Post by: Ron Kimball on June 26, 2009, 03:13:10 PM
!
Post by: Greg Cameron on June 26, 2009, 07:50:53 PM
Now the cases where peak limiting may not be all that helpful is if 1) your power amps are too big for the speakers, 2) you overdrive the shit out of your signal chain or processor, 3) the limiters are of shitty quality and can be "punched through" easily, 4) sine wave heavy trance type music where you're pushing your system to the edge (in that case having your amps sized smaller than program power rating & closer to the rms rating would be a good idea). I find with cheaper DSPs, the limiters often not only don't sound very good when you're hitting them, they also may not limit all that well either. FWIW, peak limiting has been doing a decent job of protecting systems for a long time. Of course, these are systems with well matched amp, speakers, and processing. The article Bennett wrote talks about using much bigger power amps than the speakers are rated for, and that can work very well with properly calibrated rms limiting. And you likely to get a noticeable bump in average output and headroom by doing so. But for the time being, that takes some big wallet amps and processing.
As for how well the limiters in your DCX work in terms of punch-though and sound quality, I haven't used one. But I suspect other here have and will chime in.
Greg
Post by: John Ward on June 26, 2009, 08:29:17 PM
Post by: Jeff Wheeler on June 26, 2009, 08:59:34 PM
| Greg Cameron wrote on Fri, 26 June 2009 18:50 |
As for how well the limiters in your DCX work in terms of punch-though and sound quality, I haven't used one. But I suspect other here have and will chime in. |
If you ride the limiters in the DCX you will get some degree of unpleasant output pumping. How much depends on how you have the release time configured, program content, and how far you are trying to over-drive it. The same is true of the limiter in the DEQ1024. I tried to explain this to some DJs at a club once, but my advice was met only with "sound man, turn us up!"
Post by: Ron Kimball on June 26, 2009, 09:02:35 PM
| Greg Cameron wrote: |
Peak limiters are usually not useless so long as your amps are relatively matched to the program power of the speakers. |
?
Post by: Ron Kimball on June 27, 2009, 08:55:21 AM
- maybe I get to blow my first drivers today 
?
Post by: Nick Enright on June 27, 2009, 10:19:05 AM
I run a 2450 into the EVM-18 (200wrms/400wpgm/800wpk) all the time, with DJ's. In a sealed cabinet.
From the QSC Manual, FTC Power Output is 450w. (FTC is the measurement method)
Well all I would do is set the limiter and using some resources.. (couple of nice excel charts to calculate power output and to convert units, they're here somewhere) ..dial the limiters the appropriate # of dB down to limit the power applied to the cabs.
If you set the limiters without math. (setup but dont attach the speakers, run the amp till the clip lights light, then dial down the limiters till they go out.) I doubt you'll kill a sub. (USE THE RIGHT HPF! or maybe even dial it up 1-5Hz.)
Post by: John Ward on June 27, 2009, 10:36:00 AM
Anybody seen or demoed the Carvin 8 x 8 processor? Looks good from the marketing and at $900. direct I would like to learn a little more about it. Although Carvin has always been more of a prosumer level product some of their stuff is pretty amazing when it holds up. Unfortunately a lot of it hasn't held up or has been dummied down as their target market has always been the working musician.
Post by: Greg Cameron on June 27, 2009, 11:35:52 AM
Greg
Post by: Greg Cameron on June 27, 2009, 11:40:52 AM
| Ron Kimball wrote on Sat, 27 June 2009 05:55 |
Well, I was hopin' to use it to protect a pair of 200wrms PV118's run off an RMX2450 as the sole subs for an outside gig today I stupidly agreed to - maybe I get to blow my first drivers today ? |
I'm not sure what the problem is here. The PV118 has a program power rating of 400 watts, the RMX2450 450 watts. That's well within the limits of what's considered feasible. It's a perfect match up. I think you're over thinking this. Just set the limiter in the processor to allow mild to no clipping. Turn on the clip limiter in the amp for extra safety if you wish. You're done.
