processors with hard limiting / dcx replacements

[Grant Bunter]

Play a test tone and the limiter will pass the signal then ramp back down to 60v. If I pump the fader up and down I can easily see 100v+ on the DMM.

What amp are you getting this with?

I'm just wondering if part of the problem is you have to much amp.
All limiters can only reduce voltage to a certain point, for example, with the DCX that's a max of -24dB. Another forum member recently had a problem with a JBL 1200 series amp and that was capable of 80V (average) output, with maximum limiting making the limit light engage reasonably constantly.

This means RMS power could possibly be getting past the limiter. Much more of a concern for voice coils than a few volts here and there...

[sine143]

lol... I think -24 db should be enough (max voltage/8)

[billkatz]

Actually max voltage/16.

24 dB / 6dB per doubling = 4 doublings. 1/2 * 1/2 * 1/2 * 1/2 = 1/16. or 20 * log (1/16) = -24.08.

I think I've got it after 3 edits...

[frpcraig]

You do need a meter to see if the limiter is working.
The amp can clip all it likes and the limiter keeps it at 60 volts that's the point.

The purpose of the limiter is to prevent amp clipping and over-excursion of the speaker.
The digital meter is fine for adjusting the limiter threshold with a steady-state sine wave signal, but if you're checking for limiter overshoot, a scope is the right tool for the job. A digital meter won't react fast enough to provide any meaningful reading.

I thought the limiter was to limit the voltage and nothing else?
The limiter doesn't know if an amp is clipping at 60 volts. All it is bothered about is supplying 60 volts.
Is that not the case? Or have I got it wrong again!

[byacey]

The limiter actually limits the line line level drive signal to the power amp, and by extension it limits the output signal of the amplifier. The limiter doesn't have any means of detecting the output of the amplifier; the operator setting it up with an AC voltmeter becomes the missing link in the feedback loop.

That being said, yes the amp can clip, but only if the limiter threshold is set higher than the clip point of the amp. Doing so wouldn't achieve any purpose, and you may as well not have a limiter. Now, with the BFM horns, we want to further limit the amps, not just against clipping, but also for a maximum safe operating voltage that can be applied to the drivers, to protect them against mechanical damage.

This is why I responded to the post regarding the "The amp can clip all it likes and the limiter keeps it at 60 volts that's the point."

If the amp is properly limited at 60 volts, it should never approach clipping unless the amp is under sized and can't swing 60 volts into a load. So no, if the limiter is properly set, the amp cannot clip all it wants; the limiter prevents this from ever happening.

[frpcraig]

I think I get that Byacey. (sorry if this counts as a hijack!)
I just remember Bill's sticky about an underpowered amp not being able to bust a speaker even if it is clipping if the speakers can handle the extra power anyway??

Can you help with this please?

My Berry EP 2500 does about 65 volts with a 60hz sine prior to clipping.
Full clip it does 100 volts.
I limit at 50 volts.

I have another amp which is in full clip at 50 volts and I can limit the dcx to 50 volts.
Other than a crap sound why would the speakers care if the amp was clipping itself to death??
I thought if the speaker can handle 50 volts no probs then that's it?
Or does the current value go up and melt things?

Please move this if it annoys the original chap who posted.

[byacey]

When an amp starts clipping the RMS value goes way up (+30%) compared to a sine wave. This creates added heat in the voice coil that can quickly burn it up if run under these conditions for any period of time. If the speaker coil is robust enough to handle this for any period of time it may survive, but it sure won't sound very good.

In the case of the amp that is only capable of 50V, I would dial the limiter threshold back so it limits at about 45 volts to prevent clipping.

[Tom Smit]

Or does the current value go up and melt things?

If the amp is often running to it's limit, the current will be high for it and thus shorten the life of the amp. This is a good reason to have a bit of headroom, keeping in mind that too much headroom is more difficult to limit with an outside processor.

[djtecthreat]

Play a test tone and the limiter will pass the signal then ramp back down to 60v. If I pump the fader up and down I can easily see 100v+ on the DMM.

What amp are you getting this with?

RMX2450, but I cant tame any of my PLX or PLX2s either.

