Thoughts about amplifiers at 3/4 volume.

[Ryan Lantzy]

Re: Thoughts about amplifiers at 3/4 volume.

Sending a hot signal and padding down the input at the amp, will reduce noise picked up in that interface between amp and the sending unit. This is less of an issue with good modern balanced inputs, balanced sends. and balanced (3 circuit XLR) wiring. In the not so good old days, using unbalanced interfaces and wiring, it was a very valid technique to manage noise in quiet ambient venues (like HOW). I can imagine a modern venue with distro or whatever issues and hum in the amps, that might benefit from hot level and pad, but with clean interfaces, not so much.

So arguably with modern gear you "should" be able to get by without amp trims, but there is a lot of inertia in the marketplace so customers will still want their amp trims for some time to come, and there could be the occasional problem gig where this technique could help.

JR

Makes sense to me. I had not thought much about amps picking up noise over long runs of wiring. All of my racks have the DSP co-located with the amps so this is never an issue. Sometimes it's hard to see potential problems when you're so used to doing things one way. IME, the mixers and outboard have dominated the noise floor of nearly every gig I've ever done (save some crazy buzz that develops in a musicians instrument).

Thanks, as always, for humoring me. It's nice to have perspective on these things and even though they aren't currently problems, they may aid in troubleshooting down the road.

 

[Ryan Lantzy]

Re: Thoughts about amplifiers at 3/4 volume.

Hello,

Errr....no. You may need to "turn down" the input stage to prevent over-loading the input which would cause distortion of the signal.

Hammer

Hammer,

As Bob Lee has itterated on these numerous times, most amplifiers have a fixed input stage, then a variable gain stage, then the amplifier stage. If you are overloading the input of an amp, the gain knobs will not do much to help you now-a-days. This may not be true of every amp on the market, but it's true of many.

 

[John Roberts]

Re: Thoughts about amplifiers at 3/4 volume.

Hammer,

As Bob Lee has itterated on these numerous times, most amplifiers have a fixed input stage, then a variable gain stage, then the amplifier stage. If you are overloading the input of an amp, the gain knobs will not do much to help you now-a-days. This may not be true of every amp on the market, but it's true of many.

and as JR has iterated,,, modern amplifier's active input stages can handle modern line level gear's output levels, so that is a non-issue.

JR

 

[Kim Watson]

Re: Thoughts about amplifiers at 3/4 volume.

I understand why they turn back the amps and the guys have covered it nicely, but to play devils advocate on running the system with amps turned right up.

We have a guy here who was working a show..... ran his amps turned back and the system gain structure set up for that (more level coming out of the board.) he turned the amps off..... and went for dinner. came back to find that the band wanted to practice and had take it upon themselves to turn all his amps on and UP FULL. he lost allot of speakers that day. when the amps were turned up full the amps were clipping. Destroying his rig.

Needless to say he now runs with the amps on full and his gain pulled down in the processor.

Have fun

Kim x

 

[Greg Cameron]

Re: Thoughts about amplifiers at 3/4 volume.

he turned the amps off..... and went for dinner. came back to find that the band wanted to practice and had take it upon themselves to turn all his amps on and UP FULL. he lost allot of speakers that day. when the amps were turned up full the amps were clipping. Destroying his rig.

I'm normally a calm person even when shit breaks or has problems. But I could see myself really loosing my shit over something like that. Of course, my system is setup with the amps all the way up and good protection in the processor so chances of catastrophic destruction are remote. But if someone of their own volition decided to go into my processor, turn off all the limiters, cranked up the band-pass gains, and blew all my drivers, I don't think I would be very calm. And they'd better be ready to write a check for drivers/diaphragms as well as a check to the venue to pay for the staff of the cancelled show. My rig isn't huge, but it's got a lot of drivers for it's size, 32 to be exact. And those 1" TAD diaphragms alone are $600 a pop, literally.

Greg

 

[Jeff Hague]

Re: Thoughts about amplifiers at 3/4 volume.

