Hi.
If one should measure an amplifier using pink noise, would the output wattage be higher than one measured using a sine wave?
Many professional audio amplifiers are not using sine waves to market their output power any longer. Instead, they choose pink noise.
As far as the manufacture states, it is a better representation of music than sine waves. I don’t agree on such nonsense. Nevertheless, how would an amplifier stand measured with sine waves under pink noise conditions?
Pink noise seems far less demanding than sine waves, which would more than likely mean the power supply can deliver more power due to an easier duty cycle.
What are your thoughts on the matter?
Cheers!
If one should measure an amplifier using pink noise, would the output wattage be higher than one measured using a sine wave?
Many professional audio amplifiers are not using sine waves to market their output power any longer. Instead, they choose pink noise.
As far as the manufacture states, it is a better representation of music than sine waves. I don’t agree on such nonsense. Nevertheless, how would an amplifier stand measured with sine waves under pink noise conditions?
Pink noise seems far less demanding than sine waves, which would more than likely mean the power supply can deliver more power due to an easier duty cycle.
What are your thoughts on the matter?
Cheers!
You could claim higher power when testing with pink noise because the supply is only partially loaded and you get full peak voltage. The actual effective power with a PN signal is less - because of a crest factor of around 10-12 dB. Amps are *supposed* to be power tested with sine wave, all channels driven, but I'm sure not all of them are. A valid sine wave power test need only be run for a few seconds - long enough for the power supply to stabilize at its fully loaded voltage, but not long enough for the mains breaker to trip. A lot of big pro amps can draw 30, 50, or even 80 amps with a sine wave for some length of time, and you can't legally do that off a standard wall outlet and they needed some way of solving this dilemma. Pink noise, with its lower effective duty factor is supposed to be used for heat dissipation and mains draw testing as it is "more representative of real world use." But that can vary substantially depending on application and how much clipping the user is willing to tolerate.
For my own applications, I'll only run an amp down to the lowest Z that carries an FTC rating (which, nowadays has the burn-in done at 1/8 power, PN compared to 1/3 power sine in years past). With most amps that's 8 ohms. Some will take 4R, and a handful will go to 2R (RMX5050 being the only one I can afford). When I build them, I power test with 20 Hz sine, and design to tolerate 1/8 power PN at 2R and 1/3 power PN at 4R indefinitely without taking components beyond their ratings (except for the mains plug which tops out at 20A) That's a tall order for a store-bought amp.
For my own applications, I'll only run an amp down to the lowest Z that carries an FTC rating (which, nowadays has the burn-in done at 1/8 power, PN compared to 1/3 power sine in years past). With most amps that's 8 ohms. Some will take 4R, and a handful will go to 2R (RMX5050 being the only one I can afford). When I build them, I power test with 20 Hz sine, and design to tolerate 1/8 power PN at 2R and 1/3 power PN at 4R indefinitely without taking components beyond their ratings (except for the mains plug which tops out at 20A) That's a tall order for a store-bought amp.
even with FTC guidelines in place, a lot of manufacturers still have some kind of "fudge factor" in place in their testing methods. i got a system in for repair recently that said "1000W RMS CONTINUOUS POWER" in big huge letters all over the shipping box. funny thing was, the line fuse was a 5 amp fuse. the other 400W must be sucked out of the ether somehow? even though this receiver has class D amps, there are absolutely NO amplifiers on the planet that run at 160% efficiency.... maybe the manufacturer has found a way to circumvent the laws of thermodynamics. this has to be the best kept "free energy" secret of all time!!!!! i'l hook up a couple of them to run each other's line voltage supply off of the speaker outputs, and never have to pay a dime for electricity ever again......
here in the US, these specs are supposed to follow some basic rules. one of them being that output power is to be measured as RMS power at clipping all channels driven into whatever load is specified in the specs. but what appears to happen with these "low-fi" and "mid-fi" systems is that they measure one channel at clipping into a load, and then multiply by the number of channels. the particular system in question required a new power supply, which is listed in the PARTS LIST as a 500W supply. being that it's a SMPS, not a linear supply, that pretty much assures that the 500W rating is pretty much a brick wall. any more than 500 watts and the power supply goes into shutdown mode, where a linear supply would just sag a bit. a SMPS also doesn't have any huge energy storage mechanism like a linear supply has. once you get to the current limit, that's it.
