Limiter Procedure and New Technique - need critique
olorin
Posts: 15
I've tried the Super Duper Limiter thingy but find that the amp still clips at what I perceive to be too low a level. Plus dealing with setting the gain on the amps on a darkened stage can be a bit of a PITA. What I'd like to accomplish is to be able to dime the amp's sensitivity so that the amps can only be truned down as needed, and prevent the amp from ever clipping by using the crossover gains on the DRPA.
Following the gain structure of the individual channels and the output of the mixer (making sure that none of them are near clipping) I would then set my xover gain so that the amp is never clipping. This way I get a brickwall limiter by using the xover gains, and I don't have to worry about where the amps sensitivities are set. Of course I would match the amp to the speaker so that it doesn't mechanically or thermally blow up the speaker, but at least that's on me.
Critique, please, and tell me where I am wrong.
Following the gain structure of the individual channels and the output of the mixer (making sure that none of them are near clipping) I would then set my xover gain so that the amp is never clipping. This way I get a brickwall limiter by using the xover gains, and I don't have to worry about where the amps sensitivities are set. Of course I would match the amp to the speaker so that it doesn't mechanically or thermally blow up the speaker, but at least that's on me.
Critique, please, and tell me where I am wrong.
0
Comments
When you did your gain structure did you not get the mixer to clip then the amps and then back the amps off a bit till they didn't clip any more? I don't understand exactly what your getting at here? by "dime" do you mean crank wide open? I've never heard that expression. Remember that if you start getting those crossover gains too divergent (like +3 and -10) your going to be moving those crossover points acoustic center VERY FAR from the electrical center you have chosen. So depending on the amps you have what you want to do might not be advisable...perhaps more info here would help?
G
Here's my system:
Board - Yamaha 16/6fx
Tops - either 2 Yamaha S115 v's (250 RMS0 or 2 PV PR12's (200 rms) - depending on the room
powered by a Crown 802xlsd - I use one side to power the tops - 800 watts into 4 ohms, other side is for monitors. I always fidn that I have to turn these down compared to the subs.
Subs - 2 Yorkville YX18S - 400 program is what they're rated at each @ 6 ohms, powered by a Crown Microtech 1200 - I use each side to power each sub - roughly 390 watts going to each sub. I use boundary loading to increase the db's of the subs, which works quite well.
And yes, I set the gain structure as per the directions, but I find the difference in volume between recorded music (which, along with pink music, is what I set the system up with) and a live band is quite significant. I also recently used the system for a local "talent" show, and the difference between the one band that had no clue how to use mic technique and the rapper that decided to clip my input channel with it turned all the way down had me thinking that i should get some better system protection than the peak plus that the DRPA is giving me.
I have also recently invested in a compressor for the vocals, which has done wonders for allowing me to bring the vocals up in volume without all the crazy dynamics that were happening. I don't know why I didn't get one sooner.
So - I just want the amps to be wide open - that's one less thing to worry about with the system. And no other person who thinks they know what they are doing can come along and mess with those knobs by trying to "turn them up". I'd rather have to lower them to achieve system balance.
Therefore - I figured that if I make sure that the mixer doesn't clip and that the signal going into the DRPA doesn't clip and I then adjust the xover gains to never clip the signal top the amps with the amps set wide open then I am A-ok. On a side note, when I have set the limiters per the Super Duper thingy they are almost always engaged when i send a signal from the board to the DRPA. I had to dial back the Crown 802 to around 1 o'clock and the Limiter at 4 Overeasy to around -3 db. That does not make any sense to me.
Then when I do the Limiter thingy I have to set it all the way down to -2 db in order for the amp not to clip. And then the entire signal is being limited by the DRPA. As I have stated, this does not make any sense to me at all.
