Frequency and Mixer Output Votage

[Tripper]

If I send a single pure test tone of one frequency (say 440 Hz) though my mixer until the output meters peak and then I send a different single test tone (say 80 Hz) through my mixer until the output meters peak, will the resulting output voltage be the same? What about a higher test tone...say 5,000 Hz?

 

Thanks!

[LeftCoast]

What you are asking about is the frequency response or the normalized relationship between input signal and output signal in the frequency domain, of the mixer.

 

In general, most mixers and other signal chain equipment are designed to be reasonably flat across a broad frequency range, but will exhibit roll off at higher frequencies above 20Khz (particularly if they employ digital effects and want to attenuate aliasing noise from sampling) and will have some roll off below about 80 hz to attenuate any 60hz/50hz noise from the mains.

[Tripper]

 

What you are asking about is the frequency response or the normalized relationship between input signal and output signal in the frequency domain, of the mixer. In general, most mixers and other signal chain equipment are designed to be reasonably flat across a broad frequency range, but will exhibit roll off at higher frequencies above 20Khz (particularly if they employ digital effects and want to attenuate aliasing noise from sampling) and will have some roll off below about 80 hz to attenuate any 60hz/50hz noise from the mains.

What I'd like to do is use these test tones to set the gain structure in my system. Currently, I'm using pink noise to do this. I'm really wanting to know if I can use a single test tone (per crossover band...2 in my case, highs and lows) to know the voltage coming out. Test tones don't fluctuate in voltage nearly as much.

 

Can I use these test tones instead of pink noise to set the gain structure?

[boseengineer]

Most mixers will give you the same output for both frequencies (provided your EQ is flat and all frequency shaping is turned off).

Since sine wave really don't behave like real music, their usefulness depends on what you are trying to do. What exactly are the knobs you are trying to set in your gain structure ?

[Tripper]

 

Most mixers will give you the same output for both frequencies (provided your EQ is flat and all frequency shaping is turned off). Since sine wave really don't behave like real music, their usefulness depends on what you are trying to do. What exactly are the knobs you are trying to set in your gain structure ?

The gain knobs.

 

I'm trying to do a couple of things. One, is to learn. I'm experimenting with different tones, pink noise, and taking voltage measurements in an effort to gain knowledge about my system and sound in general. I'm a software developer so being able to logically connect how everything works under the hood helps me understand the system. The more I understand, the better choices I can (hopefully) make. Using pure test tones, the voltage coming out of the mixer is more constant and helps me understand what's going on.

 

Second, I'm trying to set the gain structure of my system accurately. I want to set the system up so that everything in my signal chain clips at the same time. After that, then I can add headroom where necessary in certain components.

 

But I would say that the intent of the question was more fact finding than anything else.

 

Thanks for your reply!

[agedhorse]

Like Boseengineer says, the output of all modern consoles are flat across the audio band within say +/-1dB. I know of no modern console that rolls off anything at 80Hz or even 50Hz. That's a falsehood, which has nothing to do with reality.

 

1. Do not use test tones with speakers, especially high frequency drivers. Power ratings of speakers do not generally corrolate well with test tones.

 

2. Do not fall into the mythhood that all stages must clip at the same time. The power amp limit point needs to be ~9-12dB BELOW the clip point of the drive electronics in order for the limiters to work properly (or at all).

 

3. All modern gear will function well, with low noise, if the gain structure of the drive electronics is set to unity (no gain boost or cut) and the amp input sensitivity reduced by 3dB. This will be default give appropriate headroom with your DR-260 (which has a lower clip point than your mixer).

 

Trust me on this.

[Tripper]

 

3. All modern gear will function well, with low noise, if the gain structure of the drive electronics is set to unity (no gain boost or cut) and the amp input sensitivity reduced by 3dB. This will be default give appropriate headroom with your DR-260 (which has a lower clip point than your mixer).

The 260 has internal jumpers which allows that input to clip at +30dB. The folks at the dbx forum suggest I set this so that the MixWiz3 and the 260 can clip at the same time.

 

I'm really liking your suggestion to just put everything at unity and reduce the input sensitivity by 3dB for when we play live. But also, when I'm just messing around in the garage, I'd like to experiment to see what I can learn.

 

Thanks so much for your reply.