Greg
Post by: Jeff Wheeler on June 27, 2009, 01:16:14 PM
The DCX2496 does have a compressor on each output that could be used as a rough RMS limiter. I am not sure how much you can do on it before you run out of processor power. I've only had one for about a week and have only had it out once.
I agree with Greg that one side of an RMX2450 is not going to threaten a PV118. My DJ colleagues and I have been abusing the heck out of a pair of similar Peavey subs (that are so old the 1/4" jacks are developing rust) with an XTi 2000 in stereo for a year, and they still work fine.
Post by: Andy Peters on June 27, 2009, 02:19:24 PM
| Jeff Wheeler wrote on Sat, 27 June 2009 10:16 |
Does amp clipping (and the harmonic distortion that comes with it) cause damage to subwoofers? I thought no, and thus have not been afraid to drive my sub amps to mild clipping all night long. Honestly I figured any woofers would not be damaged by a clipped waveform. |
Of course they can cause damage.
Obviously the speaker cannot reproduce the harmonics. So those harmonics are just dissipated as heat in the voice coil. Extra heat doing nothing good for you.
-a
Post by: Greg Cameron on June 27, 2009, 02:31:11 PM
| Jeff Wheeler wrote on Sat, 27 June 2009 10:16 |
Does amp clipping (and the harmonic distortion that comes with it) cause damage to subwoofers? I thought no, and thus have not been afraid to drive my sub amps to mild clipping all night long. Honestly I figured any woofers would not be damaged by a clipped waveform. |
While Andy is of course correct about clipping being potentially harmful to any driver, the larger voice coils used on many sub drivers tend to deal with it a bit better than smaller drivers. I don't necessarily advocate mild clipping of sub amps, but it's a fairly common practice even with larger rigs to get a bit more omph out of the system, not to mention the added harmonics might sound desirable to some within limits. FWIW, my sub amps are run into mild clipping. Be aware that not all amps are created equal when hitting clip in terms of the way they sound. Some are less harsh then others, so even mild clipping may not sound very good with certain amps.
Greg
Post by: John Roberts {JR} on June 27, 2009, 03:04:16 PM
| Jeff Wheeler wrote on Sat, 27 June 2009 12:16 |
Does amp clipping (and the harmonic distortion that comes with it) cause damage to subwoofers? I thought no, and thus have not been afraid to drive my sub amps to mild clipping all night long. Honestly I figured any woofers would not be damaged by a clipped waveform. The DCX2496 does have a compressor on each output that could be used as a rough RMS limiter. I am not sure how much you can do on it before you run out of processor power. I've only had one for about a week and have only had it out once. I agree with Greg that one side of an RMX2450 is not going to threaten a PV118. My DJ colleagues and I have been abusing the heck out of a pair of similar Peavey subs (that are so old the 1/4" jacks are developing rust) with an XTi 2000 in stereo for a year, and they still work fine. |
It's been a while since we had this discussion..
The larger issue with driving sub amps past clipping, is that the main way you get amps to clip is by turning up the system gain. While the loudest parts are now clipping, the quieter parts are also playing much louder.
Arguably a very dynamic signal where the momentary peaks are many dB above the average level, you could clip all the time with little risk of overheating. For typical, less dynamic program material, turning up past clipping is accompanied by higher average power levels.
Note: Compressing the program material is counter productive for loudspeaker protection. Compression raises the average level relative to the peaks. The amp will be heating the drivers even more at clipping than before the program material was squashed.
JR
Post by: Jeff Wheeler on June 27, 2009, 04:56:15 PM
| John Roberts {JR} wrote on Sat, 27 June 2009 14:04 |
The larger issue with driving sub amps past clipping, is that the main way you get amps to clip is by turning up the system gain. While the loudest parts are now clipping, the quieter parts are also playing much louder. |
Your point is well-taken. I assume the sub peaks will feel pretty muddy (from clip distortion) before this becomes a factor anyway.
| Quote: |
Note: Compressing the program material is counter productive for loudspeaker protection. Compression raises the average level relative to the peaks. The amp will be heating the drivers even more at clipping than before the program material was squashed. |
I figure unless the speaker companies are going to provide a de-rating chart for increasing ambient temperature (e.g. 100F outside jobs) I will continue to be conservative and bring enough rig that I don't need this kind of output compression to manage RMS power.
