PA pyrotechnics on power dropout-help please

[Mike C]

"If the connection from the board to the amps is properly balanced you'll never get a ground loop. I've worked with hundreds of systems in major concert venues with the FOH a hundred feet or more from the stage, using both FOH and separate backstage monitor consoles, and never had a problem. However, not all devices, especially signal processors, are properly balanced. Another reason for having all signal processors next to the board is that allows everything to use the same breakout box, and if there are improperly balanced devices or even some that are unbalanced the interconnects are short enough to keep ground loops from arising."

Generally in large production systems there is a common power distro that all but takes care of power grounding issues between equipment.
In Marvin's church that's probably not the case.

[marvin]

Mike & Bill,

My memory failed me again There was a time when we lifted the ground at the mixing booth, but that must have been with a prior system. I looked today at the booth and amp rack and found everything to be in order respecting power cord ground pins. Now we all feel better.

Back to the original power failure plosive problem.
How would you recommend proving a UPS could solve the problem? When we lose all power we get that horrible BANG through the main speakers. I suppose I could simply power everything on and then switch off the culprit amp or unplug it while maintaining power to all upstream devices. However, I hate to even do that for testing purposes because I am afraid it might damage a driver. What do you think? Is it safe to test this way?

[marvin]

BTW, here is a line diagram of our PA system.

Church PA line dwg08112022.pdf

The SEQ1, 2, or 3 denote which step of the Furman power sequencer the device is plugged into.

[Mike C]

The switch box is interesting, is that used to turn on and off the speakers in the different sections of the church? Is it maintaining a balanced line connection through the switching?

It maybe be hard to tell when it happens but do all the speaker zones, stage monitors, nursery, overflow and the main speakers all make the the loud pop or is it just
certain speaker zones?

If possible try to set up a test to see if you can narrow down if there is a single piece of equipment causing the pop.
Maybe turn the levels on the amps very low and pull the power that is feeding all the amps and the processing to see if you hear any noise at that point.

To check the input side of things turn the amps off, plug a pair of headphone into the mixer and pull the power feeding the mixer, wireless equipment and anything
connected to mixer inputs and see what you hear in the headphones.

If it's something connected to a mixer input from the stage that is the biggest noise pop issue a UPS on the mixer and processing would not really solve the problem.

[marvin]

The switch box is interesting, is that used to turn on and off the speakers in the different sections of the church? Is it maintaining a balanced line connection through the switching?

The switch only opens the tip (#2 pin) circuit. The 1 and 3 lead continuity is maintained. There may be a small capacitor across pins 2 and three but I would have to confirm that. Is this a valid way to switch outputs? It works perfectly well.


It maybe be hard to tell when it happens but do all the speaker zones, stage monitors, nursery, overflow and the main speakers all make the the loud pop or is it just
certain speaker zones?

Unknown. When it happens I, and most people, are in the sanctuary and it is so loud there is no way to hear anything else. The ancillary rooms are mostly unoccupied at that point of the service. I do know that during normal shut down sequence, I can hear a small pop from another room which indicates one of the amps is not shutting off before the upstream devices. However that pop doesn't sound destructive.



If possible try to set up a test to see if you can narrow down if there is a single piece of equipment causing the pop.
Maybe turn the levels on the amps very low and pull the power that is feeding all the amps and the processing to see if you hear any noise at that point.

That sounds like a good test. I will try it and let you know.



To check the input side of things turn the amps off, plug a pair of headphone into the mixer and pull the power feeding the mixer, wireless equipment and anything
connected to mixer inputs and see what you hear in the headphones.

Connect headphones to the mixer headphone jack? Or the main output of the mixer?



If it's something connected to a mixer input from the stage that is the biggest noise pop issue a UPS on the mixer and processing would not really solve the problem.

Thanks, Mike

[Mike C]

"
The switch only opens the tip (#2 pin) circuit. The 1 and 3 lead continuity is maintained. There may be a small capacitor across pins 2 and three but I would have to confirm that. Is this a valid way to switch outputs? It works perfectly well"

That is not correct to switch a balanced line audio circuit on and off.
Pin 1 is ground, pin 2 is positive and pin 3 in negative. Pins 2 and 3 both carry audio they are just opposite polarity/phase, disconnecting just pin 2 would still
pass audio on pin 3 to the connected equipment the resulting level would be lower but still there.
Maybe the switch box is not actually wired for balanced line switching.

