Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Scott Holtzman]

And do that using only 50% of the mains power the "big lungs" require to move the same amount of air, while weighing in at less than 1/3 the weight!

And as usual Art for the win.  I thought we were going to have a car fight for a second, was getting my safety glasses and mask out before the fur started flying <g>

[Jack Hawkins]

Great i'm starting to understand amplifiers a bit now, now am I right in saying the reason amplifiers such as the Behringer NX6000 cannot output their rated power continuously (not burst I mean continuously) is because they A, don't have a big enough power supply and B, because they don't have the thermal capability to be able to keep the amplifier cool enough? I know the old Crest 9001 and 10001 amps used to have overbuilt power supplies so they could provide their rated power output continuously all night long no matter how harsh the low frequency requirements were due to the music having obcene amounts of low frequency energy. I know they'd go all night at their rated power if need be without breaking a sweat. Is that all in all because they had massive power supplies, very big power capacitors and enough output devices? I've heard from people before that to run one of those amps on bass heavy music it ideally needs a 32amp supply and if run on a standard British 13A plug socket could easily be run into clipping because there wasn't enough power from the 13A socket for what it needed.

[Chris Grimshaw]

The reason they can't do continuous sine waves all night is because they're not designed to do continuous sine waves all night. It's that simple.

A full-power sine wave is:
1 - Extremely rare in music. Even drum-n-bass.
2 - Abusive for the speakers connected

So, amplifier designers don't bother to optimise for that. Why should they?

It would be much more useful to have an amp that'll deliver 1000w continuously and 5000w short-term. That way, your short-term peak (kick drum) remains unclipped, and the bass line also holds up fine. If you had an old-fashioned sine-waves-all-night amp capable of 2000w, it'll be clipping the kick drum and still have capacity to spare for the bass line.

I think it's important to note that the old iron amps you're comparing to were top-of-the-line in their day, and this Behringer amp certainly is not. It's exceptional value for money, but it's still a lower-budget piece of equipment.

FWIW, the fuse fitted in every 13A plug will pass 30A for a second. You can run a surprising amount of gear on a 13A supply*, although I'd suggest that if the power levels are going to be high, the 13A plug should have a 32A socket on a short cable, and then run 32A cable to where the power needs to be. That way, you've minimised cable losses. A long 13A extension lead will happily drop 10s of volts under heavy load - believe me, I've watched the meters.
Those old iron amps were usually fairly low-efficiency (think about all the heat coming out), so if they're putting 2000w into the speakers, they might be pulling 4000w from the wall and pushing 2000w out the vents. That's the other reason that large mains supplies were generally recommended.

* I took a Powersoft T602, bridged into a 4ohm 2x18" sub, 6dB into the limits playing Truth - Devil's Hands on a single 13A supply. Held up just fine. Even though the bass gets quite intense at times, it's still dynamic enough that the mains draw (reported by the amp) was well within the limits.

I think it's worth your time familiarising yourself with concepts like "duty cycle" and "crest factor", and thinking about how those relate to speakers and amplifiers.

Chris

[Jack Hawkins]

The reason they can't do continuous sine waves all night is because they're not designed to do continuous sine waves all night. It's that simple.

A full-power sine wave is:
1 - Extremely rare in music. Even drum-n-bass.
2 - Abusive for the speakers connected

So, amplifier designers don't bother to optimise for that. Why should they?

It would be much more useful to have an amp that'll deliver 1000w continuously and 5000w short-term. That way, your short-term peak (kick drum) remains unclipped, and the bass line also holds up fine. If you had an old-fashioned sine-waves-all-night amp capable of 2000w, it'll be clipping the kick drum and still have capacity to spare for the bass line.

I think it's important to note that the old iron amps you're comparing to were top-of-the-line in their day, and this Behringer amp certainly is not. It's exceptional value for money, but it's still a lower-budget piece of equipment.

FWIW, the fuse fitted in every 13A plug will pass 30A for a second. You can run a surprising amount of gear on a 13A supply*, although I'd suggest that if the power levels are going to be high, the 13A plug should have a 32A socket on a short cable, and then run 32A cable to where the power needs to be. That way, you've minimised cable losses. A long 13A extension lead will happily drop 10s of volts under heavy load - believe me, I've watched the meters.
Those old iron amps were usually fairly low-efficiency (think about all the heat coming out), so if they're putting 2000w into the speakers, they might be pulling 4000w from the wall and pushing 2000w out the vents. That's the other reason that large mains supplies were generally recommended.

* I took a Powersoft T602, bridged into a 4ohm 2x18" sub, 6dB into the limits playing Truth - Devil's Hands on a single 13A supply. Held up just fine. Even though the bass gets quite intense at times, it's still dynamic enough that the mains draw (reported by the amp) was well within the limits.