While I don't have hard figures for my program material, I would guess my typical sub operating range of ~37Hz to 80/100Hz has peaks at least 10dB over RMS anyway, so I will continue to size subs for peaks and not worry too much about long-term heating.
My operation is small enough that I do not need to worry about perfectly sizing amps, either. I guess if you are bidding on the U2 360 tour you have to figure everything out pretty carefully, but this is not the case for me, as my biggest jobs involve an unimpressive two half-empty amp racks.
Post by: John Chiara on June 27, 2009, 09:49:51 PM
| Ron Kimball wrote on Fri, 26 June 2009 15:13 |
Does anyone know FOR SURE if the limiters on the DCX2496 are peak or RMS? If peak (= kinda useless?), can the dynamic eq be used as an RMS limiter? Does anyone know FOR SURE if the thresholds are offset off of +22dbu or ? Behringer manuals sure suck ! |
I know they click pretty good when you hit them!
John
Post by: Ron Kimball on June 30, 2009, 01:08:32 PM
?My question to Behringer support:
>Is the limiter "peak" or "rms"?
>I'm guessing that to set the limiter to output 0dBu I need to set
>its threshold to -22dB to compensate for the +22dB maximum
>output?
Answer from:
>Glenn S.
>Your BEHRINGER Customer Support Team
>You set the threshold at the db where you want the limiting to
>start. Set it for +22db if that is the level where you want the
>limiter to start working.
Of course the limiter only goes up to 0dB so setting it at +22db (or even +1dB) is impossible. I just "updated" the question with this:
>Since you can't set the limiter above 0dB I'm sure you are
>completely unfamiliar with this product. Any chance of passing
>it on to someone knowledgeable about Behringer's products?
Post by: Greg Cameron on June 30, 2009, 03:37:14 PM
Greg
Post by: Greg Cameron on June 30, 2009, 03:44:03 PM
Greg
Post by: Scott Smith on June 30, 2009, 03:51:30 PM
I like to set mine so they start working the same time the amp's limiter starts flashing. I find that each device seems to have a slightly different life of it's own, and they seem to react differently, to different signals. Ideally, you should only be seeing the occasional flicker, from either. How hard you push your system's average output will probably be the deciding factor.
Post by: Ron Kimball on June 30, 2009, 04:11:52 PM
| Greg Cameron wrote: |
After a quick glance at the manual, it's pretty simple. The limiter can be set for limiting between 0 to -24dB which should equate to +22dBu to -2dBu of actual output. |
. Certainly you can feed a sine wave into it and - without speakers attached - set the limiter so your amp outputs the proper voltage sine wave to match the rms rating of your sub but if they are peak and not rms limiters you'd be severely limiting your real "music power" to well below the sub's ratings. The only loudspeaker management system that I've verified can do both peak and rms limiting is the one built into the I-Tech. I personally would be happy enough with a properly sized amp with integral clip limiter to limit the peaks and a DSP based crossover/rms-limiter to limit the temperature of the voice coils.
Post by: Greg Cameron on June 30, 2009, 04:20:49 PM
Greg
Post by: Greg Cameron on June 30, 2009, 04:38:21 PM
And as I've stated before, don't over-think this. If you're amps are sized such that they're close to the program power rating of your speakers, just set the peak limiters on the processor to keep the amps from clipping or at least clipping hard on peaks. If you want the 'tits' setup with hardcore rms and peak limiting, you shouldn't be looking at Behringer gear anyway. You'll need to plunk down some cash and get I-Tech amps. Or better yet, buy self powered boxes where all this stuff is figured out and implemented already. Going a piecemeal route with getting another external device to handle rms limiting and such is adding complexity that you really don't need to for a basic setup. Sure, you can do it, get the compressor/rms limiter, calculate the max rms voltage your speakers can handle, set the compressor according, etc. But it's really not necessary. There are tons of rigs out there running simple peak limiting and not blowing up. If you're going to be pushing your particular setup that hard, you should get more rig.