Playing around with it I was able to get the knobs all the way up on the 802, the limiter up to 6 db with Overeasy 4, and the high db on the xover at -3 db. I have the low xover db at +7. I can then balance the system by simply turning one of the amps knobs down, and no one can mistakenly turn the amps knobs up and cause damage. The amps will only clip if I really push the master fader close to clipping. When I have it at 0 on the mixer I probably have a good 6-8 db of headroom before the amp, mixer and DRPA will clip. This makes more sense to me.
I measured the voltage out of the 802xls and found to my surprise that with my method I could only get roughly 18-20 volts out of the amp, with the amp gain dimed, and the mixer and DRPA's levels up fairly high.
When I turned the amp's gain knobs down and increased the xovers db's on the DRPA I got the voltage up to around 35 volts. The xover's are very close to each other now, as far as db's - the high is +4 and the low is +7. I believe the top voltage that should be coming out of the amp is 56 volts, based on the specs. So I still have some playing around to do. But that is a significant increase in voltage.
Has anyone else measured the AC voltage at the speaker outs after their signal is processed via the DRPA? I wonder why I can't get the near top voltage level that should be coming out of the 802xls?
lets start there and see if we can find out whats happening here. Do you know what the input sensitivity of the amps is? I'm awaay from home now and can't spend a bunch of time looking stuff up..
G
A Yamaha 16/6 fx mixer. Max output is 24 dB. Nominal is +4. For pink noise at the input channel I have the pfl set at +3 on the board's meters. It seems that this board's individual channels register at +3 on the meters vs. the Master sends - when I pfl the input channel it's at +3 on the meter. When i check the afl and bring the channel fader to 0 and the master fader to 0 the afl meter level is at 0.
So with 0 coming out of the board the DRPA input meters read between 0 and +5. I currently have the Graphic EQ fairly flat and no PEQ's activated. For the xovers I have a BW18 on the low with @ 42.5 Hz and +7 db's. For the high xover i have a LR24 set at 118 Hz at +5 db's. The output meters on the DRPA read between +5 and +10.
I have the limiters on both set at 4 overeasy. The High Threshold is set at 6 db's and the Low is set at 12 db's.
I use a Crown 802xlsd (one side which specs at 800 watts into 4 ohms, 1.25 v sensitivity) for the highs and a Crown microtech 1200 (1.4 v sensitivity, each side which is roughly 370 watts into 6 ohms) for the lows. Speakers are either Peavey PR12's (200 watt RMS @ 8 ohms) or Yamaha S115v's (250 watts RMS @ 8 ohms) for the highs and Yorkville yx18s (400 wattt program @ 6 ohms) for the lows.
With this setup I am getting only around 30 volts coming out of the 802xls when the master fader is at 0. If i dime the master fader the 802 amp clips like crazy. The Microtech doesn't have clip indicators but I am measuring roughly 45 volts coming out of that amp. If I understand correctly as long as the voltage that the speaker is rated at is not being exceeded I am fairly safe as far as speake protection goes. But i don't think I'm getting as much volume as I should.
the instructions in the START here thread specifically say TURN OFF ALL LIMITERS AND COMPRESSORS! Then we are are trying to get all the devices talking the same language (voltage wise)...there are a couple of ways to do this...the START HERE thread is for beginners and novices because it has less potential MIXER to JUST start clipping, on input or output (here we suggest -0- for beginners on the xover gain controls) then, make sure the levels in the driverack are NOT clipping. Then with the speakers disconnected from the amps use the gain controls on the amps to get them to JUST clip... that should be marked as the highest point the amps ever get turned up to.
If you want to set the amps to wide open you will need to use the driveracks xover gain controls to get the amps to JUST clip then back them down a bit so that mixer is clipping, the driverack is not, and the amps are not...
What your posting is not what I expect to see, so maybe I'm not reading what your saying right...if you have the compressors or limiters on THIS METHOD WILL NOT HAVE ANY MEANING, and will be null and void!
