Loudspeakers Looking for the right speakers for PA System

[Meyer]

Hi all. I'm shopping for speakers. I stumbled across this forum and found an excellent response that completely cleared up the whole issue of Continuous, Program and Peak power ratings for speakers which definitely helped.

So here is the situation: I was providing sound for an outdoor stage a couple of weeks ago at Seattle's Pike Place Market and really kind of wished I could get a little more juice out of my system.

My amp is a Mackie M1400, great amp, but in stereo mode it pushes 300W per side. So I found a second Mackie M1400 on eBay and its on its way. I was thinking that I could run two matched amps in bridged mode, one per channel, and then I'd be running 1000W per side. That is more power than I suspect I'll ever want or need.

And of course that kind of power is going to punch the cones right out of the old 15" two-way speakers I've been running for the last fifteen years. I don't even know what they're rated at, actually, so now I'd like to replace them with something newer, lighter, and matched to my amps.

I do run a compressor/limiter between the board and the amp, BTW.

When I play in one band and do sound for two others, I get to use my friend's JBL JRX115 speakers, and I really like them. But the power rating is 250W program (I think) and 1000W peak. I read that you really want your amp to be around 1.8 to 2.2 times the program rating of the speaker, yes? So that speaker would be a bad match.

So I could spend the money and go to the JBL MRX 515s, which are rated at 400W/800W/1600W. Not ideal, but I like the company.

If I do that, can I not ever run the speakers on just the one amp at 300W per side? I've been told that if I go below the continuous rating, I could damage the voice coil or other components. That's one question.

Another is do the JBL MRX515s seem like an adequate match for the two Mackie amps?

And the last one is do any of you have a favorite speaker that WOULD be a great match for the two Mackie amps running in bridged mode at 1000W per side?

And one final question: these amps have the usual gain control, one for each side. When I turn that gain control down (say to 50%), am I effectively reducing the wattage by 50% and could go under a speakers' minimum power requirement, or are the two unrelated?

Sorry for the long post, and any help would be appreciated.

Rob

 

[David Ashton]

What you absolutely must do is read up on gain structure, this is the most vital knowledge you will ever gain[pun intended]
Google gain structure, there are articles by Rane, Yamaha, Mackie and most serious players.
The gain control on your amplifier will almost certainly NOT need to be at full to give you correct gain structure.
Correct gain structure means that each component in your sound system is giving you the optimum signal and the lowest noise, basically all units produce a constant noise [the noise floor] if your amplifier is running at full then you are amplifying that noise to the max, and to avoid feedback you are having to reduce the signal to compensate, this reduces your dynamic range and reduces the quality of your sound.
Many pro's fail to understand this basic rule.
The gain on your amplifier is there to match the input of the amp to the optimized output of he previous stage.
http://www.live-audio.com/studyhall/gain.pdf
So you adjust the input of your mixer to match what signal is coming in, you then optimize the output of your mixer to about 0db, you then adjust the input of the next stage to give optimum output and so on till the last stage is to set the gain on your amplifier.
This is a simplified version of what is required, but I'm pointing out how important gain structure is, not writing a book.
Chuck Mcgregor has written possibly the best article, live_audioStudyHall-Gain Structure if you can find it.

 

[TassieBogan]

If you're going to run in bridged mode, be aware that the minimum impedance for bridged amplifiers is double the minimum impedance of the same amplifiers in dual mono operation.

Example - Two amplifiers, each rated 100 watts maximum into 4 ohms, in bridge mode they will appear as a mono amp rated 200 watts into 8 ohms.

Also, speakers do not have a minimum amp requirement per se. But your amps do have a max, and trying to drive a speaker on an under powered amp will lead to damage to both components.

Edit: if you do run in bridged mode you will also likely need a specially wired speakon-4 cable. Check the wiring of the amp - it should be somewhere on the back

 

[museav]

If I do that, can I not ever run the speakers on just the one amp at 300W per side? I've been told that if I go below the continuous rating, I could damage the voice coil or other components. That's one question.