I agree that most of the noise in a system originates upstream of the amps so S/N really is no reason to turn the trims down anymore.
I am curious about another possible reason for doing so and that is to keep the crossover outputs close in level to each other to avoid bands overlapping.
I know the graphics from Driveware may be a bit oversimplified but I think these illustrate my point.
In the first attached graphic, all 3 outputs are set to 0dB and there is minimal overlap between bands. (It actually seems like in reality there should be a bit of a gap between the bands so they sum at 0 but again, the graphics probably arent very accurate.)
In the second attached graphic, the mid output is at +6dB relative to the low and high outputs and it shows significant overlap between bands.
It seems to me that leaving the crossover outputs near equal to each other and using the trims on the amps to control the overall output of the individual bands reduces that overlap.
Ofcourse in a perfect world the output power amps would be a perfect match for the efficiency and # of drivers in each band but how often does that happen?
Does this sound right or am I missing something (again)?

PS - Isnt it possible to somehow insert these graphics inline with the thread?

 

[Silas Pradetto]

Re: Thoughts about amplifiers at 3/4 volume.

...
In the second attached graphic, the mid output is at +6dB relative to the low and high outputs and it shows significant overlap between bands.
It seems to me that leaving the crossover outputs near equal to each other and using the trims on the amps to control the overall output of the individual bands reduces that overlap....?

Hey Jeff, unfortunately, the real-world acoustic response of the speakers being powered with the Driverack will look nothing like DBX's pretty picture. In fact, the picture doesn't even mirror the electrical response, and should not even be shown because it is misleading.

Using amp gains or the bandpass gains on the DSP will yield the same net electrical response.

You are right though, that relative bandpass gain will definitely change the acoustical crossover! This is why aux subs are a bad idea: they are variable. If one decides to adjust the aux for a channel to taste, they are effectively changing the acoustic crossover for that source, possibly causing all kinds of phase problems, depending on how compatible the sub and main phase responses are in their overlapping region. Personally, I prefer group subs, but those can have issues also.

I've actually seen people use the picture of the crossover in the Driverack to set up their system; it is surprising how much faith we put in pictures if they are available. I think the same thing happens with digital console EQ...

Yes, you can put those pictures inline, you have to 'preview' your post, then click the picture, get the link of the big one, then go back and insert a picture using the link that vBulletin provides....it's a huge pain.

 

[Bennett Prescott]

Re: Thoughts about amplifiers at 3/4 volume.

Yes, you can put those pictures inline, you have to 'preview' your post, then click the picture, get the link of the big one, then go back and insert a picture using the link that vBulletin provides....it's a huge pain.

http://www.soundforums.net/live/threads/864-Picture-Posting-Problems

 

[Greg Cameron]

Re: Thoughts about amplifiers at 3/4 volume.

You are right though, that relative bandpass gain will definitely change the acoustical crossover! This is why aux subs are a bad idea: they are variable. If one decides to adjust the aux for a channel to taste, they are effectively changing the acoustic crossover for that source, possibly causing all kinds of phase problems, depending on how compatible the sub and main phase responses are in their overlapping region.

If you fly your tops or have them at any substantial distance from your subs, shifting the relative acoustic crossover by changing the sub send gain from unity on a channel or two doesn't mean much anyway as you've already got the issue of non-coupling low frequency drivers in the separate bandpasses. And you know the saying, "if it sound good, it is good." I try not to get too hung up on the physics if things are reasonably setup and it sound good. That said, I only occasionally make tweaks to the sub send above or below unity. When I do, it's because it 'sounds' like the right thing to do.

Greg

 

[Jeff Hague]

Re: Thoughts about amplifiers at 3/4 volume.

Hey Jeff, unfortunately, the real-world acoustic response of the speakers being powered with the Driverack will look nothing like DBX's pretty picture. In fact, the picture doesn't even mirror the electrical response, and should not even be shown because it is misleading.

I agree but as a quick visual aid I do think they provide some value.

Using amp gains or the bandpass gains on the DSP will yield the same net electrical response.

I wish I had thought of that... But, ultimately what matters is the accoustical output and hopefully the bands are pretty close in level by the time sound leaves the speakers.

I've actually seen people use the picture of the crossover in the Driverack to set up their system; it is surprising how much faith we put in pictures if they are available. I think the same thing happens with digital console EQ...