[agedhorse]

On the DR260, the input may clip at +30 but the ooutput is still limited to about +20, so you are straddling the fence since the MixWiz clips at about +26dBu so IMO, you are better off leaving the DR-260 alone and let it clip at it's default. They can not clip at the same time due to this. This will be also part of the headroom assumption. (note there is a conflict in MOL's in the manual, I have found it to be closer to +20dBu)

 

You do not want to reduce the sensitivity of the power amp by 9-12dB because you will have just eliminated ALL of the potential for limiting, and you will just pass the clipped drive electronics signal onto the speakers at high level.

 

With the DR-260's maximum output level of +20dBu, this is the limiting factor in the signal path and subtracting the amp sensitivity from this number gives the actual operating headroom the the amp's limiter may act upon. Generally, this will be 20-6dBu = 14db. Now subrtact 3dB per my suggestion and this gives 11dB which is about right.

 

Alternatively, you could calculate this a different way using the DR-260 as the limiter but you still have the output level to contend with, so even if you were to set the limiter to +20dBu, and the amp sensitivity to +20dBu, and the input sensitivity to +30dBu and subtract the 4dB difference between the clip point of the MixWiz and the input sensitivity of the DR-260, you stand to gain maybe 6dB of noise improvement but this is also the maximum system headroom under ideal conditions and there's no account for boosts and cuts due to eq under full scale conditions.

 

There are a lot of potential approaches to the headroom/noise tradeoff, but rarely are the technical aspects of the inner workings of the operating level equations discussed, nor the likely limits that such approaches may encounter.

[dboomer]

 

2. Do not fall into the mythhood that all stages must clip at the same time. The power amp limit point needs to be ~9-12dB BELOW the clip point of the drive electronics in order for the limiters to work properly (or at all).

 

 

Here's where we perennially disagree. I say BOTH things can happen at the same time. You have to include the basic understanding of how to "drive" the system. If you frequently clip you mixer (or other gear in the line) then Andy's method makes lots of sense. But understand you are making the choice to give up lots of system power in trade for safety/reliability. 9 db below clipping means that your amps begins to limit itself at roughly only 10% of its potential output.

 

In my method, by safe driving, you do not give up that power ... but to be perfectly fair you can get into big trouble quickly if things go bad.

 

What I think you need to do is truly understand what costs the trade-offs carry with them and then decide how you should proceed. If you couldn't be bothered then by all means go Andy's route.

[Tripper]

 

Here's where we perennially disagree. I say BOTH things can happen at the same time. You have to include the basic understanding of how to "drive" the system. If you frequently clip you mixer (or other gear in the line) then Andy's method makes lots of sense. But understand you are making the choice to give up lots of system power in trade for safety/reliability. 9 db below clipping means that your amps begins to limit itself at roughly only 10% of its potential output. In my method, by safe driving, you do not give up that power ... but to be perfectly fair you can get into big trouble quickly if things go bad. What I think you need to do is truly understand what costs the trade-offs carry with them and then decide how you should proceed. If you couldn't be bothered then by all means go Andy's route.

Thanks for the reply. I'm trying to make an informed decision. All this stuff is VERY interesting to me as well....almost makes it fun.

 

Part of the problem is it's taking me a while to figure out the different schools of thought on this. Since it appears there is more way to skin this cat, I, as a noob to all of this, have to decipher the differences to try to make sense out of it. Right now, I think I'm going to have to play it safe until I know what I'm doing.

 

A 150 MPH hotrod may not be the best car for a kid with a new license. Trouble is, I LIKE hotrods!

[agedhorse]

 

Here's where we perennially disagree. I say BOTH things can happen at the same time. You have to include the basic understanding of how to "drive" the system. If you frequently clip you mixer (or other gear in the line) then Andy's method makes lots of sense. But understand you are making the choice to give up lots of system power in trade for safety/reliability. 9 db below clipping means that your amps begins to limit itself at roughly only 10% of its potential output. In my method, by safe driving, you do not give up that power ... but to be perfectly fair you can get into big trouble quickly if things go bad. What I think you need to do is truly understand what costs the trade-offs carry with them and then decide how you should proceed. If you couldn't be bothered then by all means go Andy's route.

 

 

No Don, you are misunderstanding this completely. You do NOT give up ANY power.

 

The headroom of the drive electronics is completely independant of the output power. The headroom is maximum amount of limiter complence available ONCE LIMITING BEGINS TO TAKE PLACE. This is, by definition on amps with limiters, the clipping point of the power amplifier, or 100% of rated power. Beyond this point you will increase the average power while limiting the peak or maximum power, thus increaseing the average to peak ratio of the program and the thermal energy delivered.