OMNIFEX
I would suggest you download ( pdf ) manuals from the various amp manufacturers, especially from Crown.
You can compare the various techniques and "industry standards" that are used to generate wattage numbers.
Pay particular attention to the power consumption measurements and how they are Qualified.
Because Big wattage numbers sell amps, it is almost a standard practice to run tests to produce as large a number as possible, for the consumer's eye.
It most cases: the conditions under which those numbers were obtained are not stated as they do not reflect actual operating conditions.
There are a lot of ways to get exaggerated power figures.
I would suggest you download ( pdf ) manuals from the various amp manufacturers, especially from Crown.
You can compare the various techniques and "industry standards" that are used to generate wattage numbers.
Pay particular attention to the power consumption measurements and how they are Qualified.
Because Big wattage numbers sell amps, it is almost a standard practice to run tests to produce as large a number as possible, for the consumer's eye.
It most cases: the conditions under which those numbers were obtained are not stated as they do not reflect actual operating conditions.
There are a lot of ways to get exaggerated power figures.
unclejed613 said:
The FTC guidelines are more lax than they used to be. And they don't enforce it anymore - it's optional for a product to carry an FTC rating. Some do, most don't. The lower-tier Crowns and QSCs just claim an EIA power rating. And with the EIA standards, you can run the distortion pretty much as high as you want for the test, as long as it's specified. Most of the time, it's way past clipping.
The power consumption numbers listed for sine, 1/8 power PN and 1/3 power PN give you an idea of how much power an amp can really put out. Many times this is way higher than the maximum 16A allowed from a standard 120V/20A circuit. It better be for an amp claiming to put out 2kW. The RMX5050 claims it draws 87 amps (!!!) at 2R sine wave.
i recently had a sony system come through the shop that was a "component" system that looks like a very large boom box. it had "bi-amped" speakers' it had a power rating of "500W/ channel * ".
if you look below the picture on the box at the small print, you saw what the asterisk was indicating.... " * at 10%THD"
500W.... yeh right.... with the usual 22guage speaker wires..... funny
even funnier, it had 2 50W/ch stereo amp chips in it 1 for the woofers, 1 for the "twids".
if you look below the picture on the box at the small print, you saw what the asterisk was indicating.... " * at 10%THD"
500W.... yeh right.... with the usual 22guage speaker wires..... funny
even funnier, it had 2 50W/ch stereo amp chips in it 1 for the woofers, 1 for the "twids".
Thank you Gentlemen.
I downloaded a few schematics from Crest, Crown, & QSC. If I wanted to estimate the Sinusoidal output, how would I go about doing so?
Should I follow the volts times volts divided by the impedance and, subtract the figure by 58%?
Peavey suggest doing so with their old CS models.
Cheers!
I downloaded a few schematics from Crest, Crown, & QSC. If I wanted to estimate the Sinusoidal output, how would I go about doing so?
Should I follow the volts times volts divided by the impedance and, subtract the figure by 58%?
Peavey suggest doing so with their old CS models.
Cheers!
if you want W/ch, drive both channels full power into whatever load (8r for instance). the formula is E^2/R so if you get 28.3V rms, it's (28.3*28.3)/8 which should give 100.111 watts. total power would then be 200.222 watts, since you're driving both channels to full power. i have no isea where Peavey came up with the 58% figure, unless it was a calculation for heat dissipation or something, or maybe it's a correction factor for watts per channel if you test the power with only one channel driven to full power,
PV probbaly came up with the 58% figure from efficiency - the CS800 should be about 58% efficient at full power. If it's putting out 400 wpc, it will draw 1379 watts. Leaving a power dissipation of 579 watts. It will draw more than 1379 volt-amps because of the non-sinusiodal AC current draw so you can't use that to calculate thermal dissipation.
Measuring output power should be straightforward in principle. Volts (RMS) squared divided by ohms. The trouble is twofold. First, most modern amps will throw a fit and shut down when asked to dissipate 579 watts of heat. Compare the heat sinks in that old boat anchor vs. its newer bretheren, vs. a "comparable" QSC - RMX1450. No comparison. That figure grows substantially with amps "bigger" than a CS800. Second, 1379 watts can be 2000 or more volt-amps. The transformers can probably take this for an hour or more, but the mains breaker will trip in a few seconds. The test needs to be run quickly enough that the thermals aren't excessive, and the average current draw well under 20A. "Burst testing" is sometimes done - but if the burst is short enough you end up with inflated power numbers. 10 or 50 milliseconds won't drain the filter caps. One second on full sine, 5 seconds off is a good compromise. And a good amp should be able to take that for what would be the duration of a gig.