Could you post a picture of the box showing the internal switches and connections?

[marvin]

I will open it and inspect it. Will let you know.

[marvin]

The switches have no capacitors in the circuits (An old drawing showed caps, but there are none).

Interestingly, the SPDT switch closes the pin 2 circuit in the on position, and shunts the downstream 2 & 3 leads going to the amp when switched off. Is this a good way to switch balanced lines? Should we use a DPDT to interrupt both circuits 2 & 3?

See this wiring sketch:
Church PA switch wrg08132022.pdf

[Mike C]

Both in the drawing and looking at the switch I can't see how shutting the south wing is not also shorting / shutting off the mains.
That switch is a hack job and not helping things,

Yes double pole double throw switches would be the correct switches to use. The input feed + & - would go to one pole set on each
switch, the common / middle switch connections would go the zone output jacks and the other pole would be shorted together, shorting an input pins 2 and 3 together is a way to mute the input on a piece of equipment,

Actually I would use the mixers group outputs to feed the zone and do away with the switch.....unless you are using the groups for your mixing operation.

Looking closer now at the one picture I can see you channel group assignments and it looks like your only using the main LR .

[marvin]

Both in the drawing and looking at the switch I can't see how shutting the south wing is not also shorting / shutting off the mains.
That switch is a hack job and not helping things,

I thought the same thing at 1st look, but they are independent. Each switch shunts it's common terminal (signal lead 2) to the switch off position terminal (signal lead 3) only on the load (amp) side of the switch. That state leaves the line side (mixer output pin 2 common feed to both switches) isolated from the shunted side of the off switch, with an intact path to feed the other switch. It works!



Yes double pole double throw switches would be the correct switches to use. The input feed + & - would go to one pole set on each
switch, the common / middle switch connections would go the zone output jacks and the other pole would be shorted together, shorting an input pins 2 and 3 together is a way to mute the input on a piece of equipment,

I will try to rebuild that switch box.. Is it necessary to short the amp input when unused, or good practice? Couldn't we just open pins 2&3 with DPDT switch without shunting 2 to 3, or does that risk a pop when the switch is thrown?

Does the DPDT need to be break before make or make before break?




Actually I would use the mixers group outputs to feed the zone and do away with the switch.....unless you are using the groups for your mixing operation.

Looking closer now at the one picture I can see you channel group assignments and it looks like your only using the main LR .

We coulD do that, however, all operators are amateur volunteers. We are maxed out on post fade buses in the mixer so the two switches on the main output are needed to turn off the South wing zone when needed.

The four post fade sub outs are used for recording, streaming, and two separate monitors. We could eliminate the other switches, but they are convenient.


Thanks a bunch,, Mike, for your time and expertise!

[Mike C]

"I will try to rebuild that switch box.. Is it necessary to short the amp input when unused, or good practice? Couldn't we just open pins 2&3 with DPDT switch without shunting 2 to 3, or does that risk a pop when the switch is thrown?

Does the DPDT need to be break before make or make before break?


Shorting the input cable will take care of any stray noise the unterminated cable run could pick up acting as it is antenna.

A standard DPDT switch will be fine.

[PWR RYD]

Hi Marvin.

Did you ever get this fixed/figured out?

Craig

[marvin]

Hi Craig,
Not yet...getting sidetracked by other things, including assembling a little speaker before the Meniscus GTG.

I ordered and have the DPDT switches from PE and intend to replace the SPDT switches per Mike's suggestions. Later, I will try to trace down the big bang problem. It isn't at the top of my list right now, but I do plan to take advantage of all the input kindly offered here.

I will update when there is some progress.

Marv

[marvin]

Hi everyone. I hope all you past responders to this thread are notified by the forum because I still need advice.

Life keeps me busy and since the “BANG! on power failure” issue is relatively rare, it has been on the back burner for a while. However, ignoring it didn’t make it go away, and a recurrence a week ago forced it to the front burner again.