I think it's worth your time familiarising yourself with concepts like "duty cycle" and "crest factor", and thinking about how those relate to speakers and amplifiers.

Chris

I guess although sinewaves are used for sub bass notes in electronic dance music they're at a low level and not full level. When I mentioned the old Crest amps I wasn't even remotely comparing them with the Behringer. When you say use 32A cable do you mean connect from a double wall socket (for each amp) like in this picture?

[Chris Grimshaw]

Yes. If you need to run power over long distances, using the thicker 32A (or 63A, but you don't see that as often) cable will minimise the voltage drop along that cable. An extreme example, but I remember running a PA + backline + lights on around 150m of 13A extension leads (actually, some of them might've been 10A - I had 75m of 13A on-hand, don't know what else was acquired). The generator (not mine) had gone down. Anyway, with the PA at idle (amps on, but no signal), I was getting around 220v. Push the fader, and I think I saw 180v once. I was being gentle, too.

A thicker cable would've helped there. NB - it wouldn't have given me a rock-steady 240v. The laws of Physics still apply. It might've got me 230v@idle and 210v under load, though, which would've been a much happier time for the amplifiers.

It's not a replacement for a big power supply, but sometimes you've only got a 13A supply to work with, and that's how to make the most of it.

Oh, and a much better way to make the most of it is to use efficient amplifiers 

Chris

[Jack Hawkins]

I have no idea what the efficiency of the RMX4050HD is but say it's 80% efficient that means i'm losing 20% of the power through heat coming through the fans right? So taking into account my subs are rated at 1,600 watts continuous that would mean i'd ideally need an amp 20% more powerful (20% higher power per channel continuously) than my subs are rated JUST to in theory get the same continuous power that my subs are rated at. Is that correct. So a lot of people with these supposedly massively powerful amplifiers could be losing 10 maybe 20% of that power.

[Art Welter]

I have no idea what the efficiency of the RMX4050HD is but say it's 80% efficient that means i'm losing 20% of the power through heat coming through the fans right?

Jack,

The RMX4050HD is only 50.7% efficient, that is to say when it produces 4000 watts sine wave at the speaker outputs, it draws 7884 watts from the mains (65.7 amperes @120vac). 49% of the power is wasted as heat.
This is pretty typical for that type of amplifier, Crest, Crown, and QSC will all have similar thermal efficiency.

The Behringer  (or other class D switch-mode PSU) amps are over 90% efficient, less than 10% is wasted in heat. When doing sine wave tests on Crest and QSC, the fan outputs were like hair driers, while at the same output levels, SpeakerPower and Behringer amps put out hardly any heat.
Since they draw far less power from the mains, they don't need as heavy cable for the same voltage drop- you will save weight and cost in that department as well.

Art

[Luke Geis]

Your subs can handle 1,600 watts continuously. The continuous rating is the wattage at which you can operate the subs essentially forever ( until mechanical breakdown ). If you have less wattage than that, the subs just don't move as much air. You can mathematically calculate what your likely output will be by using the specs. If your subs have a sensitivity of 100db @ 1watt, 1meter, then if you have only a 1-watt amplifier you will achieve 100db in output. If you have a 2-watt amplifier you will achieve 103db in output. A 4-watt amp will net you 106db in output. Every doubling of power nets you another +3db in output. You can see how you get to a point of diminishing returns quickly though. By the time you get to 500 watts, you're doubling things up really quickly as you go to 1000, 2000, and then 4000 watts respectively to acquire each +3db gain.

There is no minimum sized amp you can use, it is simply that you will not get the output you want or have the headroom you need to make it work as loud or as safe as you desire. As mentioned, when you hard clip an amplifier it can output much more power than the numbers on the box state. Where the whole too small of an amp thing came from is because people simply cut corners and pushed things to the max. A speaker coil is nothing more than a light bulb element really. If you can imagine running the light bulb at half the voltage, it will be dim and because there is less heat, it should last forever. Conversely, if you power that 120 volt light bulb with a 240-volt outlet, the bulb will flash brightly for a second and then burn out. The speaker coil takes the wattage you give it and converts only a small percentage of that power into acoustic output, the rest is converted into heat. If you provide an excessive amount of power long enough, the coil will get really hot and start to burn/cook. Eventually it will burn up until it delaminates from its former or physically burns the wire open, in which either way the speaker ceases to work.

