timeout waiting for acknowledge from dsp 0

[asktoby]

Periodically, my soundcard stops working with the error message:

Code: Select all

System error
timeout waiting for acknowledge from dsp 0 on Pulser2 (brd 0)
(communication stalled at Cebulon (async))

The hardware will no longer respond to any changes, but you might be able to save your project before terminating the application!

[OK]

Cebulon.jpg (28.97 KiB) Viewed 9774 times

Do you have any idea what might be causing this and how to stop it?

The system is two PulsarII cards connected via S/TDM.
Scope v7.0.1705
Windows 10 64Bit

[Ripper]

I would think that it is a bad capacitor in a board.
I recently fixed a problem like this in 3 of my Pulsar II boards.
Already have the other pc opened to do the same to the other boards i have in that one (2 Scope 14DSP and 1 Pulsar II).

Start with the 100uf ones and go testing after the changes are made.
Yep, someone will have to solder

[Ripper]

Following this matter, as i mentioned, i'm changing my capacitors for new ones.
I did it on my Power Pulsar but now i found out that the other 14 dsp board, that is Scope Professional (I thought they were the same model, but they don't) has different capacitor printings.
I have these 47 16F printings on the caps while on the other board they were 47 6V. Hum...
Are 47uf 16V be the same as these 47 16F?

[jksuperstar]

Yep still Aluminum 16v 47uF caps.

You must run in a very hot and/or dry environment, to have so many caps dry out on you at the same time.

[Ripper]

Yep still Aluminum 16v 47uF caps.

You must run in a very hot and/or dry environment, to have so many caps dry out on you at the same time.

I don't think the environment is a problem. I think it is a matter of board's age.
I'm doing it more as a precaution, as the real problems i notice are on the DSP0 (and some DSPx, these will be next) of some of the boards, which are already solved by recaping the 100uf caps and the 47uf 6v ones closer to the IO boards and reflowing/adding solder to the rest as a test measure (it worked till now where i did it)
To prevent more issues i am changing all the other caps. It's not that hard to do, just a bit tedious on the big boards and, in this case, have to wait for the 47uf 16v caps that i didn't have in stock.

Oh, and thank you for the confirmation about the 16f / 16v doubt.

[astroman]

you need very (!) good soldering skills to mess with these multi-layer boards.
I don't question Ripper has - just a warning to the unexperienced or those equipped with non-adequate gear... see below

But in fact a worn out cap can trigger bizarre messages: my Pulsar One someday was commented to have a Ram error
It was a cap gone loose. Holy crap... btw a cap I had formerly replaced (for obscure suspect)

[asktoby]

Despite being a competent solderer I was nervous of taking the caps off my precious Pulsars!
So I've been getting by the last four months by keeping the side off my PC and giving the soundcard a nudge in the PCI slot every time it failed. It was very annoying though because every time the dialog popped up I'd have to restart the Scope software. It was popping up once or twice per evening.

On about three or four occasions I took both Pulsar IIs out of my PC, removed the ADAT daughterboard from one, and used electronic cleaning solvent on all the internal connections. It never really helped for more than a few days.

Last week I had the idea to move the ADAT daughterboard from one card to the other so tried that and am happy to report that I haven't had the issue since then - it seems to be fixed without needing to replace the caps.

[sepuka]

Got this error on XITE-1.

xite.png (9.31 KiB) Viewed 5898 times

[astroman]

I‘ve had this error numerous times on a Pulsar, and it had nothing to do with the SRam.
If the error persists, check the cap located between wordclock and midi, it may have come off.
(it might be faulty, but that‘s not very likely, considering it‘s relatively short time of usage)

Of course that‘s just a hint for a quick check. If you feel uncomfortable with opening the unit, contact support.
In my case replacing the caps (4 in a row on a Pulsar) solved the problem.
(the io system with the 2 SRams seems identical for many Sharc designs)