G
However, even with the limiters off I cannot get anywhere near the 56 voltage rating out of the speaker terminals on the 802xls without the 802 clippping like crazy. Is it the amp, is it me? That is my conundrum. The readings I listed are the ones set per the gain structure method. it seems that diming the 802xls knobs is couunterproductive as far as trying to get the most power out of the amp given my equipment.
DRA
Do you have an oscilloscope? That might be a better way as you can calibrate the display, and get a real clear picture of what your dealing with.
G
That brings up a question, what ARE you using to determine the output voltage? What signal, and what meter?
BTW that thread is here:
http://srforums.prosoundweb.com/index.p ... msg_329854
You should peruse it...
Else you have to decide on which kind of failure you want to protect against. The slow, melt the voice-coil kind, or the not-so-slow, beat the cone to death kind. (don't take too literally)
DRA
Knowing WHEN your equipment is in danger of clipping and AVOIDING it is the beat way in my opinion.
If you have brick wall limiters and have WAY too much power use this method:
viewtopic.php?f=60&t=2242
to determine what amp power and speaker power you need.
G
The pink noise source I'm using is a wav file thru my laptop into my mixing board. I'm bringing this level at pfl to +3 on my meter - this is equivalent to 0 on my afl with both the channel fader and the master fader set to 0.
I am measuring the voltage out of my amps speaker terminals with a multimeter. I do not own an oscilloscope.
I understand that I might not be able to measure the voltage exactly but there was a significant upward trend in the voltage when I backed off the amps' knobs and played around a bit with the gain structure in the mixer and the DRPA. I'm simply trying to get this voltage as high as possible while at the same time protecting the speakers from blowing up due to catastrophic mechanical failure - NOT due to operating them all night at too high a level (heat failure).
Set the amps sensitivity @ the 4oclock position and then use the xover sensitivity to just (not) clip the amp... It looks like there really isn't much more gain available, so you could "dime" it and set the sensitivity on the DRPA xover points to just below clipping, when the mixer IS @ clip (give yourself some headroom on the amps...a few dB at least)
G
Then, do a little experimenting with the "knee" and see what you like best. set up the limiter @ 1 for soft knee operation and 10 for hard knee operation (or somewhere in between I use 4 and 6 a lot) and then slowly reduce the "threshold" till the meters start to react, the gain reduction meter should not be doing anything yet. EVEN with hard knee limiting you shouldn't hear much at this point. In hard knee operation, when the gain reduction meter shows activity...you will most likely NOTICE the effect on the sound, in 1, or 2 (soft knee) you may not notice till you get the meter moving some...
NOTE: you can now compare what you know you have for a max signal level, and what the limiter is set ant and know how much overshoot you have before clipping occurs...
NOTE: you can lower the threshold more in soft knee operation without it being noticeable, but it WILL overshoot more when "events" happen...BUT you can also get into a war with the limiter REAL easy by not remembering you have the limiters set, and trying to get more volume with the master faders... it is a vicious circle when that that happens so remember to back off the limiter .. set the volume, and then reset the limiter...
If you see the meters higher than you expect and system volume NOT tracking with that ... check the limiter level first.. it's probably going to be showing LOTS of activity on the gain reduction meter
hope that helps
Gadget
G
We played a gig this past Saturday at this tiny bar and with the board going to only +3 at it's peak - and staying at 0 or less most of the time - my Crown 802xls with Peavey PR12 mains could not be heard that well at the front of the bar (I was actually told it sounded like a stereo system at the front of the bar, which was only about 30-40 feet away). It was a long and narrow bar - I felt like I was playing in someone's basement.
I had assumed with 800 watts into 4 ohms (I used the other side of the 802 for monitors) and with my subs I should have been peeling the paint off the walls of this tiny bar if I wanted to.
Do you measure with an SPL meter? Is that full band going at top volume, vocals only, etc. etc?
Obviously we need the vocals to cut, and we keep our stage volume nice and low. So I always try and get the vocals to where I want them first.
It appears that the cabs are 95db@1w 1m. At a distance of 50 ft, you should get about 105db with only 125w or so. You do have something soing on.