I guess the bottom line then is, the graphics being as poor as they are and the fact that the "summation" between the bands doesnt actually happen until there is actual accosutical output are simply more reasons to use tools like Smaart to align systems.

Thanks!

Jeff

 

[David Wilkins]

Re: Thoughts about amplifiers at 3/4 volume.

Gentlemen,

It seems our fellow forum member; Shane Presley has stumbled upon a question that has challenged the knowledge of our esteemed group. Though there have been 29 responses (at the time of this writing) I do not see the real answer to his question. Maybe those of you who really do know have stayed quiet, waiting to see what people really know and do. So here goes, flame suit on:

There are but three reasons that a competent installer/system engineer would raise the sensitivity of the power amplifier's input stage. In order of importance, they are:
1) Obtaining a system’s maximum dynamic range. If the mixer is capable of a +26dBu output and the amplifier will deliver max gain voltage with a signal of +4dBu when the attenuator is at its highest sensitivity, then that system has lost 22dB of dynamic range.

2) Maintaining the system’s lowest possible overall noise floor. Every component in the audio chain has a noise floor. If one component in that chain is operated close to its floor the system's overall noise is raised. You may have heard an improperly set gain structure that was so far from optimum that the entire system sounded "thin."

3) Because the system is capable of much more SPL than the gig or venue needs and the user never intends to operate at a point of over driving the amplifier's input--even if it is one of the few amps whose input can be overdriven by the mixer's highest signal.

I've only come across one amplifier whose input attenuator didn't operate like a true adjustable pad.

I’ll point you to what was given to me years ago that has been one of the most valuable tools to help me understand gain structure and to get the most out of a system.
Take this paper on gain structure to woodshed for a little study time.
http://shure.custhelp.com/ci/fattach/get/58/0/filename/gain_structure.pdf

It has also be updated and is on the ProSound Web here: http://www.prosoundweb.com/article/how_do_you_set_system_gain_structure/

 

[Brad Weber]

Re: Thoughts about amplifiers at 3/4 volume.

If you have a DSP, turn down the crossover outputs, if not... bring down the master fader by the same amount. You'll see that the same result is gained, plus like another poster mentioned, you don't have to worry about someone fiddling with the knobs "on accident."

Adjust things in the DSP gains.

Is adjusting the DSP output providing the same result? It might be a very similar result if you are adjusting is the analog signal level of the DSP after the output D/A but not necessarily the same result if what you are adjusting the digital signal level before the D/A.

 

[Tim Padrick]

Re: Thoughts about amplifiers at 3/4 volume.

3/4, 13/16 - whatever it takes.

 

[John Roberts]

Re: Thoughts about amplifiers at 3/4 volume.

Gentlemen,

It seems our fellow forum member; Shane Presley has stumbled upon a question that has challenged the knowledge of our esteemed group. Though there have been 29 responses (at the time of this writing) I do not see the real answer to his question. Maybe those of you who really do know have stayed quiet, waiting to see what people really know and do. So here goes, flame suit on:

Thanks for clearing that up... I was thinking it was answered too many times.

There are but three reasons that a competent installer/system engineer would raise the sensitivity of the power amplifier's input stage. In order of importance, they are:

This ASSumes the church in the OP's question was set up by a "competent" anything...

1) Obtaining a system’s maximum dynamic range. If the mixer is capable of a +26dBu output and the amplifier will deliver max gain voltage with a signal of +4dBu when the attenuator is at its highest sensitivity, then that system has lost 22dB of dynamic range.

While correct wrt peak voltage and peak power, as we should have observed by now typical system operators often drive amplifiers past clipping to increase the average voltage and average output power further. Hopefully not so much in churches, but it is no accident that amps have the capability to be used that way, and clip limiters have tens of dB of gain reduction range.

So yes you can maximize (peak) dynamic range by lining up clipping to be the same, but many amps have the capability to be operated beyond clipping with clip limiters. So this becomes a trade off between lowest possible noise floor, and pulling max power from the amps. In an ideal world people (and i guess "competent installer/system engineers") don't clip amps. In the real world they do.

In church applications as in the OP's question, this is IMO the exact tradeoff being considered, louder output or less residual noise. They went for less noise.