Greg
Post by: Nick Hickman on June 30, 2009, 05:12:03 PM
| Ron Kimball wrote on Fri, 26 June 2009 20:13 |
Does anyone know FOR SURE if the limiters on the DCX2496 are peak or RMS? If peak (= kinda useless?), can the dynamic eq be used as an RMS limiter? Does anyone know FOR SURE if the thresholds are offset off of +22dbu or ? Behringer manuals sure suck ! |
I don't have any inside information, but I did do some tests and there appears to be a lot going on with the DCX2496 limiters. They appear to combine both RMS and peak detectors.
The threshold you set appeared to be the RMS limit level relative to +22dBu, and the peak limit level appeared to be 3dB above the RMS limit level. This was based on testing a full bandwidth output and hitting the limiter hard. Things could be different based on the high-frequency limit of the band-pass or how hard it was hit.
Here's the amplitude-over-time output from a 1kHz sine wave and 1kHz square wave, both with the same peak amplitude, and both in a full-bandwidth output under heavy limiting, on the same vertical scale. The sustained RMS level is the same in both cases, and the initial "peak" of the square wave is limited 3dB higher.
Limiting peaks to 3dB above RMS (if that is indeed what's happening) would seem a dubious design decision. 6dB above RMS (i.e. corresponding to the peak level entailed in AES loudspeaker power ratings) would seem more sensible.
Nick
Post by: Ron Kimball on June 30, 2009, 07:57:30 PM
. If the limiters are rms I would expect the output to be only down 3db (=.707 of) from the sine - this seems to show much more?
Post by: Nick Hickman on July 01, 2009, 09:00:22 AM
| Ron Kimball wrote on Wed, 01 July 2009 00:57 |
Interesting data! It would be more useful if there was a vertical scale and even more useful if there were graphs of the inputs also. Pink and/or white noise would be good to try too . If the limiters are rms I would expect the output to be only down 3db (=.707 of) from the sine - this seems to show much more? |
Exactly so. Output measurements were as follows:
Sine wave (limiter bypassed): +1.9dBu RMS, +4.9dBu peak
Sine wave (limiter engaged): -2.2dBu RMS, +1.0dBu initial peak
Square wave (limiter bypassed): +4.8dBu RMS, +4.8dBu peak
Square wave (limiter engaged): -2.1dBu RMS, +1.4dBu initial peak
I'd encourage you to go ahead with your own measurements!
Nick
Post by: Scott Smith on July 01, 2009, 09:51:46 AM
| Nick Hickman wrote on Tue, 30 June 2009 17:12 |
...I did do some tests and there appears to be a lot going on with the DCX2496 limiters. They appear to combine both RMS and peak detectors... |
GREAT TESTING! Is that typical in limiters? It seems to be a good solution for my needs.
Post by: Ron Kimball on July 01, 2009, 11:56:09 AM
| Scott Smith wrote: |
GREAT TESTING! Is that typical in limiters? It seems to be a good solution for my needs. |
. The DriveRack 260 also has crappy docs as to what theirs does .Post by: Ron Kimball on July 01, 2009, 12:02:48 PM
| Nick Hickman wrote: |
Output measurements were as follows: |
Post by: Nick Hickman on July 01, 2009, 12:22:19 PM
| Ron Kimball wrote on Wed, 01 July 2009 16:56 |
The reason for my post is that there doesn't seen to be any published data as to how these loudspeaker management limiters work. It does start to look like the Behringer does RMS limiting which is exactly what I want . |
If my observations were correct, it effectively does RMS limiting whilst the signal crest factor is less than 3dB. Above that, it limits based on peaks. That margin really should be user-adjustable (or 6dB!).
I agree that loudspeaker processor limiters aren't always well-documented. The BSS FDS-366, for one, offers an adjustable margin between its RMS limit and its peak limit.
Nick
Post by: Ron Kimball on July 01, 2009, 12:28:21 PM
| Nick Hickman wrote: |
The BSS FDS-366, for one, offers an adjustable margin between its RMS limit and its peak limit. |
.