And whoever told you that could not be more wrong. Not only is running well below the rated power completely acceptable and common, it is virtually inevitable. Think about it this way, say your speaker is rated at 100W continuous, 200W Program and 400W peak. The continuous rating is 6dB below the peak rating, which is a result of the test signal used for the ratings, however the crest factor, the difference between the peak and average levels, of an actual live music signal is often 10dB to 20dB or even more. If you have a 400W peak level and a 10-20dB crest factor that then relates to an average level of 40W to as low as 4W, well under the 100W continuous rating. So it is quite normal for a speaker that is being pushed to its limits for peaks to still to spend most of the time at a fraction of the rated power, much less speakers not being pushed that hard.

Another is do the JBL MRX515s seem like an adequate match for the two Mackie amps?

And the last one is do any of you have a favorite speaker that WOULD be a great match for the two Mackie amps running in bridged mode at 1000W per side?

First, by changing from one amp in stereo mode to two amps in bridge mode you went from 250W per speaker to 850W per speaker (those are the specs for the M1400 in full range operation). With the speaker remaining the same that's a 5.3dB increase, definitely noticeable and probably around a 50% perceived increase in loudness, but perhaps not as significant a difference as some people might think (3+ times the power does not equate to 3+ times as loud).

As far as speaker selection, there are many other potential considerations including who is operating the system, transport and storage considerations and the acoustics of the environment in which the speakers will be used, but in general selecting a speaker should always consider a) what area are you trying to cover, b) how loud you need/want to get, c) what type of music or content is involved and d) how much you have to spend. Selecting a speaker based on it's power rating relative to the ratings of an amplifier does not really address any of these factors.

On even a most basic level, the rated power is simply how much input signal the speaker can handle under certain conditions without some type of failure, it does not reflect the output level of the speaker which in turn is not the same as perceived loudness. Don Davis, author of 'Sound System Engineering', liked to comment how people bragging about how many watts their systems had were often simply bragging about how inefficient their systems were. The audience dopes not care how much power you are putting into the speakers, they care about how much sound comes out of them.

If Speaker A and B are otherwise equal but Speaker A outputs 100dB at 1m with 1W and Speaker B outputs 90dB at 1m with 1W of power then with the same amplifier you will get very different results with the two different speakers. This is reflected in the speaker Sensitivity, which is usually noted as the sound pressure level output at 1M with a 1W or 2.83V (which is 1W into 8 Ohms) input. The higher the sensitivity, the more output with the same power. Going back to the 5.3dB you gained going to two amps in bridge mode, that same difference could have been obtained by using the same single amp and speakers with a 5dB or more greater sensitivity.

Then you get into patterns, response, size, weight, mounting options and so on. So you could 'match' a speaker based on the amp but then that is all you have done, you have not really addressed anything relating to the performance of the system or the compatibility with the potential applications. So it is important to consider some of these other factors when considering or recommending any speakers.

And one final question: these amps have the usual gain control, one for each side. When I turn that gain control down (say to 50%), am I effectively reducing the wattage by 50% and could go under a speakers' minimum power requirement, or are the two unrelated?

Simple answer, in virtually all cases what you are referring to are are level controls rather than gain controls. They adjust the level of the signal going into the amplifier circuit, they do not affect the amplifier gain itself or the potential output, only the input signal level needed to get a certain output. An X Watt amplifier is an X Watt amplifier regardless of where the level controls are set, it will just take a different input signal level to get that output depending on where the controls are set.

Some amplifiers do indeed have adjustable or variable gain, however these are typically software or switch settings that are independent of the level controls and are usually found on higher end products.

 

[Meyer]

Arggh. I probably spent a half hour composing a good followup to those excellent responses, and the board had logged me out meanwhile then lost my response on submit - so I'll have to recap in short form.

