Gain Structure and Premature Clipping

Tripper

As I was going through the Gain Structure instructions I noticed what I thought to be premature clipping in the 260. I noticed that if I turn down the input gain on both channels (-6 to -8 in my case) that the "clip" indicator on the LCD screen would come on when the signal that was fed from my mixer resulted in a +18 flash on the threshold meters of the 260. All 260 functions were turned off. I'm not using the GUI (yet...USB High Speed Serial Adapter ordered)

When the input level is set at 0, then the clip indicator on the LCD illuminates at the same time the +22 LED illuminates on the threshold meter, as expected.

I'm assuming that anytime the "clip" indicator comes on in the LCD that it really is clipping. So, should I ignore the meters?

Thoughts?

Thanks!

Gadget

I have noticed that the Meters on the front panel fall below the performance of the clip indicator, and I think we can blame something called meter ballistics for that. In the old days when needle VU meters were used overshoot was common place, then that added a peak led, and then full LED meters, Today I believe the limitation is a combination of the reaction time of the circuit, and the turn on/off speed of the LED, and those are the limiting factors... the GUI however reacts far more accurately.
Gadget

Tripper

Thanks for the reply! I'll use the GUI when I get my cable and we'll see what happens.

Thanks again!

Tripper

The GUI had the exact same effect. If I run a near clipped pink noise signal out of my MixWiz 3 (full red) then even if I pull the input faders down in the GUI (not the master) to around -8dB then I still get significant clipping (as indicated on the unit LCD) and neither the GUI or unit input meters are in the red. It's "less" clipping than if the fader is at 0dB. Maybe I'm significantly over 20 dB coming out of the MixWiz. I'll have to check with a voltmeter later.

So, it sounds like I'm clipping the input signal (which I'm assuming is +20 dB for the 260). I'm sure this has been covered but does the input need to always be less than +20 dB from upstream equipment regardless of what we set the 260 input faders for? The faders are POST AD conversion?

Yes, no, maybe?

Thanks!

Gadget

The Mixwiz has a max output into a 600 ohm load of +26dBu. Remember also that the pink noise has a crest factor of typically either 6db or 12db depending on the type, Some audiologists would argue that pink noise is not the best tool to use for setting up the actual peak system gain structure...for just that reason. Your certain that the limiter AND the comps are off also right?

G

Tripper

Yep, I'm sure the limiter and compressor are both off as well as all other modules.

I also tried it (just for giggles) with a pure 440 tone and it didn't clip even as the MixWiz was pinging red. I need to do some more research on how different frequencies, pink noise, etc... effect the voltage coming out of the mixer.

What I don't understand is....voltage is voltage so why would it matter what frequencies were used to setup the gain structure so long as the voltage coming out of the mixer is enough to clip downstream equipment. For instance, why can't I just use a 440 tone to run through the system and make sure everything clips at the same time...and then adjust for any headroom...and then call it a day?

Gadget

I guess that's pretty much what I'm saying.. pro's use pure tones to set gain structure and Pink noise to adjust for tonal balance and flat response...

You see, as I mentioned earlier, pink noise has a crest factor... an ever changing peaky mass of frequencies... kind of like a bubbling cauldron of frequency peaks... these instantaneous peaks can, and do exceed the major body of the sound. As a visual exercise put your hands side be side with your fingers pointing up...the fingers are the peaks and your palms are the main body of the sound (analogy) Those ever changing frequencies (like your finger length difference) all vary, and are up to 12 db in ...well crest... Like transient peaks on your sound system they are difficult to catch because they happen so fast...

Most of the traffic here is beginner to novice level ... and We try to keep the content ...well... less technical :oops: without drumming it down... so to speak...Once in a while we get someone, like yourself, that needs more...and so, I guess I have been wanting to modify the gain structure post to reflect more REAL WORLD PRO techniques without over doing it techspeak wise...there is a fine line where TMI causes overload, and were already WAY over the line with many here as it is...

Hope that helps?
G

mikejoeydbx

Hello All,

The Clip indicator, on the display, will light whenever any of the inputs or outputs clip.