DRA
I then test the vocals - I'm using an SM58 - and while I realize there's going to be minimal bottom end I can't get the vocals that loud. At a shout into the vocals I am nearly peaking the pfl - when I back off and sing "normally" it is getting up to around 0. With my board that's an 18 db difference. I have also incorporated a compressor on the vocals to get rid of the crazy dynamics. It has actually improved the volume.
Am I being too conservative with the individual trims on the vocals? How do you guys set your trims for dynamic vocalists - my voice is very much like Springsteen - from a growl to a whisper and back.
Yes, @ the mix position, usually 7 meters in a small bar. Yes I can peal the paint off the walls, I know when I crank the subs "things" mysteriously fall off the walls. But then I have more power running my horns than you have for your tops. But as I said that's a "more system" thing. Because I HAVE it doesn't mean I HAVE to use it all...and the band is a full band, playing classic rock. I also run bi-amp monitors, that have more power than your tops.. again, I have it I don't usually use it...
Hmmm, sounds to me like there are some "holes" in your sound. Are we talking about tops only here? you have never mentioned subs but once If this is a fullrange application, and your using a 2 way system, well the low end would be missing.
Those tops, and th4e Yamaha's have very peaky performance, and are quite harsh in the high end... Auto eq makes a world of difference in those speakers...but lets look at a couple setups first...
Fullrange tops:
1. In the wizard select a 2x2 configuration
2. either select your speakers from the menu or select "custom" it really makes no difference.
3. set the HPF @ 65hz BW18 (minimum... peavey doesn't provide a -3dB spec that I could find -10 is 54hz)
4. run the tops off the left hi out and the monitor off the right high out
2 way system:
1. Set up a 2x4 config
2. again either select the speakers you have or custom, it matters not
3. Set the HPf for the subs @ 45hz BW18 and the LPF @ 100hz LR24
4. set the HPF for the tops @ 100hz LR24
5. run the tops out of the left high out
6. run the subs out of the left low out
Are you running monitors? you could run those out of the right hi outs, they would share the HPF of the tops.
Perhaps if you tell us what YOU have set up we might see where the problems lies...
G
Again, those tops are peaky in the HF region and require lots of eq to make them sound smooth
I assume you don't have the measurement mic?
In a 2 way setup what do the tops sound like without the subs turned on? what do the subs sound like with the tops turned off...
G
Again though, that was recorded music, so I wonder if live music would be different?
And yes, at the gig I had my subs. I coupled them together and tried to boundary load them next to the wall. They don't seem to be the problem. It's the tops and lack of volume of the vocals.
Here's exactly what I used:
Yam MX16/6 mixer to the DRPA to the 802xls (sensitivity was around 1 o'clock - anyhting above this clips the amp) using one side to the Peavey PR12's for tops. I use the other side for monitors.
Crown Microtech 1200 (one side per sub) to the Yorkville yx18s for subs. Sensitivity was wide open. This amp doesn't have a clip indicator, and I didn't hear anything distorting in the subs.
XOver set at BW18 42.5Hz for lows, and LR24 set at 118Hz for highs. I also have the highs and lows crossing at 118Hz. The Xover highs were set at +5 dbs and the lows were at +7 dbs.
I pfl'd the main vocals to give me just below clipping at the highest shout I could do. Average was probably around 0 on the meter. I then set the background vocals, mic'd guitar amp, acoustic guitar, kick drum, bass ,digital piano and synth/midi controller at roughly 0 pfl on the mixer. We keep a low stage volume so that I can put everything thru the PA. On the afl the meter never got above 3. With my board this gives me around 10 db's of headroom.
The meters on the DRPA read roughly +5+10 on the highs and 0 on the lows for inputs. The outputs read around +5 on average. I had the limiter set at 4 Overeasy for both highs and lows, at roughly 8 db for the highs and 12 db for the lows. I did notice that the limiter was in effect for nearly the whole show. Lightbulb just went off I think -is this the issue - should I set the threshold for around 16-18db's if I really only want catastrophic protection?
