I am having some issues with what seems to be a problem with a DSP processor. I have installed a DSP into a client’s board room and every morning it has to be reset. By this I mean that after the unit has been rebooted all is well.

Every morning for the last 3 weeks after install of the unit we come into the meeting room and the audio levels are spiked. What I mean by this is we have a very load noise floor and loud crackling from the 70v speakers. I check the levels on the signal meters and check all of the DSP settings and no settings have been changed.

Though now I reset the DSP processor and all audio levels go back to normal with the same DSP settings stated above. I have checked every phoenix connector, xlr input jack and every RCA connections within the room. All the cable is shielded microphone except for the 14/2 from the 70v amp feeding the 6 speakers in the room.

The whole rack is on a dedicated circuit (or what we specified that we needed was a dedicated circuit if the electrician provided is unknown at this time) with a Surgex SX2120. We even have a UPS on the system from the outlet to the Surgex.

This anomaly seems to happen every night. This is a very basic system with a third party controller for raising the line level signals to the amp only.

Does anyone have a idea of what is happening here. The engineers that made the DSP are at a loose and at this point so am I. The only thing I can rationalize at this point is that there is a bad processor within the DSP unit that over a period of hours just shuts it’s self off.

I can't tell you how many times I've run into this only to find out that a factory assembly line process happens every day at 9:00pm or something like that, and it coincides exactly with a failure. Further inspection shows a ground loop or an induced spike in a signal line. These things are 90% observation and 10% deduction. Don't guess, look.

Thanks for taking the time to send me your recommendations. It is very much appreciated. In working on this system I have checked and verified the power with an analog scope. This is what has been looked at so far.

I measured the ac line with an oscilloscope, both the “conditioned” and the “unconditioned” circuits. Although the “conditioned” line had a slight intermittent noise pattern, it was far below a level that would interfere with the installed equipment (less than 5v of transient at approximately 4KHz). Oddly, the “unconditioned” line was cleaner than the “conditioned”; no noise at all.

It looks like I am going to have to rent a digital storage oscilloscope and let it monitor the system for a period of 48 to 72 hours and check the readings daily. I think that with will be the next step in your recommendation of “These things are 90% observation and 10% deduction. Don't guess, look.”

Once again Wayne, thanks for taking the time out of your day to help with a problem. It is nice to know that a few other people are willing to help in solving audio problems such as these.

Sincerely

ELK

• http://www.biamp.com/index.php?pg=audia/AudiaFLEX