Before the service began, we experienced a few second power dropout with all lighting going out at the same time. We were all scared by that very loud BANG from the PA speakers that accompanies all our power failures. The lights and general power were restored within a few seconds, but it happened again within 15-20 seconds. This time the lights just seemed to flash or brighten/flicker, but the BANG was as strong as before. Everything was fine for the remainder of the evening.


Before I lose you with the full account that follows, I will list my bottom line question of the day here: What happens when an amp power supply gets a power spike from the utility? Could that cause the loud BANG we experience? I hope you engineers can help answer that.


During the following week, I performed the following tests per your previous recommendations:

• I had rewired the homemade switch box per Mike C’s recommendation some months ago. That has worked great, but made no change to our problem. Its great to have it wired properly, though—thanks, Mike. No known ground loop hum at this time.
• Only one device in the mixing desk (TASCAM digital audio recorder) had a ground lift plug in place, which I removed with no ill effects. Now all PA equipment at the desk and amp rack are properly connected to power.
• Per Mike’s troubleshooting advice, I performed the following:
  • Disconnected &/or switched off (both by switch and plug pulling) every device in the signal chain to the main amps (mixer, DFR11EQ/limiter, S200EQ, Peavey IDL1000 digital delay, both with reduced and full main amp input trim.
  • Even switched off the mixer and/or pulled the main plug to the entire mixer desk.
  • None of the above produced any more than a gentle “pop” or short hummm.
  • Pulling the power plug to a Shure BLX4R wireless receiver produced a decent SNAP only at normal use channel level setting, and less snap at lower slider levels, and no snap at all with slider fully down. Definitely nothing like the big BANG from power failure. All input channels down was the state of the mixer during our last power fail event so I know the mixer inputs are not the issue.
• I also shut off the building 400 amp MAIN 3phase service switch with the PA system in the same state as the last event to simulate a utility power failure. Again, just a minor thump from the PA system.

My extensive testing could not replicate the big BANG we experience with power failures. Nearly everything produced only a mild “thump”.

My conclusion is that the amps are seeing a voltage spike from the utility (ComED) at the moment of these power dropouts. I filed a case with ComEd last Saturday and they promised to check their equipment, but I have not received a report yet.

Our mixing desk and amp rack are both protected by whatever the Furman PS-8 power sequencers offer. I don’t know if that is enough suppression to prevent a spike from affecting the amps.

What happens when an amp power supply gets a power spike from the utility? Could that cause the loud BANG we experience?
Somehow the spike produces a current spike that slams the driver voice coil into the back plate? Isn’t it just as likely to launch the coil and cone out of the basket?
I have the impression the amps may produce that bang due to power surge regardless of whether an input is connected to it or not. Please correct me if that is wrong.

If you made it this far, thanks for your interest and diligence. I appreciate any comments you can offer.
Marv

[devnull]

I wonder if it could be something inside the church that's causing your power issues. I'm thinking a failing motor or motor controller on an air handler or something else tied to all 3 phases. A large inductive load can look like a power outage and won't necessarily pop a breaker, depends on the time curve, and the power can come back after the motor is up to speed.

Have you had other outages when you weren't having services? Talked to the neighboring businesses if they've had power issues? When there's a power event is it across all 3 phases? Does it occur same day of the week and approximately the same time?

[marvin]

I wonder if it could be something inside the church that's causing your power issues. I'm thinking a failing motor or motor controller on an air handler or something else tied to all 3 phases. A large inductive load can look like a power outage and won't necessarily pop a breaker, depends on the time curve, and the power can come back after the motor is up to speed.

Have you had other outages when you weren't having services? Talked to the neighboring businesses if they've had power issues? When there's a power event is it across all 3 phases? Does it occur same day of the week and approximately the same time?

Thanks for your comments, dev...
After all my testing and the history of this problem, I really don't suspect equipment inside the building. It only happens when we have a power outage, and it can be months between events.

Our most recent event seemed to coincide with a dropout a few blocks away, yet another parishioner two houses away witnessed no clocks needing to be reset, but it was a very short dropout.