duplex loopback tests:

1) with the ADA8000 running from internal clock, and the RME slaved to this. For this test it doesn't matter if you sync the RME to WC or ADAT. This is your baseline.
2) with the ADA8000 locked to WC from the RME
3) with the ADA8000 locked to ADAT from the RME

Get yourself some analysis software that will do a high-resolution FFT analysis. Maybe your DAW has it integrated, otherwise you could probably trick RMAA into it, or http://focus.ti.com/docs/toolsw/folders/print/adcpro.html (thanks 12afael for that link). Compare the frequency plots of all three experiments, and pick WC or ADAT sync depending on which of (2) or (3) is closest to (1). In my experience peaks in this spectrum are 'worse' (ie more audible) than an overall rise of the noise floor.

Random notes:
- these results are valid for the setup that you tested only. Switch between 44k1 and 48k, or run your desk chair over the WC cable a few times and the outcome may well change.
- you want to use as high a test frequency as possible, as jitter artifacts become more visible for higher frequency. Then again, you want to stay well away from the LPF corner set by the antialiasing filter. IME 16kHz is a good compromise.
- my software suggestions are rather vague as I do most of my FD testing with custom tools on UNIX systems, and have little experience with pre-cooked solutions on either Windows or MacOS. If you want I can dig up my source code, but be warned that it's pretty ugly code with few comments, never intended for an external release.

Good luck, and do let us know what you come up with,

JDB.

[EDIT 20080902: Fixed the test procedure by changing 'external clock' to 'internal clock'.]

How about using a single ADA8000 as a master and the rest as slaves?

thank you very much for the detailed reply...

I'll try this now....

Please post your results.

I am using 2 ada8000's with 2 terratecs over optical with the terratecs as masters. I would like to know what you find.

:Ron

Has anyone else ever studied this further?

Just picked up a Behringer ADA8000 ADAT converter to add 8 inputs to my Motu 828 mk2 and am wondering what might be the best setup. I've borrowed some coax and a t-piece and terminator from work, so can sync via wordclock. Recording on Saturday, so it would be good to hear any opinions.

Roddy

THE PROBLEM NEED A CENTRAL CLOCK MASTER GENERATOR!

you can reach a compromise about but ever as a nightmare the jitter
run on the signals

only a unit with 0,5 ppm (the best at today.....)
can kill the nightmare!!!
as
rosendahl hd
mutec ic clock hd
believe me
:wink:
have an ice :guinness:

[quote author="SIXTYNINER"]THE PROBLEM NEED A CENTRAL CLOCK MASTER GENERATOR![/quote]
No need to shout.

While a central or master clock is an answer, it is one to a completely different question.

In general, using the converter's internal clock will give best jitter performance. In cases where that's not an option (like the situation which Roddy describes) you'll need to have some kind of clock linking. The question at hand is if the ADA8000 is least sensitive to jitter when slaved to WC or when slaved to ADAT. Whether the reference clock is supplied by the Motu or by a dedicated master clock has no impact on this.

Roddy, I suggest you try the procedure outlined above, if you have the time, as the performance of the entire system depends on, well, the entire system. If I were in your situation and I didn't have the time to test, I'd get a good WC cable (where 'good' equals 'not been run over by a desk chair several times') and gamble on that.

JD 'six of one...' B.

You should be able to take the wordclock out of the ADA8000 and run it into the Motu 828 mk2 . Set the Motu 828 mk2 to external wordclock and the correct conversion frequency. Set the ADA8000 to Master and the correct conversion frequency. Put the T on the wordclock in of the Motu 828 mk2 and terminate the cable there. I am going to try to do this with a Focusrite Saffire which is a similar unit. This is the method they recommend in their manual. I'll let you know how that works out.

[quote author="burdij"]You should be able to take the wordclock out of the ADA8000 and run it into the Motu 828 mk2.[/quote]
While that will work, it will also degrade the jitter performance of the Motu's converters. Then again, that degradation may be low enough for you not to notice or care.

From a technical POV, when you start out with a decent crystal oscillator in your converter it's not possible to improve its jitter performance through locking it to an external signal. The best you can hope for is to not have things deteriorate too much, which is just about doable with a low noise PLL design or a proper ASRC implementation. The ADA8000's clock sync is not very good in this respect, I have no idea how well either the Motu or the Focusrite handle this specifically.