Tripper,

for this application you will want to do this:

Set the input jumpers to the +30 dBu position (these jumpers are on the mainboard and described on page 73 of the owners manual). Setting these jumpers to the +30 dBu position , will allow the 260s analog circuit to accomodate the +26 dBu signal level from the mixer (along with 4 dB of headroom). You would now adjust your input gain (in DSP) to +4. Now the 260s inputs will clip at the same point as the outputs of mixer.

As for the test tones vs pink noise discussion. Test tones are more acurate, but take more care to get correct (as you have to take into account EQ and such and need to have a better understanding of audio and DSP) and pink noise is less acccurate, but easier to use, so I usually recommend this to the novice users.


Hope this helps.

Keep up the good work Gadget!

Tripper

Thanks guys! I really appreciate the help.

Tripper

This weekend, I set the jumper to +30dB. This worked out well and took care of the input clipping.

I have a question though. The instruction manual says the if you set the jumpers to other than factory default (+22dB) that the output meters would not be calibrated. But what about the GUI and particularly the various modules?

For instance, if I use the limiter and set the threshold to 15 dB can I be assured that the levels coming out of the 260 will never be greater than 15 dB? Is it accurate...or do I have to account for changing the input jumpers.

As a follow up question, is there any need for me to change the output jumpers to match the input jumpers? I didn't do this, so all the output jumpers are still set to +22 dB.

Thanks for your time!

Dra

In the end the calibration is not a real issue for you. The meters will still be usable as reference to peak and you still use the amps clip lights to set everything up. They will still be calibrated "to each other". So, I would not worry about it.

DRA

Tripper

Thanks for the reply!

In my case, I'm using an amp that is rated higher than what the speaker can handle so I want to limit to the speakers rating and not necessarily to the amp. The amp clipping won't be as helpful.

Of course, I can always measure the voltage coming out of the speaker cable and set the limiter to the voltage the speaker can handle. I suppose mathematically that would also answer my question as well.

I was also just curious.

Thanks!

Dra

That is what I did. My amps are rated 600w. My top speakers 400w (prog). That's 3x continuous (a bit much, for me). I did the math to get the volts and dialed the amps back to the desired voltage at mixer peak. Speakers get all they want and I have tons of headroom.

DRA

Tripper

Dra wrote:

That is what I did. My amps are rated 600w. My top speakers 400w (prog). That's 3x continuous (a bit much, for me). I did the math to get the volts and dialed the amps back to the desired voltage at mixer peak. Speakers get all they want and I have tons of headroom.

I'm running an RMX 2450 (500 Watts/side) for the subs. The subs I have are Peavey SP118 (600 rms/ 1200 program). Over on the Harmony Central forum, some of the folks keep telling me not to run bridged mono on the 2450 (1200 watts per speaker) but that seems to me to be a good match. 1200 watts into 1200 program speakers. I certainly don't want to blow the speakers.

Yesterday, I setup gain structure so that everything clips (after I changed the 260 input jumpers) and then I limited the subs until the RMX (in bridged mono) stopped clipping.

What do you think about this and what are the risks?

BTW, the I ended up setting the crossover gain with a -6dB differential between the tops and the subs because the tops kept overpowering the subs.

Dennis

Unless I missed it, you haven't said what Mid/Hi cabinets and amps you are running. If I read correctly, after you bridged the 2450, you still had to run your Mid/Hi's at -6dB(assuming subs were 0dB)...seems a little excessive.

Dennis

Tripper

Dennis wrote:

Unless I missed it, you haven't said what Mid/Hi cabinets and amps you are running. If I read correctly, after you bridged the 2450, you still had to run your Mid/Hi's at -6dB...seems a little excessive.

Dennis

Hi Cabinets: 2 Peavey SP5s (400 Watts RMS) driven by QSC GX5
Low Cabs: 2 Peavey SP118 (600 Watts RMS) driven by QSC RMX 2450

I've crossed over at 100 Hz. I've set the high gain at -4 dB. I've set the low gain at +2 dB.