[BrentEvans]

I can offer that the PeakStop limiters on dbx products operate on a VERY short but >0 attack time. The difference is that they don't square off the wave like the DCX does. They will allow a brief peak to pass over the threshold, but will still hold it to a much reduced level so that the wave doesn't square off and sound quality is maintained. Try backing off the threshold by a db or two and see if the results aren't more along the lines you want.

BTW... the 2450 is only rated to 450 watts (60v) at 8 ohms, and the clip lights start to come on a bit before that anyway, so it doesn't sound like you're in much danger with things the way they are with the dbx. I can, however, speak very highly of the Symetrix and Ashly products, with an edge to the Symetrix.

Symetrix Solus has been talked about here a few times, but they offer a more basic product called Jupiterthat works quite well as a speaker processor. The Jupiter is technically fixed architecture but offers a number of "apps" that can be used for various purposes.

The Sound Reinforcement 6 and 12 apps work great as a speaker processor and has 15 band graphic EQ, parametric EQ, HP and LP filters, limiters, and feedback fighters (which work REALLY well). The main difference between the arrangement of the components. You can get Jupiter in 4i/4o, 8i/4o, and 4i/12o versions, and while it's a bit pricey ($1500 for Jupiter it's less than Solus and works really well.. a Jupiter 8 could do stereo mains, mono subs, plus 5 monitor mixes. That means you really wouldn't need any other drive rack processing... so it's about the same price as three dbx Driverack PA2 units and gives you one extra channel of monitor processing you wouldn't have with 3 Driveracks. The only thing it doesn't do is Auto-EQ, but since it has parametric EQ, you can use RoomEqWizard's auto-eq feature with the =http://www.parts-express.com/dayton-au ... erence mic and just input the EQ curve manually.

[escapemcp]

To avoid the squared off waves from the DCX and emulate a behaviour more like the Driverack (I think), would a touch of heavy compression a few dB below the limiting threshold work to make the limiting less severe, (i.e. a bit more like analogue gear I'm thinking)??

Something like the graph below. Blue is the compression threshold and red is the limiting threshold. Excuse my dodgy sinewave, but I think you get the idea.


It will take more voltage to actually hit the limiter, and when it does it doesn't hit it as 'hard'. Effectively, it's just 2 stage compression, but would it work (i.e. sound better) for all source material?

[BrentEvans]

To avoid the squared off waves from the DCX and emulate a behaviour more like the Driverack (I think), would a touch of heavy compression a few dB below the limiting threshold work to make the limiting less severe, (i.e. a bit more like analogue gear I'm thinking)??

Something like the graph below. Blue is the compression threshold and red is the limiting threshold. Excuse my dodgy sinewave, but I think you get the idea.


It will take more voltage to actually hit the limiter, and when it does it doesn't hit it as 'hard'. Effectively, it's just 2 stage compression, but would it work (i.e. sound better).

You can add a compressor to improve sound quality as the thing hits the limiter, but you shouldn't increase the limiter threshold. You should add the compressor with a lower threshold than the limiter after the limiter is set.

[Grant Bunter]

This is getting weird.

BTW... the 2450 is only rated to 450 watts (60v) at 8 ohms

and further looking says that the RMX 2450 is rated at 750W/4 ohms, or roughly 55V.

So how the heck does one end up with 100V readings?
I understand that transients can perhaps do that, yet my thoughts are that transients are often so brief that, as such, a DMM may not even pick them up if you're playing music content.

Looking again at the RMX2450, 100V at 4ohms is 2500W, and that is greater than supposed bridged output of 2400W...

[byacey]

Don't forget what I mentioned earlier. The digital meter measures RMS voltage. When clipping occurs, the output becomes more of a square wave, and the RMS value goes up to 100% instead of the normal 70.7% of peak. This would account for the misinterpreted and erroneous readings.

[88h88]

Last time I was getting weird readings on my voltage meter it turned out it was fried and giving me false readings.

I managed to get my hands on another relatively quickly and it determined my limiting was spot on but the meter was toast and giving me ridiculous readings of ~100v.