Measuring output power should be straightforward in principle. Volts (RMS) squared divided by ohms. The trouble is twofold. First, most modern amps will throw a fit and shut down when asked to dissipate 579 watts of heat. Compare the heat sinks in that old boat anchor vs. its newer bretheren, vs. a "comparable" QSC - RMX1450. No comparison. That figure grows substantially with amps "bigger" than a CS800. Second, 1379 watts can be 2000 or more volt-amps. The transformers can probably take this for an hour or more, but the mains breaker will trip in a few seconds. The test needs to be run quickly enough that the thermals aren't excessive, and the average current draw well under 20A. "Burst testing" is sometimes done - but if the burst is short enough you end up with inflated power numbers. 10 or 50 milliseconds won't drain the filter caps. One second on full sine, 5 seconds off is a good compromise. And a good amp should be able to take that for what would be the duration of a gig.
My confusion lies with the Crown PSA 2 & Macro-tech 2400 amplifier.
The PSA 2 offers 75-volt rails, two Power Transformers each, having its own 20-amp fuse.
The MA 2400 offers 57-volt rails, one power Transformer for each channel in addition to a 15-amp circuit breaker for each transformer.
The PSA 2 is 275 watts per channel @ 8 ohms, 460 watts in a 4-ohm load.
The MA 2400 is 515 watts per channel @ 8 ohms, 710 watts in a 4-ohm load.
When I calculate use the Volts times Volts divided by 4 ohms, I get a sum of 812.25 watts for the MA 2400.
When I calculate the PSA 2, I get 1406. 25 watts.
Can anyone explain what is happening here?
Cheers!
The PSA 2 offers 75-volt rails, two Power Transformers each, having its own 20-amp fuse.
The MA 2400 offers 57-volt rails, one power Transformer for each channel in addition to a 15-amp circuit breaker for each transformer.
The PSA 2 is 275 watts per channel @ 8 ohms, 460 watts in a 4-ohm load.
The MA 2400 is 515 watts per channel @ 8 ohms, 710 watts in a 4-ohm load.
When I calculate use the Volts times Volts divided by 4 ohms, I get a sum of 812.25 watts for the MA 2400.
When I calculate the PSA 2, I get 1406. 25 watts.
Can anyone explain what is happening here?
Cheers!
First off, the MA2400 is a bridge configuration with floating supplies for each channel. The effective rail voltage is +/-114 using a normal topology.
And it isn't anything special like class G or H to reduce current consumption or dissipation - it is a hog.
Now in order to get stated performance for the 2400:
Unloaded supply = +/-114V
Assume 75% regulation: Vcc = 85.5V
Two sets of saturated transistor banks @ 5V each: Vpeak = 75.5V
Output voltage = 53.4V RMS, 712W/4R.
At 65% efficiency and 65% power factor the AC current draw would be 3370VA.
Now to get stated performance for the PSA2:
Unloaded supply = +/-75V
Assume 88% regulation: Vcc = 66V
One set of output transistors in saturation: Vpeak = 61V
Output voltage 43.1V RMS, 465W/4R.
At 65% efficiency and 65% power factor, the AC current draw would be 2200VA.
I would say that the PSA2 is more conservatively designed - that is, designed to be pushed harder or run at higher duty cycles, based on the overcurrent protection.
And it isn't anything special like class G or H to reduce current consumption or dissipation - it is a hog.
Now in order to get stated performance for the 2400:
Unloaded supply = +/-114V
Assume 75% regulation: Vcc = 85.5V
Two sets of saturated transistor banks @ 5V each: Vpeak = 75.5V
Output voltage = 53.4V RMS, 712W/4R.
At 65% efficiency and 65% power factor the AC current draw would be 3370VA.
Now to get stated performance for the PSA2:
Unloaded supply = +/-75V
Assume 88% regulation: Vcc = 66V
One set of output transistors in saturation: Vpeak = 61V
Output voltage 43.1V RMS, 465W/4R.
At 65% efficiency and 65% power factor, the AC current draw would be 2200VA.
I would say that the PSA2 is more conservatively designed - that is, designed to be pushed harder or run at higher duty cycles, based on the overcurrent protection.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.