[Tripper]

On the DR260, the input may clip at +30 but the ooutput is still limited to about +20, so you are straddling the fence since the MixWiz clips at about +26dBu so IMO, you are better off leaving the DR-260 alone and let it clip at it's default.

I also noticed that the 260 also has output jumpers as well that you can set to +30. Would it make sense to set these as well IF I were to keep the inputs at +30? Then, perhaps the mixer and 260 would be able to work better together?

[agedhorse]

 

I also noticed that the 260 also has output jumpers as well that you can set to +30. Would it make sense to set these as well IF I were to keep the inputs at +30? Then, perhaps the mixer and 260 would be able to work better together?

 

 

Unless there was a change made recently, the MOL is still +20 or 22dBu. The output jumpers are there to provide the necessary makeup gain to normalize operation back to unity. It's another confusing aspect of the gain/noise/headroom equation, and poorly explained or called out in the docs. They would have had to change the analog +/-Vcc to +/-18v to make this happe, and IIRC it's roughly +/-12v.

 

This is why I am kind of a hardass on this stuff, it's plenty confusing and there's so much bad information out there that's presented in a way that makes it SEEM reasonable unless you understand what's happening behind the front panel. Since I design this kind of stuff day in and day out, it's trivially obvious to me. Not so to the average Joe, or even some of the folks who should know better, because the information is poorly presented.

[agedhorse]

 

I also noticed that the 260 also has output jumpers as well that you can set to +30. Would it make sense to set these as well IF I were to keep the inputs at +30? Then, perhaps the mixer and 260 would be able to work better together?

 

 

They need to be set together to work properly, in order to normalize gain at unity. Otherwise there would be a ~10dB loss through the unit.

 

Note that the meter calibration no longer tracks when set off of the default values.

 

I would leave them where they are.

[dboomer]

 

You do NOT give up ANY power. .

 

 

Now that i read it again ... you are saying leave 9-12 dB of available "drive" level. That's the same thing I am saying ... I'm saying do it by restraining yourself where you would limit the possibility.

[agedhorse]

 

Now that i read it again ... you are saying leave 9-12 dB of available "drive" level. That's the same thing I am saying ... I'm saying do it by restraining yourself where you would limit the possibility.

 

 

My method doesn't limit the available power to 10%, it limits it to 100% and does not allow clipping of the amp or the drive electronics. I don't understand how you came to that conclusion.

[dboomer]

 

My method doesn't limit the available power to 10%, it limits it to 100% and does not allow clipping of the amp or the drive electronics.

 

 

I am in agreement that your method can allow 100% of amp power.

 

The drive electronics can still clip ... neither your or my method can prevent that. Your method has a 9-12 dB chance of preventing the mixer output from clipping by accepting 9-12 db loss of s/n ratio. Mine has the potential is flexible in that it can do either or split the difference. Please don't misread it and think that it can do both at the same time ... it cannot.

[Tripper]

 

3. All modern gear will function well, with low noise, if the gain structure of the drive electronics is set to unity (no gain boost or cut) and the amp input sensitivity reduced by 3dB. This will be default give appropriate headroom with your DR-260 (which has a lower clip point than your mixer). Trust me on this.

 

Tonight, I did just this. I set everything to unity (really, just insured the DR-260 inputs were at 0dB) and reduced the amp input sensitivity of both my GX5 (sp5's) and RMX 2450 (sp118s) to -3dB. Actually, they were both set about 3 o'clock. Nothing bridged. However, the tops were louder than the subs. So, in the crossover in the 260 I set the gain for the tops to be -7dB and the bottoms at 0dB. To my ears, this seemed to balance nicely.

 

It sounded pretty darn loud to me!

 

Does this sound right?

[agedhorse]

Yes, this is right in themiddle of the ballpark.

[LeftCoast]

 

Like Boseengineer says, the output of all modern consoles at flay across the audio band within say +/-1dB. I know of no modern console that rolls off anything at 80Hz or even 50Hz. That's a falsehood, which has nothing to do with reality.

 

 

The entire Mackie VLZ-Pro line of mixers have a 75Hz 18db/octave high pass filter on the mono inputs (the +0/-1 20hz to 20KHz spec is only from mic/xlr inputs).

 

The Soundcraft MH3 has a variable, bypassable 30Hz to 400Hz high pass filter on the mono inputs and a fixed, bypassable 80Hz high pass filter on the stereo inputs.

 

The Allen and Heath GL2400 has a 100Hz 12db/octave (bypassable) HPF on the input channels.

