ADA 8000 crackling

[a.d.a.m. baby]

It's been a great couple of months - everything from synths, drum machines, even laptops, going pffft!
Now it's the turn of my be!@#$%^& ADA 8000. After being on for a while, there is crackling on the audio outputs. The settings have not been changed (it gets clock via ADAT optical from a MOTU 828).
Before I get my hands dirty, any ideas or experience to share?

cheers
Adam

[Alistair]

Try replacing the ADAT cable before you do anything else. I have had them go bad on me before and had crackling.

[Thirteen]

They have power supply issues with underrated components, that may be your problem, I think that there was a discussion with fixes over at The Lab.

[a.d.a.m. baby]

Thanks guys. The problem occurs only after the unit heats up. As Steve says, they have PSU issues and get really hot. Replacing the regs and big caps was no help. I've bought another one and will be modding the PSU or replacing the toroid to make this one behave better

cheers
Adam

[rob]

so the big question is "why does it crackle?" Replacing the power supply could end up being a lot of work that doesn't fix your problem.

[Chris H]

This is a long shot but maybe the cracing is caused by a crack in the PCB that becomes wider as the unit heats up and the metal case expands. A possible test ..... Tap around the unit once it has warmed up .

[The Tasmanian]

It could be because its Christmas ......pork ?

[Kurt]

From memory the "PSU issue" was only ever an issue in 110V countries.. I've had two for years with no problems at all.

Does it crackle on all channels or only one/some? Does it crackle when passing audio or when it's doing nothing?

I'm with Rob
"It crackles, I read somewhere that the power supply may have certain issues in some situations, it must be the power supply that causes the crackling. Therefore I will spend money fixing a problem that does not exist on a new unit." does not seem like very thorough troubleshooting.

[a.d.a.m. baby]

Well well... this IS embarrassing.. you guys have correctly identified my lame-ass fault-finding as...faulty!
It wasn't the bloody be!@#$%^& after all!
In fact, I'm not really sure what the problem was, but it recurred with the new unit. Somehow, it would seem that the clock over the ADAT connection is not behaving as it used to.
Now, my understanding of these scenarios is, when you are sending audio via ADAT optical to a unit, then that unit should get its clock via ADAT also. That is how my MOTU 828 and the B-unit have been connected for years and worked well. Recently, I had to replace the logic board in the iBook that controls all this stuff - not sure if that is relevant, but after that I noticed this distortion on the audio out s of the B-unit. The settings of the main MOTU unit have not changed (clock is internal).
Anyway, by flicking the clock switch on the B-unit over to internal, the crackling in the ADAT audio stopped, and it continues to work fine.

Why would just the clock and not the audio in the ADAT connection be faulty?

[Kurt]

Clocking over lightpipe, in my experience, is only solid if there are only two units, one master and one slave. This is one area where the ADA8000 sucks, there is no word "through". I have tried all sort of scenarios with my soundcard and two ADAs to try and get solid clocking, none have been wildly successful. I am currently running both units as Master and set my soundcard to accept clock signal from one of them. This works fine 99% of the time, though I have had a couple of sessions where I have had to repatch everything because of clock errors (like awful one sample glitches and ticks in the recorded tracks from one unit)

The solution would be an external clock generator I guess but it hasn't caused me enough problems to follow it through. Yet.

[Alistair]

With more than two units, connected in anyway, you need an external clock in my opinion. I'm sure someone else will have pulled it off, but I've never found it reliable enough to be confident.
Also, as I mentioned before, try a different ADAT cable just in case. They are quite fragile and susceptible to damage.

[a.d.a.m. baby]

Thanks guys. There's only ever been two units in my set-up, the MOTU master and the ADA slaved to optical clock from it. No changes then suddenly this. Alistair would a faulty cable pass audio but not clock? I mean, aren't these things just go or no-go? Or do they deteriorate, with maybe the high frequency stuff being affected first, which could explain the scenario...

[Alistair]

I'm not sure exactly- like analog cable, there are endless arguments on the net about what is what. I do agree that it's all binary code, so it either gets there or not, but I do know my thin flimsy ADAT cables (cost around $2) crackled fairly often, and my thicker and more rigid ADAT cables have been trouble free (and were still only $10 each). Nothing else was changed in my setup, but better quality ADAT cables fixed all crackles straight away, and the noise has never returned.
With clocking we are talking about fractions of a second, so I guess minute changes in anything could possibly make a difference. There are some guys who believe if you are using a master clock, all wordclock cables should be the same length so the signal gets there at exactly the same time- others have systems that run perfectly fine without this. Who knows??

[Alistair]

Once I moved to 3 devices I grabbed a BLA clock as well. What PS are you using Angus? I got it second hand off this forum and it's got a US plug switchmode with an adapter- I've been meaning to replace cause adapter plugs annoy me.
Great little clock, but it shits me that the switch for 41/44/88/96 is on the same side as the cables! Bloody annoying to have to reach behind to change the rate.

[no-fi]

check the MOTU wasn't somehow set to receive audio clock via ADAT... (if that's even possible? never used the MOTU)

I assume you have the 2 units setup with 2 adat cables, and if one is being synced from the other while also syncing the other, that could get funny.

Otherwise, does the MOTU have a BNC wordclock connector and do wordclock out? if so - get a short run of coax with BNC connectors and use that to sync the AD8000.. that's likely to be better than trying to recover clock off the audio bitstream.

[a.d.a.m. baby]

Just an update - this problem continued to intermittently frustrate me, but appears to have been fixed by the use of a dedicated wordclock BNC cable between the units as some suggested.
cheers
Adam