Personally I'd slave the ADA8000 from whatever your main converter is. I've just gone over the ADA's schematics, and its converters have a pretty noisy on-chip clock PLL. It is highly likely that the Motu has more to lose jitter-wise. If you want to gild the lily some more, dedicate the Motu to your most 'precious' channels, the ones with the most hi-freq content and/or the sharpest transients, as that is where jitter is most noticeable.

But like I said, depending on your source material and your customers/audience, the impact of who-clocks-what may well get lost under the noise floor.

JDB.
[and then there are those, mostly master clock salesmen, who claim that their clocks inject euphonic jitter, providing what is effectively frequency domain dither. While this may be true for some clock-converter combinations it is impossible to hold for all, and I personally prefer to do my tracking without any such blanket predistortion]

Thanks for your suggestions.

JD, sounds like you know your stuff which is nice to have around here as this place tends to be quite analogue-biased. I never believed too much about the difference between converters and clocking, etc. but recent comparisons have surprised me a lot.

Let's keep this ball rolling, the theory behind it in combination with practical evaluation of proposed setups [salesman-crap]can only lead to informed decisions and sonic-bliss[/salesman-crap] well, what I wanted to say, please keep us posted Roddy about your findings :wink:

BTW, got your ADA8k going yet JD ? Here I'm using the original one I had, but the second hasn't seen a soldering iron yet.

Bye,

Peter

[quote author="rodabod"] recent comparisons have surprised me a lot.[/quote]
Hi rodabod!
Could you elaborate on what you were comparing with what?
I, too, have a 828+ada8000 sitting here, havent tried hooking them up yet...
I rechipped one output in 828 and the sound improved alot, so I plan to rechip the rest whenever I get to it...
It'd be interesting to hear about what can be done with clocking here..

[quote author="clintrubber"]Let's keep this ball rolling, the theory behind it in combination with practical evaluation of proposed setups [salesman-crap]can only lead to informed decisions and sonic-bliss[/salesman-crap] well, what I wanted to say, please keep us posted Roddy about your findings :wink: [/quote]
Indeed. Maybe it's even time for a digital-META. I'll see if I have time to collect links to a few threads.

[quote author="clintrubber"]BTW, got your ADA8k going yet JD ? Here I'm using the original one I had, but the second hasn't seen a soldering iron yet.[/quote]
Nah, mine is still in storage. One of these days, though...

JDB.
[my newly acquired Multiface may just give me enough of an incentive to get that ADA8000 working]

OK, some practical experience from my side (no measurements, just listening tests with very good monitors)
I tried several setups and always use my main AD clock as master, in this case very good ones (Mytek ad 8x96 or Swissonic AD8, to my surprise, both clocks sound equally good as master clocks in identical environment - really good indeed).
DA's tested are Swissonic DA24, DA96 and DA96MK II as well as ADA8k.
The Swissonic DA's really improve from beeing WC'd by the AD clocks
(which leads me to the assumption that the clock of Swissonic DA's is worse than the WC signal of Swissonics and Myteks AD clocks...hmmm)
However, the ADA8k DA's benefit as well from beeing WC'ed by the Master AD.
I use T adapters and 75 ohm terminators on both end of the bus.
generally, use bussing, not daisy chaining. on busses shorter than say 5 meters you may get away without Terminators and get no perceivable loss due to this (this is very similar to bnc ethernet bussing, I guess, where you *might* get away without terminators up to 10 meters)..
Please note, I do use ADA8k for DA only, I *might* choose lightpipe clocking when using AD's as well, hmm, but I never did right now....

Kind regards,
Martin

PS: Great explanation, jdbakker!!! :thumb:

OK some additional thoughts for you:
When using ADA8k for recording (more than a few additional tracks) I preferred lightpipe only in the past! It is incomplicated and sounds OK really.
Here you may not benefit from WC from my experience ! (Say if you record 24ch and the 16 you do with ADA8ks contain essential material)
Please don't ask me for the theory behind this....it's my subjective experience.
As for dedicated Word Clock (distributions) I do not 'believe' in it for small to medium setups. If you record, it *will* degrade your master AD. Always.
Also, dedicated word clock modules are often heavily overpriced in my opinion....
Clocking is always a compromise if you have a setup with several and different model converters - and marketing theory is not what you listen to...