2) Maintaining the system’s lowest possible overall noise floor. Every component in the audio chain has a noise floor. If one component in that chain is operated close to its floor the system's overall noise is raised. You may have heard an improperly set gain structure that was so far from optimum that the entire system sounded "thin."

"Thin" sounds like it is describing something other than a simple gain trim. I can't guess what, so I won't.

While correct about noise floor contribution from later gain stages, this is confusing in the context of the power amp trim, since you are in fact operating closer to it's noise floor by trimming back the input gain while simultaneously improving the system S/N. While you may be adding some tiny fraction of a dB of the amplifiers noise, the several dB of reduction to the much louder system noise will swamp this out. So this is a true but completely irrelevant factoid. System noise floor will typically be dominated by the microphone and preamp, unless the preamp trim is mis-adjusted.

3) Because the system is capable of much more SPL than the gig or venue needs and the user never intends to operate at a point of over driving the amplifier's input--even if it is one of the few amps whose input can be overdriven by the mixer's highest signal.

This is actually a possibility but I would expect pretty rare.

I've only come across one amplifier whose input attenuator didn't operate like a true adjustable pad.

One more than I've ever seen... but who knows? Most do have an active stage before the passive pad so can still be overloaded in theory, but not in practice for decent designs.

I’ll point you to what was given to me years ago that has been one of the most valuable tools to help me understand gain structure and to get the most out of a system.
Take this paper on gain structure to woodshed for a little study time.
http://shure.custhelp.com/ci/fattach/get/58/0/filename/gain_structure.pdf

It has also be updated and is on the ProSound Web here: http://www.prosoundweb.com/article/how_do_you_set_system_gain_structure/

Yup, Chuck's old article is a classic.

JR

 

[Ryan Lantzy]

Re: Thoughts about amplifiers at 3/4 volume.

Gentlemen,

There are but three reasons that a competent installer/system engineer would raise the sensitivity of the power amplifier's input stage. In order of importance, they are:
1) Obtaining a system’s maximum dynamic range. If the mixer is capable of a +26dBu output and the amplifier will deliver max gain voltage with a signal of +4dBu when the attenuator is at its highest sensitivity, then that system has lost 22dB of dynamic range.

2) Maintaining the system’s lowest possible overall noise floor. Every component in the audio chain has a noise floor. If one component in that chain is operated close to its floor the system's overall noise is raised. You may have heard an improperly set gain structure that was so far from optimum that the entire system sounded "thin."

3) Because the system is capable of much more SPL than the gig or venue needs and the user never intends to operate at a point of over driving the amplifier's input--even if it is one of the few amps whose input can be overdriven by the mixer's highest signal.

I've only come across one amplifier whose input attenuator didn't operate like a true adjustable pad.

No one is arguing that you shouldn't optimize the systems gain structure for low noise floor, maximum dynamic range, and maximum system output.

The question, is whether doing so on the amp matters. I contend, that it in fact does not. The corner case is what JR mentioned, having long unbalanced runs of signal cable or maybe some other noise that crops up in the connection from the piece of gear immediately before the amplifier.

Because amplifiers have 20-30 dB more of SNR than the rest of a system, all that needs to be done is to bring them inline with one another. It really doesn't matter if it's done at the DSP output, mixer output, or amplifier (or via inline pads). (Edit: as long as it's done immediately before the amp.

I'm not sure I understand #3. Turning down the amp gains never sets the maximum output of the system.

 

[John Roberts]

Re: Thoughts about amplifiers at 3/4 volume.

I'm not sure I understand #3. Turning down the amp gains never sets the maximum output of the system.

This one is a little convoluted. If you pad down the input to the amplifier "enough", the previous stage will clip or run out of drive capability before driving the amplifier to full power. Crude but arguably a limit to max power output. The power limit is not very precise as operators can still drive this earlier stage past clipping and increasing the average power, just like clipping an amp (that doesn't have an active clip limiter).

Since this could require 15-20 dB of pad, a popular myth is that turning down the amp less than that will have any affect on max power.

JR

 

[Brad Weber]

Re: Thoughts about amplifiers at 3/4 volume.

I'm not sure I understand #3. Turning down the amp gains never sets the maximum output of the system.