First, thanks for the great responses - this is really helping me. My followup by numbers:

1) David and museav: gotcha on gain and I'll do the reading. But I'm satisfied that I'm not endangering the speakers by merely turning those gain detents down. And actually, the manual says that leaving them all the way up is fine, as that matches most pro +4 dBu inputs (and I run a Behringer UB2222FX-Pro mixer and a Behringer MDX2200 Composer Pro Compressor/Limiter between the board and amp).

2) Thanks TassieBogan, and I'm in good shape on running Bridged: Mackie makes it easy - I just flip a switch to 'bridged' and plug the banana connector in to the positive terminal of both channels and off I go.

3) On the underpowering issue, I may still not be understanding so here is a concrete case: If I use my one Mackie amp in stereo mode pushing 250W into 8 Ohms to run that JBL JRX515 that is rated at 400W/800W/1600W (continuous/program/peak), would that matchup be likely to cause harm to the speaker (or amp)?

4) But museav - THANKS! You pointed out that my amp bridged actually pushes 850W into 8 ohms bridged. I was going by this website: Mackie - m1400 But that 1000W figure is either incorrect or I am misinterpreting it. The manual that you referenced is a later edition than the one I have, but I revisited my manual and found a reference buried in there that confirms that yes indeed I would be running 850W per side as opposed to 1000W. This still puts me out of the range for the JBL JRX115, as JBL recommends an amp in the 250W to 500W range, unfortunately. Still, I'm now working with better information.

5) As museav points out, I'm not giving the whole picture. Whatever speaker I get will be purely portable and will be stored in a closet or the garage (no extremes of heat, cold or humidity) when not in use. They need to be relatively light: 40-50 pounds is fine. I use a standard pole mount.

They could be used anywhere, so I just need a good general speaker. I do outdoor events, bars and coffe houses. The application ranges from a single singer/songwriter to large rock bands (where the PA is usually used for just the vocals).

This is very important: I would like to be able to run the speakers with just one amp stereo at 250W per side or with both at 850W per side. This allows me to limp along if anything ever goes wrong with one of the amps, and also allows me to just run one amp when I really don't need both. (I like to build as much redundancy as I can into my equipment selection to reduce the possibility of show-stopping technical failures. I often carry a small backup mixer and use good powered speakers as monitors that can become the mains in a pinch.)

And I can really spend as much as I want, but of course would like to keep it reasonable. $750 to $800 a speaker would be about as high as I'd ever want to go - and this is only because whatever I buy now will most likely last the rest of my life (or its life): I don't really plan on growing all that much.


So thanks again! And any further advice you might have would be greatly appreciated. I definitely have found the right place!

Cheers,

Rob

 

[waynehoskins]

Here's a brief explanation of what I think is behind the "underpowering" myth.

Underpowering itself doesn't damage speakers any more than driving your car below top speed damages your car. What kills drivers in this scenario is actually excessive power.

In the classic underpowering case, a person has an amplifier that's too small for his application. To get the desired average volume, he turns the console up, overdriving one or more stages of the amplifier, or possibly equipment in-between. This introduces clipping, and in the classic case it's excessive clipping. This introduces strong high-order (and probably odd-order, though that doesn't matter here) harmonic content, from the more nearly square-wave signal compared with the normal more-nearly complex-sine-wave signals we expect. It also, as he expects, increases the average power delivered to the speakers. The peak power is still no more than the "undersized" amplifier can deliver peak, but the average power is higher, and most importantly, the high-frequency power is much higher.

In most speaker systems, the high frequency drivers are rated for about 1/10 the power of the low-frequency drivers, so the "100-watt" speaker likely has a 10-watt compression driver. The "50-watt" amplifier is capable of delivering, in simple terms, 50 watts at any frequency. In normal applications this is perfectly sufficient, so long as clipping isn't introduced, so the frequency-power relationship stays where we expect it should. If a sufficiently strong high-frequency signal is applied to the amplifier, it is perfectly happy to amplify that to the 50-watt level, unaware that the high frequency driver in the speaker it's powering isn't capable of handling anywhere near that. This is, somewhat simplified, how clipping an "undersized" amplifier blows speakers.