Again, heavily subjective opinion....

Kind regards,
Martin :wink:

[quote author="jdbakker"][quote author="clintrubber"]Let's keep this ball rolling, the theory behind it in combination with practical evaluation of proposed setups [salesman-crap]can only lead to informed decisions and sonic-bliss[/salesman-crap] well, what I wanted to say, please keep us posted Roddy about your findings :wink: [/quote]
Indeed. Maybe it's even time for a digital-META. I'll see if I have time to collect links to a few threads.[/quote]
Nice :thumb:

BTW... :wink:

Nah, mine is still in storage. One of these days, though...

Click to expand...

Let's see which one of us has it running first... I wouldn't bet on myself though :wink:

[my newly acquired Multiface may just give me enough of an incentive to get that ADA8000 working]

Click to expand...

Nice. Multiple ADAT-I/O by chance ?

Bye,

Peter

[quote author="clintrubber"] [/quote]
[For the Dutch-impaired: the title reads Fokke & Sukke know what it's all about in Science; the text is Very impressive, my dear colleague ... but does it also work in theory?]

Good point. I briefly touched upon it a few posts back, but given smallbutfine's observations it may be worthwhile expanding on it a bit more:

The setup which sounds best (subjectively) does not necessarily have the lowest jitter (objectively), or vice versa.

I know that some people prefer the sound of a given converter which is clocked in a way that introduces more jitter than the minimum achievable. That's fine, but it's also entirely subjective. There's no two ways about it: jitter introduces irreversible distortion. Still, so do many mic pres (like the Hamptone, to name but one). If anyone prefers the sound of their jittery converter that's fine by me, there's no arguing taste. Having said that, I do feel that the function of a converter is to translate between the analog and the digital domain with as little loss of distortion as possible. As others have said here before, if you want to color the sound, do it deliberately with an effects unit (and sample clock jitter isn't too hard to simulate in the digital domain).

[quote author="clintrubber"]

[my newly acquired Multiface may just give me enough of an incentive to get that ADA8000 working]

Click to expand...

Nice. Multiple ADAT-I/O by chance ?[/quote]
No, that's the Digiface. The Multiface is a laptop-connected 8in/8out-module with a single ADAT I/O. I do have a multi-ADAT PCI card.

JDB.
[lusting for an AES-32 or some other good multi-AES card -- no such thing as too many digital I/Os]

Nicely said, JDB.
You are completely right about jitter reduction as an ultimate goal to prevent distortion (therefore I talked about clocking more than one converter as beeing always a compromise because there will always be quite some jitter introduced in all converters but the master!)

So if not measuring, how do I judge the configurations by listening?
I use test cds, for example. These contain excellent natural instruments, drums, violin, voice and maybe acoustic jazz in natural rooms recorded to come very very close to real life with uncolored mics, amps and excellent converters.
Know how these instruments sound in real life!
Comparing by listening to the transient behaviour (e.g. acoustic drums).
Listening closely to the depth stages of the recording! Distortion nearly always flattens these!
Listening to synthetic sub basses near sinus - one can clearly hear distortion and frequency range reduction.
Listening to natural voices and complex electric guitar arrangements, music that is recorded/mixed state-of-the-art to check the mids for any kind of mud....
I currently use a chevsky test disc, madonna's ray of light, Talvin Singh and Foo Fighters - Echoes, Silence, Patience & Grace to get a good picture of a setup's sound right at the moment.
If it's fast, open, deep and low, it is good.
No way to be sure of the best setup if you do not measure or have very good monitoring (yeah, our monitoring is the most important part of our setup and we always try to upgrade and spend more money into this than into converters actually!!!)

I hope you did not get the impression that I am the enemy of digital theory, I come from electrotechnics and physics and am an IT guy for a living.... BUT the deeper you get into the theory the harder it is to proof it in your actual setup because there are so many unknown factors that can degrade clocking and sound. I really would like to have an easy/not too expensive way to measure things like jitter objectively....and still, there are also other effects that influence sound in the digital domain, like different hardware latencies in the converters and such.....
'Digital' sounds so predictable and easy - but if one digs deeper, it is not! It is a much underrated complex topic.


Kind regards,
Martin