Set the gain for normal singing and use mic technique (pull away and turn your head) for screaming.
DRA
viewtopic.php?f=60&t=2228
Please detail how you accomplished this.
**NOTE** what ever you set the HPF for the tops at will be the HPF for the monitiors as well
Is that a DRPA or DRPA+
Looks like you were limiting the HELL out of the system!
Looks like you need to review the process more carefully. ESPECIALLY the limiter procedure. Note that I said get the system running (with the band) and settle on a volume that gives you (at least 8-10 dB) headroom, THEN set the limiter with the threshold ALL THE WAY UP start lowering it till you just start to see the meters flicker, at this point you won't hear any effect from the limiters, lower it further till the gain reduction JUST starts to flicker, then back it off a tiny bit till it no longer flickers... IF YOU PUSH THE MASTER FADERS UP NOW THE GAIN REDUCTION WILL TAKE OVER AND YOU WILL SQUASH THE SOUND.. and you'll hear it! You cannot set a hard and fast "I'm going to limit the signal to
+10" you have to use the gain structure to determine the limiter setting.
There are some variables that can cause this NOT to work well...did you do it on a hard surface? was there a bunch of stuff near the speaker, or was it clear for at least 5-8 feet around it? Did you try a nearfield (on axis, or aimed at the point between the woofer and horn, mic a few feet away method and compare? If it's the DRPA+ the thing works remarkably well even in a room, so try it...
I find that on my systems, that they sound fine after auto eq for well done pre recorded music...and live sound. I generally test with the Eagles "Hell freezes over" Hotel California, Steely Dan, Dread Zelppelin, and pink floyd as well as some rap if I'm doing a DJ KJ thing...They do require a different gain technique however in that live sound is DYNAMIC where recorded is squashed pretty much lifeless...
I find that I generally run the show well below -0- on the master faders, and never even reach +10 EVER... so I set the limiters on the loud dynamic passages and I never usually even see the yellow or red lights on the meters on the DRPA even flicker...
I don't use the DRPA for monitors. I simply send the signal out of the board to a Peavey EQ, to the 802xls and then to the monitors. I have never had any issues with this method. I cut the eq slightly at 2k, and cut by 12 db everything under 100Hz.
For the indoor mic technique I mimiced what you did - set a piece of wood on a mic stand between the woofer and tweeter of the tops. Pointed the speaker on a downward angle. I also used Steely Dan on playback (My Old School) as well as some Keith Urban stuff.
I have always tried to set the limiter using the recorded music, and not the band playing live because I knew I wouldn't have time to do it at a gig. I am the guitarist, lead singer, and de facto sound man, all at once.
I guess I'll set everything (gain structure, pinking the system, etc.) all over again, and not touch the limiters until we do a sound check? Does this make sense?
Once again, I can't thank you guys enough for taking the time to walk me thru this.
G
When I redid that voltage test at the speaker outs of the 802xls guess what - I got to the 57-62 volts exactly when the mixer, DRPA and amp all clipped. So I was able to get to the rated output of the amp.
What I found was VERY interesting. When I was at 0 on the channel and Master faders (with the pink noise) I would only get around 20 volts coming out. When I increase the Master Fader to around +5 on my board I got the volts up to 30. When I then increased the channel to +5 I got up to around 40 volts. My meters on the board went to around +3. This would leave me with around 12 db's of headroom.
Now I finally understand the whole concept of RFTG (rig for the gig). You want the most powerful amp possible so that you don't have to push any of the other points in your signal chain close to clipping in order to get the volume you need. It's quite interesting when you can actually see this in practice by measuring the voltage. I've been running the system way too low all this time. If my math is correct 20 volts was only sending out roughly 50-60 watts to the speaker. - 120 watts to the pair. And I was wondering why my volume was so low? Yikes.