|
Amplifier power rating vs. frequency |
Post Reply 
|
Page 12> |
| Author | |
Carl Klinkenborg 
Registered User 
Joined: 26 May 2020 Location: UK Status: Offline Points: 16 |
 Post Options
 Thanks(0)
 Quote Reply
 Topic: Amplifier power rating vs. frequencyPosted: 26 May 2020 at 10:01am |
|
Hi All,
Newbie here; apologies if this in the wrong forum - perhaps mods could move if so! I am considering purchasing a 4-channel Class D amp from a reputable company who have performed their own testing and advertise it with real-world power ratings under varying conditions such as EIAJ burst and continuous RMS, but it is not recommended for sub use, when overheating would cause shutdown. Stated simply, what makes an amplifier unable to output equal power into the same load (let's assume a purely resistive load for now) at low frequencies compared with higher frequencies, when mathematically the power in a sine wave remains independent of period? Cheers, Carl.
|
|
 |
|
|
Earplug
Old Croc 
Joined: 03 January 2012 Location: Europe Status: Offline Points: 7216 |
Post Options
Thanks(0)
Quote Reply
Posted: 26 May 2020 at 10:46am |
|
The difference is a prime example of the difference between theory and practice! And why you need to be very careful about the specs given by some manufacturers.  To keep it short, I´d say that the lower frequencies are (usually) served by larger drivers, with larger coils and so need more current drive to move. As simple as that. |
|
|
Earplugs Are For Wimps!
|
|
|
|
|
studio45
Old Croc
Joined: 16 October 2007 Location: United Kingdom Status: Offline Points: 3864 |
Post Options
Thanks(0)
Quote Reply
Posted: 26 May 2020 at 1:53pm |
|
I forget the exact figures but something like 75% of the electrical energy required to reproduce popular music (the kind the kids are into these days) with a moving coil speaker, is used to produce frequencies under 200Hz.
A different way of saying that is that the duty cycle (time at full power vs time at zero power) for an amp powering bass/sub, is much higher than for an amp just powering mids or tweeters. So for these class D amps where the power supply is a bit too small, and they're relying on charged capacitors to see you through the power peaks - you can have trouble playing bass/sub, as there's not enough time between peaks for the capacitors to recharge. The power supply can overheat. |
|
|
Studio45 - Repairs & Building Commotion Soundsystem -Mobile PA
|
|
|
|
|
Conanski
Old Croc
Joined: 26 January 2006 Location: Ottawa, Canada Status: Offline Points: 2543 |
Post Options
Thanks(0)
Quote Reply
Posted: 26 May 2020 at 7:47pm |
|
A linear power supply recharges the caps at line frequency which is 50 or 60hz, a switching power supply operates at hundreds of kilohertz and therefore has no problem keeping the caps charged... so that isn't the problem.
|
|
|
|
|
studio45
Old Croc
Joined: 16 October 2007 Location: United Kingdom Status: Offline Points: 3864 |
Post Options
Thanks(0)
Quote Reply
Posted: 27 May 2020 at 2:18pm |
|
Depends on how fast the supply can recharge those caps though! If the MOSFET is a cheap one with significant RDSon....and the transformer is a bit cheap with thin wire....and maybe the feedback circuit is a bit too simple, slow to react.....it won't matter if there is a LEGION of reservoir caps after that, if you're asking too much of it they won't get charged back to full in 1/1000th of a second, and the rails will droop! Then the transformer will heat up as it's trying to pass too much current...then the smoke comes out.
Same problem can happen with linear supplies too, when the transformer isn't quite big enough. EG all those Maplin/Prosound/Soundlab "2000 watt" amps that can be picked up with one finger. You ask them for 2000 watts continuous for longer than 5 minutes, and that transformer will just evaporate. Sure it can do the voltage, but not the current. This theoretical situation is for a bass-only amp, being fed a low-passed signal. The duty cycles can get quite high in that case. I mean you could get it to happen playing full-range too, probably, but at that point you'd be doing several things wrong, like turning it up far too much and not really having enough rig for the gig. |
|
|
Studio45 - Repairs & Building Commotion Soundsystem -Mobile PA
|
|
|
|
|
Earplug
Old Croc
Joined: 03 January 2012 Location: Europe Status: Offline Points: 7216 |
Post Options
Thanks(0)
Quote Reply
Posted: 27 May 2020 at 3:59pm |
|
"
I forget the exact figures but something like 75% of the electrical
energy required to reproduce popular music (the kind the kids are into
these days) with a moving coil speaker, is used to produce frequencies
under 200Hz. " Yes, definitely. Likewise the use of underrated/cheapo components for the supply - or just not enough o/p devices to supply the current. I had some idiots here the other day wondering why their '1500W' Skytec amp couldn´t properly power 4 x 8" mid-tops. I opened the thing up, and it only had 4 o/p transistors per side. More like 150W than 1500. |
|
|
Earplugs Are For Wimps!
