ADA8000 sync problems - new findings

I found another reason for ADA8000 sync errors: they can be caused by bad ADAT encoder ICs (Wavefront AL1401AG) in the device.

There are some production lots of this IC are sensitive to vibration (!). The error can be provoked by knocking on the ADA's main board in the area of the AL1401AG and causes a badly synced ADAT signal on the output. I know at least two people which can confirm this; the affected AL1401AG datecodes are 1038 (and maybe later), earlier production lots (e.g. 0846) work flawlessly.

Anyone here who can confirm this? I have working replacement ICs here, but if this turns out to be a common problem, we should first put some pressure on Behringer...

Do you know it's internal to the chip and not just a bad production run of boards or wave soldering?

Well I'm about to return three to Thomann but won't be able to open them up unfortunately.

All three have the same behaviour in different extremes. What's interesting is they'll all clock to each other but not through the appsys extender pass through.

What's also interesting is when two faulty units are connected together, the position of the sync switch makes little difference, providing at least one is master. Set both as masters, one at 44.1 and the other at 48 really shouldn't work yet seems fine.

The other thing that's interesting about these units, and may be useful info, is they won't clock to anything else, but they will clock to each other.

As I say, I have three, each more useless than the last. If I connect the two most faulty together they'll clock perfectly. Even if both are masters. Even if both are masters at different clock rates. So an input to unit one that's a master at 44.1, connected through to output of unit two that's a master at 48. That really shouldn't work, but it does.

Quote:

---

Originally Posted by JLepore

Do you know it's internal to the chip and not just a bad production run of boards or wave soldering?

---

It's definitively internal to the chip, as it appears also when the ICs are used in custom circuits outside the ADA8000.

From, possibly, Mr B

Dear Matt,

thank you for your comment and sorry for the trouble you are going through.
Since this issue was mentioned earlier, we have investigated the matter. This is also the reason why we have delayed the introduction of the ADA8200.

The Adat clocking is managed by the Alesis/Wavefront chips AL1401/1402 which are used in the ADA8000. There seems to be an incompatibility with certain products in particular with Motu devices.

We are currently in touch with the Wavefront distributor. Interestingly, we have found differences between the previous Alesis chips and the newer Wavefront versions, but at this stage we don't know if this is a Wavefront problem or an issue caused by Motu etc.

Unfortunately this problem is out of our hand as it is an Adat incompatibility. I will ask one of our engineers to explain the matter in detail and also keep you posted on any progress.

We are also aware of some issues related to the power supply of the ADA8000. Here the challenge is the high power consumption and the corresponding heat the unit creates. While we have upgraded several capacitorsfrom 80 to 105 degree versions, we recommend to provide for enough air flow when operating the ADA8000 in a rack. This way the unit will work perfectly fine.

Like the DCX2496, DEQ2496 etc. the new ADA8200 will now feature an auto-range switch-mode power supply which will resolve the heat issue.

Uli

Maybe these things are linked somwhow...and I believe that there has been really a bad production lot. At least I have several already "Wavefront" labelled chips from datecode 06+ which run perfectly ok, even with MOTU's, but later ones apparently do not. I think that at least the distributor is aware of the problem....they agreed to return the chips with the questionable behaviour.

So apparently now the answer is don't use ADA8000's or 8200's with MOTU setups. Yet another reason to dump the ADA's.

And nice play on the power supply issues.