So how does having too small of an amp cook a speaker? It is more about the abuse. If you have an amp that is banging away to the point of clipping, the first thing most people try to do is make the red light go away without turning anything down. Enter your basic limiter. It chops off the peaks and the red lights diminish. So you turn it up some more until the red lights come back and let it set there. The limiter does reduce the peaks, but it increases the average power. Let's look at crest factor and how it affects amplification. Crest factor is the ratio between the media's peak amplitude and its RMS value. So a lower crest factor means a much higher amount of energy. A crest factor of 0 means that you have a signal in which the peak is the same as its average rms signal. A common crest factor that you can relate to would be commercial media on a cd which has a typical CF of around -5. This is extremely compressed and hot. You have to imagine that the peak signal is only 5db higher than the average level of the media. For comparison, a live band with little or no compression has a CF of between -15 and -20. So what does this have to do with limiters? As you compress media, you increase its RMS power. If you turn the media back up to the point where you're clipping, you have also then reduced the crest factor. If you compress the peaks 3db and bring the level back up 3db, you have doubled your RMS power, as you can imagine, compressing stuff as much as 10db is not unheard of and then if you use a master bus compressor you can also squeeze another couple dB out reducing the CF to -10 or as much as -8 if you are persistent. Then as we clip the amp we introduce a limiter which further increases that potential.

As you clip the amp the little red light generally means that you have achieved audible distortion. So when it flickers, it is not such a huge deal, but as it goes solid red on each kick hit for instance, it is physically chopping the top off the signals. This increases the output power significantly as was mentioned earlier that it too increases the RMS output. When the amp is clipped, the speaker is still held in that position for the length of time it is asked before it moves back the other way and repeats the cycle. So as you hard clip at around +6db you have now quadrupled the RMS power potential. The peak signal stays put, but the rest of the media now comes up. So now the majority of the signals cycle is spent at 100% positive or negative and only swings briefly back through zero. This reduces the speaker coils actual movement as it is just being slammed in or out and held there for longer periods of time. This lack of motion reduces the ability to cool the coil as now less air is moving through the gap ( the dust cap is actually used to help move air through the coil gap acting as a lung or diaphragm ). This added heat also reduces the coils ability to convert the incoming power into motion. This is known as power compression. As heat rises, it increases the resistance of the coil, the increase in resistance/impedance reduces the conversion of the amplifier power into acoustic output. Power compression can reduce acoustic output by -6 to as much as -10db in extreme cases. The typical answer to power compression and the subsequent loss of output, is to turn it up more which only further exacerbates the problem. So pretty soon you have an underpowered amp that is simply producing nothing but pure heat due to all the clipping going on, and the extreme efforts of the engineer to maintain a desired acoustic output by using limiting, compression and everything else. Eventually, the speaker's coil just burns up from all the stress. It may not have been overpowered, but all the power that was there was converted into heat instead of coil motion.

Too large of an amp does this sort of destruction in a different way. Typically it just boils down to too much power for too long. But it can also kill the speaker from too much instantaneous power causing the coil to jump the gap, or bottom out, which causes the coil to distort and move in the gap and potentially rub the sides. Rub too much and the coils lacquer coating will rub off and then the coil will short against the gap walls. It will do this until poof, the coil splits, and no more worky. Generally having a larger amp is safer only because if ran under sane levels, there is no clipping, and the speakers coil is always moving as the incoming signal is actually asking it to. Because a larger amp typically only buys you headroom, you can reduce compression and limiting, thereby reducing the RMS power and CF. The power compression still occurs, but it takes longer and is reduced because the speaker is at least able to attempt to stay cool by having full unclipped motion. With an amp that is too large, the thing that generally pops drivers are mistakes like turning things up too loud for too long, having a mic drop that simply causes the speaker to literally try and punch through the speaker grill, or just gross negligence where you turn it up to eleven and the speakers coil just fries from all the power that it was not designed to deal with.

Speakers convert wattage into output, period. The sensitivity is what determines how loud it can get given X wattage. The peak wattage CANNOT ever be exceeded and the program rating is the ideal wattage that you should use to power it with in most circumstances. The continuous rating is just letting you know the wattage at which the speaker is happy to run at forever. You can go as low as you'd like, it is essentially like using it with the volume turned down. The only real rule is to not clip the amp. As long as you are not clipping the amp, it is pretty hard to blow a speaker up.

[Keith Broughton]

Before I launch into reading Luke's latest novel ...er post , I'm curious about this statement...

With regards to the piezo and a 400hz sinewave tone you should know how to use a piezo for gain structure.

Enlighten me.

[Chris Grimshaw]

I have no idea what the efficiency of the RMX4050HD is but say it's 80% efficient that means i'm losing 20% of the power through heat coming through the fans right? So taking into account my subs are rated at 1,600 watts continuous that would mean i'd ideally need an amp 20% more powerful (20% higher power per channel continuously) than my subs are rated JUST to in theory get the same continuous power that my subs are rated at. Is that correct. So a lot of people with these supposedly massively powerful amplifiers could be losing 10 maybe 20% of that power.

Read Art's post.

The efficiency determines how much power needs to go in (mains) to produce enough for the speakers. A 100% efficient amp will have power in = power out.

50% efficiency means 2w in = 1w out.

Chris

[ProSoundWeb Community]