|
|
|
|
|
Jo bg
Young Croc 
Joined: 08 March 2017 Status: Offline Points: 552 |
Post Options
Thanks(0)
Quote Reply
Posted: 28 May 2020 at 11:12am |
|
Other things to look at, every time you go down an octave, you need four time the driver displacement to keep same level, so more power to get more excursion (or to drive more drivers at same excursion, but with more diplacement given by Sd increase).
also, as I think Martin papers demonstrated, music usually has higher crest factor in the highs and lower in the lows, so more average power needed. so you need more voltage swing for highs (which are usually more efficet devices than bass drivers) and more current for lows; usually current limits first, in particular with cheap class d multichannel amps. |
|
|
|
|
Carl Klinkenborg
Registered User
Joined: 26 May 2020 Location: UK Status: Offline Points: 16 |
Post Options
Thanks(0)
Quote Reply
Posted: 28 May 2020 at 12:02pm |
|
Thanks for all the replies, however I think there may be some misunderstanding of my original post so I will restate the question.
Firstly, forget loudspeakers completely - assume that the amplifier is driving a simple resistive load. Secondly, forget power supply capacity - assume it is just capable of driving the amp at 1kW output into the dummy load continuously at 1kHz without it or the amplifier stage overheating. Thirdly, forget about music signals - I am referring here to pure sine waves. In our lab test scenario the only thing that changes is the frequency - the load resistance, amplifier output voltage, and supply voltage remain constant. Why is the amplifier's continuous output capacity reduced if we drive the same load at 20Hz, when the energy contained in both output sine waves, and therefore the heating effect in the load, is identical? Even the suppliers who tested the amp have no answer, so top marks to the first with an explanation!
|
|
|
|
|
Earplug
Old Croc
Joined: 03 January 2012 Location: Europe Status: Offline Points: 7216 |
Post Options
Thanks(0)
Quote Reply
Posted: 28 May 2020 at 2:16pm |
|
I think that you´ve already got your answer! Why do you not think so? To sum up:- 1) You can´t 'ignore' the speaker. It is part of the circuit. Just like you can´t ignore the cabinet in which you place the speaker. It´s a complete system. Everything influences the other. And definitely not the same as driving a simple resistive load. It´s quite likely that you amp might be able to drive a 'simple resistive load', but what good is that? Your manufacturer is obviously aware of the limitations of their product and so print it on the box! 2) Under rated components/engineering and cost cutting. Not all '1kW amps' are born equal! |
|
|
Earplugs Are For Wimps!
|
|
|
|
|
Jo bg
Young Croc
Joined: 08 March 2017 Status: Offline Points: 552 |
Post Options
Thanks(0)
Quote Reply
Posted: 28 May 2020 at 5:13pm |
I am thinking out loud here, but think about pink noise and white noise. Equal energy per cycle gives you white noise rising response on log scale (and flat power spectrum), to get flat equal level you need to halve the energy every octave you go up (that's pink noise). I you look at pink falling noise power spectrum you get why to produce that 20hz sine you need far more power than to produce 1khz at same level. what you think guys? just a brainfart? |
|
|
|
|
Carl Klinkenborg
Registered User
Joined: 26 May 2020 Location: UK Status: Offline Points: 16 |
Post Options
Thanks(0)
Quote Reply
Posted: 28 May 2020 at 10:37pm |
|
Hi jo,
Thank you so much for your well-reasoned reply; certainly food for thought. he ratings I have seen seem to be related to pure sine waves, when a lower frequency is related to a lower continuous power output. The pink/white noise reference certainly makes sense from a 'music' point of view though. Please think again about the sine wave ratings I have seen quoted. Thank you and stay safe in these wierd times, Carl. |
|
|
|
|
Fred_dibna
Registered User
Joined: 24 September 2018 Status: Offline Points: 213 |
Post Options
Thanks(0)
Quote Reply
Posted: 30 May 2020 at 10:21am |
Could it not be that the amplifier power supply isn't big enough so cannot drag enough power from the mains supply? We all know frequencies below 100hz need far more power than midrange or high frequencies. Obviously because like has already been mentioned that the low frequency cones are much bigger and voice coils too. Also it might be because the amp has inadequate cooling to keep cool so many amps can go into thermal shutdown if you ask it to do too much in terms of low impedance loads of run them too hard with bass heavy music material.
|
|
|
|
|
Post Reply
|
Page 12> |
| Tweet |
| Forum Jump | Forum Permissions You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You cannot vote in polls in this forum |
Topic Options
Carl Klinkenborg wrote: