As I said previously, this current style of limiter doesn’t actually do this as much as people think, and because they sound better than non-lookahead hard brick wall clip, tend to be more likely to be abused and cause thermal damage to woofers than just leaving the limiter off.
An automatic gain control that simply cuts the volume down to some amount – say, 25% voltage of prior volume position – for a few seconds when the threshold (top LED of -1dBFS?) is lit might do the trick, though. Then give it a few seconds gradual release.
A compressor-limiter with the necessary adjustable parameters is going to have a more difficult time doing this because the gain reduction is versus the amount past threshold. The current limiter is probably infinity to 1 compression ratio so nothing gets past 0dBFS. You can put the total gain reduction at like -12dB (4:1) at a lower threshold (say, -9dB), minimum hold to like 3 seconds, eventual gradual release to about 2 seconds, and attack time necessary to trigger to 0 seconds. That’d probably trigger at the second to top LED, but how effective, I don’t know.
Increasing the attack would give something closer to a thermal limiter, where they have to be past the threshold for at least a little while before gain reduction occurs. True thermal limiters, though, tend to analyze and use the average loudness over time rather than just the peaks… think of them not just using the top of the meter bounce, but the troughs, too.
Our more advanced compressor, by the way, would not need ‘makeup gain’, since we have trim-gain controls already and we’re not looking to do production-style compression ever without us already boosting the level into compressor stage first.
I don’t understand: if I have the limiter activated, how can I damage the woofers? You can tell me that when the signal is too high and the limiter starts working then I get a distorted signal, and on this aspect I agree. But why when the limiter goes into action, the woofers could be damaged?
I think it is easier to damage the woofers when the limiter is deactivated (admitted to have under-sized woofers compared to the power of the final amplifier).
I would’ve thought that in these modern times, any amp that would be between the Prime kit and the woofer would have the technology to sense impending doom and deal with it before it reached the cone/coil.
It’s not easier to damage a woofer when the limiter on a digital DJ mixer is deactivated. It is rare to kill pro woofers with over excursion due to peaks. Thermal damage is the more frequent cause of woofer damage, and that’s the result of average level, not peak. You can increase RMS both by increasing the volume of the signal or instead from increasing how compressed it is by bringing the troughs up as the peaks stay the same. Either will cook the woofers. High-frequency clip harmonics are capable of killing tweeters, but that’s very rare nowadays with the way pro tweeters are usually limited or voltage protected already (old days, you did it with a light bulb in the wiring path), and the hard 0dBFS brick wall inherent in a digital mixer is going to prevent those ultrasonic peaks from being louder, anyway. And as I said, a limiter doesn’t sound as bad and therefore is more likely to be abused and crushed into than with it off, and is therefore more likely to result in increased RMS and woofer destruction. BTW, you can’t really kill a woofer with high-frequency clip harmonics as a woofer can’t do anything with them anyway.
Most pro amps now have output limiters to prevent the up-to-4X-the-rated-voltage-of-the-amp clip harmonics that would potentially easily kill drivers if the amp was pushed hard into clip. An amp’s power rating specification is pre-clip. This is particularly important with class D amps that have the power supply coupled with the outputs, and particularly more so with certain cheap class D amps that don’t have certain discrete components to prevent the power supply unloading on clip directly into the drivers even more than 4X. On these class D designs, since they all usually require digital conversions stages, anyway, designers tend to simply lookahead-limit peaks in the digital domain on the amp, with the power supply basically acting as an additional sort of crude thermal limiter, since switching supplies that are sort of the basis for class D anyway tend to have difficulty putting out continuous high voltage for much length of time. So if you match a class D amp appropriately to drivers’ capabilities, you can actually get quite good protection against killing the drivers if you do it right. Some really well-designed amps also have very flexible compressor/limiter settings and dedicated thermal limiters.
As far as I know, killing your high end is usually not a result from high voltage, but from the signal going from sinus (nice wave) to block as a result of clipping. And clipping is what is being prevented by the typical limiter, this should keep the waveform intact, albeit substantially flattened.
The “block” you’re talking about is the aforementioned squaring shape that produces the ultrasonic harmonics. In power amplifiers that are not output-limited in some way, these harmonics can be far beyond the amp’s rated power which is not going to happen on most front-end gear, especially on digital stuff. Hell, the Prime player processing right now produces a ton of ultrasonic garbage, but it’s not at risk of being 4X the hard 0dBFS digital brick wall.