I don't think that is what David said, I took his comment to be more along the lines of being able to adjust the system for a desired nominal output when running everything prior to the amp with good gain structure. For installed systems or any systems with a specific application it is not uncommon to select speakers based on factors such as coverage, off-axis response, sound quality, response, etc. and to end up with more output capability than required for the use. In that case you can run with the amps wide open and turn down the levels elsewhere or turn down the amps (or insert a pad as Ivan noted). The latter often makes more sense especially if you have a DSP before the amps that doesn't support adjusting the analog output levels. You are right that this is not setting the maximum potential output of the amp, however it is setting the nominal output.

 

[Kip Conner]

Re: Thoughts about amplifiers at 3/4 volume.

I have a passive Meyer UPA rig with the Meyer processors. While I was hunting down the right amp for the system I was told by the Meyer engineers that in order to get the correct balance for the limiters to work I needed to either roll back the input gains on the amp -10dB on the modern amp gain or put a simple whirlwind pad on the amp inputs. The limiter on the processor wants to operate with 26dB sensitivity while the newer XTI series operate at 34 or so.

The conversation started with Meyer because I was looking for an amp with low outputs for the 1401 HF drivers and I wanted to buy the same amp for all the drivers in the system. Four XTI 2000's power the 2 tops and 2 usw1 subs beautifully with the pad usage.

 

[Ryan Lantzy]

Re: Thoughts about amplifiers at 3/4 volume.

I have a passive Meyer UPA rig with the Meyer processors. While I was hunting down the right amp for the system I was told by the Meyer engineers that in order to get the correct balance for the limiters to work I needed to either roll back the input gains on the amp -10dB on the modern amp gain or put a simple whirlwind pad on the amp inputs. The limiter on the processor wants to operate with 26dB sensitivity while the newer XTI series operate at 34 or so.

The conversation started with Meyer because I was looking for an amp with low outputs for the 1401 HF drivers and I wanted to buy the same amp for all the drivers in the system. Four XTI 2000's power the 2 tops and 2 usw1 subs beautifully with the pad usage.

I'm not going to contrdict Meyer engineers, but if you are using the M1-A processor there are sense lines that connect from the amplifier output terminals back to the processor. I'm not sure what the amp gain has to do with it. The processor should just be looking for a peak voltage (or other undesirable condition) to begin limiting or corrective EQ. For instance, older Nexo processors that have sense lines also have a amp gain adjustment (26, 32, or 38 dB) but the manual states that this has no bearing on the limiting or protection, it just helps to adjust the output of the process to be more inline with the amplifiers gain (i.e. pad to correct gain structure).

Additionally, I don't think using XTi amplifiers which have a DSP stage (and AD/DA stages) with a processor that has amp sense lines is a good idea. The delay in the audio processing incurred inside the DSP of the amplifier will throw off the calculations/comparisons being done by the external speaker processor (if sense lines are being used). I've never set up UPAs myself or used the M1-A processor and I don't know which Meyer processor you are using for certain, but these are just some friendly warnings for you.

 

[Brian Jones]

Re: Thoughts about amplifiers at 3/4 volume.

The amp is not being "throttled back", the control affects the input sensitivity(voltage needed) to achieve rated power at the proper point along the curve. Amplifier gain is fixed. Having faders running around "0" is better for a few reasons, mainly the beasts are designed to operate best there.

I think you have that backwards. I own a Crown iTech and in the application guide it talks about attenuators and input sensitivity. Unlike most amps, the iTech's input sensitivity is adjustable. It also has attenuators like most amps. It talks about the best way to reduce maximum output power being adjustment of the input sensitivity because that will result in better SNR. This is especially useful when using it with a speaker with a lower than recommended power rating. The 8000 produces 4000w @ 4 ohms per channel and I use it to power an SRX728S on one channel and a pair of SRX712M's as mains on the other channel. Both present a 4 ohm load and are both rated at 1600w RMS/3200w program so I adjust the input sensitivity and leave the attenuators at 100%.

Technically, with the Itech, you would use the same knob to adjust both input sensitivity and attenuation with it's menu driven system but it doesn't have a level marking on the knob for obvious reasons and when not in menu mode, the knobs adjust attenuation.

Most amps though, like I said, have fixed input sensitivity.