Power handling tests, to be useful, are typically destructive. IEC 268.5 is one standard, there is also a DIN standard. They are filtered noise signals. These standards companies are very protective of their intellectual property. You might find a description of a filter to approximate them if you are very persistent.
Loudspeakers & Power Ratings: What's the Deal Part I? | Audioholics shows an old standard which looks like ~400Hz 6db/octave high and low pass on a pink noise signal
http://eaw.com/docs/6_Technical_Information/StudyHall_and_TechNotes/Power_Handling.pdf mentions AES signals.
http://www.doctorproaudio.com/doctor/temas/powerhandling.htm shows some curves.
Loudspeakers & Power Ratings: What's the Deal Part I? | Audioholics shows an old standard which looks like ~400Hz 6db/octave high and low pass on a pink noise signal
http://eaw.com/docs/6_Technical_Information/StudyHall_and_TechNotes/Power_Handling.pdf mentions AES signals.
http://www.doctorproaudio.com/doctor/temas/powerhandling.htm shows some curves.
Last edited:
I managed to make a signal through LTSpice with the method describeb. Full-range for now.
I wanted to calculate if there was any difference between that and a sinewave when it comes to power dissipation. And through a resistor there is none what I can see. Peak to average differs as expected.
Still not sure if a driver would survive its rated power with s sinewave, so Ill probably keep halving for sinewave ratings in the future as well.
I wanted to calculate if there was any difference between that and a sinewave when it comes to power dissipation. And through a resistor there is none what I can see. Peak to average differs as expected.
Still not sure if a driver would survive its rated power with s sinewave, so Ill probably keep halving for sinewave ratings in the future as well.
I'm not an expert at power handling specs, but I suspect bass speakers may handle close to rated power with sine waves from some frequency above where they are mechanically damaged to some frequency where there is too little cone motion to effectively cool the coil.
Tweeters are specified much differently from woofers as far as I know. Usually the ratings I see are "system ratings" with a specified crossover. I think KEF used to give sine wave power specs as a voltage for their T series tweeters and they were very low - like 6-8 volts RMS.
I'm not really sure what good power handling specs are to anyone anyway, to be honest - you can't really predict how a person may abuse a speaker.
Last edited:
For subwoofers there is the CEA-2010 test, there's an app note on that for Clio here: http://www.audiomatica.com/wp/wp-content/uploads/2012/03/appnote_007.pdf
The REW V5.01 beta signal generator can generate the CEA-2010 burst, to measure results use the RTA with FFT Length 65536 for 44.1k or 48k, 131072 for 88.2k or 96k, Rectangular window.
The REW V5.01 beta signal generator can generate the CEA-2010 burst, to measure results use the RTA with FFT Length 65536 for 44.1k or 48k, 131072 for 88.2k or 96k, Rectangular window.
FWIW I make Guitar speakers, one audio field where amplifiers are *guaranteed* to be overdriven 🙁 and squarewaves are thought of as "cool". 😱
So my only way to *guarantee* survival is to measure using the worst of the worst case: first measure minimum impedance (usually between 250 and 350 Hz) and then apply a squarewave within that range, 10 minutes at 5 Watts and then raising in 5 watt steps until speaker self destroys.
Results are humbling.
Speakers which could be rated "250W program" or "continuous Music power" (which are NOT continuous sinewaves ) die at, say, 60W .
Oh well.
So my only way to *guarantee* survival is to measure using the worst of the worst case: first measure minimum impedance (usually between 250 and 350 Hz) and then apply a squarewave within that range, 10 minutes at 5 Watts and then raising in 5 watt steps until speaker self destroys.
Results are humbling.
Speakers which could be rated "250W program" or "continuous Music power" (which are NOT continuous sinewaves ) die at, say, 60W .
Oh well.
Yes there is a standard. The test signal is pink noise limited to a 10db crest factor. This may be used with filters to limit the signal to the required bandwidth. The level is increased until there is 3 dB power compresion. That should be the power rating.
You use a pink noise source with diode clippers is the old method. Set the RMS value to .316 of the peak value.
- Status
- Not open for further replies.