In the classic case, by substituting a higher-power amplifier, the system is able to operate linearly, and usually within the power limitations of the drivers, at least the "program" rating (which is a time-weighted average based on some "normal" program material's dynamic and harmonic content). The desired volume is achieved without damaging the speakers.

"Underpowering" (relative to ratings) doesn't blow speakers, but heavy clipping in a system that's underpowered for its application does.

 

[museav]

"Underpowering" (relative to ratings) doesn't blow speakers, but heavy clipping in a system that's underpowered for its application does.

I think this is the crux, there is no such thing as an underpowered speaker, however you can have an underpowered system. Trying to get more from a system than it can provide can cause problems but if the system can provide the desired performance with a 1,000W rated speaker being powered by 1W then there is nothing wrong with that.

For just vocals, acoustic guitar, etc. you might look at something with a 12" woofer, maybe something like the JBL MRX512 that you could use as a main or as monitor for larger gigs. If you'll use it for full range music playback, keyboards and similar and won't be using subwoofers then you may want a 15", an 18" or two 15" woofers in order to get more low frequency response.

If you did not already purchase the second amp I would suggest looking at powered speakers as having everything in one package and properly matched including processing and limiting specific to the speaker can have advantages when you'll use them for a variety of uses.

 

[David Ashton]

Just by way of explanation to novices, an amplifier has a + and a - dc rail, and the transistors switch to create the output waveform, now as the amplifier is driven above it's rating and "clips" then the output contains more and more dc content.A speaker relies on the cone moving back and forth to cool the coil, but the dc just produces heat, not movement and this is why the coil burns out.
Apologies to the excellent explanations by Wayne and Brad but many students don't grasp exactly why clipped signals blow speakers.

 

[Meyer]

Eureka!!!!!!

If I could fax you all a beer, I'd do it, thanks all. Those explanations on "underpowering" have helped me out immeasurably". And that last post that specifically addresses the voice coil removes my last reservation, as somebody had specifically mentioned the voice coil as being the target of the damage to me once, and now I know what that's all about.

So running 850W per side puts me nicely in the range for the JBL MRX515s, as I'm hearing that its best to run at roughly twice the continuous number or roughly the program number, which turns out to be a nice match. And now it appears that it is safe to run these speakers off of one amp in stereo at 250W, so long as the lower wattage is appropriate to the application and I don't push it. (And yes, the second amp is already on the way.)

Interestingly enough, I learned that my friend's speakers that I've used in the past with just the one amp are actually the MRX515s and not the JRX115s as I had previously thought - meaning that I've already test driven them and like them a great deal.

So although I haven't made a firm decision, the MRX515s are currently the speaker to beat. Now if anybody would like to purchase my 1965 VW Variant (Squareback), I'm ready to start the next round of upgrades.



Cheers,

Rob

 

[museav]

Just by way of explanation to novices, an amplifier has a + and a - dc rail, and the transistors switch to create the output waveform, now as the amplifier is driven above it's rating and "clips" then the output contains more and more dc content.A speaker relies on the cone moving back and forth to cool the coil, but the dc just produces heat, not movement and this is why the coil burns out.

Good point. It really is a matter of energy. The energy a signal represents is the area under the waveform (the area between the waveform and 0, you always knew that learning about integrals would pay off some time). If the peak levels of a signal exceeds the maximum level possible for the electronics then the electronics cut off or 'clip' the peaks of the signal at that maximum level. This clipping results in having the same peak level but more area under the curve, thus the same level but more energy. Take a sine wave and severely clip it and it starts to approach a square wave, which contains much more energy than does a sine wave with the same peak level. So an increasing amount of clipping causes an output with the same peak level but with greater energy.

At the same time, where the signal is so clipped as to be flat represents a period where a speaker driven by that signal would not be moving, thus also reducing the cooling of the drivers. More energy with less cooling is not a good combination, especially for high frequency drivers that typically have lower power ratings and limited thermal mass and inertia.