8x8 turn-off thump
 

I was playing around with my 8x8 after the Redskins game today, and accidentally unplugged it. MASSIVE turn off noise. I had my amp set to turn on via the miniDSP's rem-out, but the remote jumper was untouched and the rem-in was not used.

The amp it was connected to (ElectroVoice CPS-8.5) has a rated voltage swing of 63V into 4Ω...and now I've officially blown my first driver that wasn't a Jordan Module or JX53. Fortunately, it was a cheap car audio speaker, and not the big, expensive Aurasound woofers or the irreplaceable JBL SUB1500s. But it has a phase plug, and the resulting surge literally changed the rest position of the cone back about 2-3mm.

Does the 8x8 still thump like that with remote-in/remote-out? I don't currently have a driver I don't care about to test, but if so then I and anyone else using these things really needs to consider keeping it on and connected to a UPS.

I am currently trying to make a relay setup for my 8x8 that mutes instantly on power down, and has a delayed mute on power up. I think this approach is the only real solution in case of "brown out".

I haven't used rem in/out but i don't see how it can help if the power is removed.

One way that I have thought of to address both power down and power up is the following concept:
Use TWO relays:

• Relay 1 is connected between the amplifier output and the speaker. It should be a 10+ amp 240VAC contact, 120VAC coil. The coil is powered via Relay 2:
• Relay 2 is a 120V contact (current rating is low), 12VDC coil.

The mains AC "live" is connected first to the Relay 2 contact, and then to coil of Relay 1. This provides two ways to turn off/on the speaker: relay 1 or the mains AC.

Relay 1 is controlled by a timer (such as a type 555/556) to provide turn-on delay of a few seconds. This provides delay-on protection for the speaker.

As soon as AC mains power is disconnected (within 7-8 msec), relay 2 will open and the speaker will be disconnected. This should be fast enough to protect the speaker from "pop" transients coming from line-level components like the MiniDSP shutting down while the amplifier is still powered up.

If someone will produce a PCB or populated PCB following this concept, I am willing to buy!!! Otherwise, this is not too hard to make using point-to-point wiring.

-Charlie

That is precisely what i am building, except i am using some 12 VDC relays for "Relay 2" and activate them with a rectified but unsmoothed DC supply for fast decay at power down.

Didn't realise i could go for relays with mains voltage coils - that is interesting and would make it a bit simpler.

For the 555 i am using a circuit from here

under "Power ON Delay Circuits" circuits number 2 or 3

But I'm not sure why the first relay is needed, unless one doesn't have a 12V trigger on one's preamp. The miniDSP's existing rem-out already delays things by a little bit. I've not experienced turn-ON thump when using the rem-out with my amps (Sherwood A-965 for mains, ElectroVoice CPS-8.5 for multisubs).

The second relay, in light of my recent experience, seems very useful.

DIY electronics design is beyond my abilities. Is something like relay 2 alone available in a plug-and-play product?

I think in Charlie's explanation of his proposed design Relays 1 and 2 may have been mixed up at some point. Anyway the circuit in theory should solve all the issues with turn-on and turn-off thump in all scenarios, where REM in/out alone might not.

You have one relay which is activated with a delay timer and switches the coil of a second relay(s) which switches the audio signal(s) to speaker/line level.

I guess you could use REM out to activate this primary relay with a delay - around 2 seconds according to minidsp datasheet. REM out provides just 10 mA at 12V so it might need to drive a transistor, which then drives the primary relay.

All is fine until power is removed - either to the miniDSP alone or everything due to a power cut. My question is how quickly does rem out decay? The miniDSP will turn off (producing a thump), REM out with drop to 0V immediately triggering the amplifier to turn off. But whether the amps are shut down quickly enough to suppress the thump through the speakers has to be tested.

In charlie's design the use of mains to power to the secondary relays means they will deactivate immediately in all cases where the minidsp loses power. No matter what happens with the amps, the thump isn't getting to them. That makes me feel a lot happier!

I am building this circuit with 24 VAC secondary relays which will decay as fast as the mains but without having mains everywhere in the box.

Because your amps have trigger inputs i would give the REM in/out feature a thorough testing with a full range speaker that can handle the thumps - that alone could be enough to solve the problem.

That's the plan right now.

I think I have a little subwoofer (Gallo Nucleus Micro, small closed box with Peerless DVC 8" woofer) that I don't really care if it blows up. Just need to go to my storage cellar and get it.

My hope is that the end-effect of this thread is simply a caution not to use the remote-out trigger without using the remote-in trigger as well. Another alternative is to consider a small UPS on the miniDSP not optional.

Hi,

The turn off bump is a significant problem with MiniDSP.
I have tried two methods , and both of these seem to work well:
1) Use speaker protector IC to short the outputs of MiniDSP when power is removed. This is perhaps the best solution overall although another board is need. I have tried it with uPC 1237 and it worked well. But its a bit mess when many outputs must be shorted and many relays have to be used.
2) Use large enough supply capacitors so MiniDSP is shut down after power amplifier does. Only useful when miniDSP is used in the same chasis as power amp. In my case it was 10,000 uF

Hope this helps.

Regards,
Lukas.

I just posted a question about this power down transient today at the MiniDSP forum. Maybe it's important to note that the transient is only severe when the AC power cord is unplugged from the power supply, not when unplugging the DC cable. It's not just a matter of the power disappearing from MiniDSP's perspective. Is it possible a different power supply would solve the problem?

MiniDSP say the above over on their own forum.

I have the 8x8 board in a box from MiniDSP ... Does anyone know if the bove solution would still be possible for me? (It sounds like the best approach).