Now, running bridged mono thumps the bottoms pretty good. It sounds good. I just don't know if it's too much.

Dra

HC - RMS rating max.
PSW - Peak rating (plus some)

They are both right. :? :shock: Headroom is meant to be just that... headroom (unused reserves, except for the instantanious need here and there).

Depends totally on what you're doing and if you run to the max capacity of the amps (even unclipped). You can use a 2000w amp on 500/1000/2000w subs, but try passing a 50hz tone at as little as 700w from the amp. You'll start smelling an odor in a matter of minutes and likely no sound in a few more. You might even see flames.

A lot of people split the diiference and match the amps to the program power of the cabinet. Yorkville, for one, recommends that for their speakers and QSC's amp selector does the same.

DRA

Tripper

Dra wrote:

A lot of people split the diiference and match the amps to the program power of the cabinet. Yorkville, for one, recommends that for their speakers and QSC's amp selector does the same. DRA

The QSC Amp selector recommends the RMX 2450 bridged mono for 2 Peavey SP118s.

But the HC forum thinks I'm crazy for doing that. At this point, I don't know what to do. I guess it's just a risk/reward thing. I want good punch but I don't want to hurt my speakers. I'm in a standard Rock band playing bars and it seems like I should be able to get good thump from two SP118's.

I think I'll try 1x RMS first and if that ain't enough I'll go bridge mono (2x RMS).

Thanks!

Dra

Logically if you can get by with less power to the speakers (NOOOOOO clipping) then that is way better for your speakers. Not a few people like to bump the clip lights on their SUB amps (never tops) to get a dirty sound. A clipped 2450 gives you 900-1000w, which is not neccessarily bad as long as you don't ride the lights. Harder on the amp, I suppose. Bridging is definately harder and bridging at 4 ohms, really harder.

DRA

Gadget

First off, 2 sp118's will, be BARELY enough even in bridged mode, remember, those speakers are rated @ 600/1200/2400 watts, the 600 figure is 600 watts pink noise 24/7... the RMX 2450 is rated @ 1/8th power @ 7 amps stereo @ 8 ohms...What that means is that most rock music uses only a fraction of the amps power capacity ...NORMALLY...Now you start adding Bridged and 4 ohms to the pot and the current draw for that amp can easily exceed 20 amps @ 50% power!

Here's the thing, you have a speaker that has a 2400 watt PEAK rating, and a 1200 watt program rating, you have an amp that can deliver 1200 watts per speaker... but no more, you can achieve the rating but you'll have NO headroom at that level. The thing about headroom and the ratings is... that if you HAVE the 2400 watts available, and run the speakers @ 1200 watts and limit them there, you'll have headroom over the amount of any peaks that exceed the limiters... with the 1200 watts available, anything OVER the rated 1200 watts is DOUBLE the output voltage and THAT is when you let the smoke out, not typically with good clean power...

Now I'm NOT saying that you should try and feed 2400 watts to those speakers all day, they won't take it! But having a car capable of 150 MPH doesn't mean you need to go that fast.. ever right? But you CAN if you really need to...

I'd run the bridged mode IF I have sufficient power (current) for the amps...turn the LIMITERS ON THE AMPS OFF, set up your subs as a MONO block (together) and then do a good gain structure....this will get the mixer clipping JUST before the amps, then set the 260 to BRICK WALL LIMITING(PKstop on), and turn the AUTO off, set the over easy @ 2 overshoot @ 3db and then lower the the THRESHOLD till the meter just starts to show gain reduction...you should be fine like that..

Gadget

Tripper

Correct me if I'm wrong but just because the speakers are rated at 600/1200/2400 watts doesn't mean you have to drive that much into them. I mean, as long as the signal is undistorted and you are using a 600 watt amp or below you can't hurt the speaker at all no matter what you do.

Of course, it's all of matter of SPL I guess. If 600 Watts per speaker doesn't give me the SPL I want then I have no choice but to run bridged....or get another amp.

Maybe this is what you mean when you say "BARELY enough".