 

Are these modern consoles in your estimation?

[agedhorse]

 

The entire Mackie VLZ-Pro line of mixers have a 75Hz 18db/octave high pass filter on the mono inputs (the +0/-1 20hz to 20KHz spec is only from mic/xlr inputs). The Soundcraft MH3 has a variable, bypassable 30Hz to 400Hz high pass filter on the mono inputs and a fixed, bypassable 80Hz high pass filter on the stereo inputs. The Allen and Heath GL2400 has a 100Hz 12db/octave (bypassable) HPF on the input channels. Are these modern consoles in your estimation?

 

 

Dude, you do know there are bypass switches on those filters. Turn them off and you get 20-20k response. The variable HPF is also defeatable/sweepable down to "nothing". If you don't bypass the channel specific filters (and eq too... duh) then of course the response won't be flat, but we're certainly not that dumb. Your comments about the Mackie VLZ models is absolutely incorrect by the way. Your frequency domain comment in your first post was technically nonsense as well.

 

I also know that both Boseengineer and I have designed our fair share of mixer and preamp circuitry for our respective product lines, and grasp the agonizing details of how a modern console works. Just a hunch, but you probably aren't quite as familiar with console signal architecure as some of us here are, your comments indicate that to me anyway.

 

I don't know if you are trying to impress or confuse us here. If you are trying to impress, you should check the accuracy of your information before posting. If you are just trying to confuse, you are doing just fine.

[SoundMan]

I don't know if they've fixed this or not, but at full gain, most Mackies are down 10dB at 20 Hz. That's only through the XLR input. At lower gain positions they are mostly flat.

[agedhorse]

 

I don't know if they've fixed this or not, but at full gain, most Mackies are down 10dB at 20 Hz. That's only through the XLR input. At lower gain positions they are mostly flat.

 

 

Hi Jim... The last one I tested, a newer version was only a couple down but the max gain may have also been a few dB lower too. Maybe 2 fixes in one? Of course full input gain for a live console is absolutely unusable though. At any useable gain positions, they are indeed essentially flat.

[LeftCoast]

Dude, you do know there are bypass switches on those filters. Turn them off and you get 20-20k response. The variable HPF is also defeatable/sweepable down to "nothing". If you don't bypass the channel specific filters (and eq too... duh) then of course the response won't be flat, but we're certainly not that dumb. Your comments about the Mackie VLZ models is absolutely incorrect by the way. Your frequency domain comment in your first post was technically nonsense as well. I also know that both Boseengineer and I have designed our fair share of mixer and preamp circuitry for our respective product lines, and grasp the agonizing details of how a modern console works. Just a hunch, but you probably aren't quite as familiar with console signal architecure as some of us here are, your comments indicate that to me anyway. I don't know if you are trying to impress or confuse us here. If you are trying to impress, you should check the accuracy of your information before posting. If you are just trying to confuse, you are doing just fine.

 

 

My original comment was simply that most mixers are flat across a very broad frequency range abut there [i should have said MAY BE] a roll off in both the HF and LF. As this was the 2nd post in the thread, at that point it wasn't even clear what that purpose of the OP question was for gain staging. It was YOU who made the comments about dealing with reality and knowing of no modern console with a LF roll off.

 

I'm not trying to impress or confuse anyone, and I am far from an expert. If I've offended your sense of expertness - my apologies. I'm not an expert. I have a 25 year old associate degree in electronics, and BS in computer science and an MBA. I have worked as a soundman and played keyboards in bands for over 30 years, but I don't do this for a living. That probably makes me more of a knowledgeable hack. A little knowledge is dangerous.

 

It has been years since I did any circuit design, and I worked in telecommunications, not audio, but in my humble, non-expert opinion, it is quite likely in any signal chain that you will see some LF roll off as anything that is transformer or capacitor coupled (which is just about everything) in the signal chain is going to exhibit roll off. So without knowing the purpose for inputting a sine wave and measuring the output, without knowing the model of mixer, or what else was in the signal chain, I commented that there might me a LF roll off.

[agedhorse]

 

I commented that there might me a LF roll off.

 

You commented that there WILL be a rolloff and gave 80Hz and 100Hz in your examples. It was misleading and confusing, and I got a couple of PM's wondering how this was possible and would it affect subwoofers.

 

Of course there will be LF and HF rolloff, otherwise the response would be DC to Light... clearly not the case, nor desireable. Our collective point is thatthe rolloff is generally sub and super-sonic.