Power into 4 Ohm "one channel driven"?

[albireo]

Hi,

I was checking the specs of a number of integrated amplifiers and noticed that several producers, when indicating the power in a 4ohm load, only report a figure for "one channel" driven. Why is that? Also what would the power in two channels be?

eg if they report
Power in to 8 Ohm both channel driven >120 Watts
Power in to 4 Ohm one channel driven >200 Watts

What can I tell about the power into 4ohm for two channels?

I read somewhere that good amplifier designs always double the wattage when the load halves hence the question. Thanks!

 

[martin clark]

It's simply because its easiest to hook-up the test gear one channel at a time (no need to duplicate the test kit). That said, in amps with non-regulated supplies the internal supply voltages may well drop slightly when supporting two channels at full output power and that makes the figures look not quite as good.

Since doubling the output power only gives you a maxium of an extra 3dB of sound pressure level in the room, it's not a difference you'd hear though. Maximum output power alone is probably one of the least useful specs there is, unless you have very large room to fill and inefficient speakers.

 

[sergeauckland]

The reason they don't quote it into 4 ohms, both channels driven is that it would look bad!

The amplifier mentioned above is already somewhat limited in its 4 ohm output, probably because of poor power-supply regulation dropping the supply volts rather than current limiting. With two channels both pulling current into a 4 ohm load, I would expect the power to drop to something like 180 watts, maybe lower, which looks a lot worse than the 240 watts one should get with an amplifier with a properly stabilised supply.

S.

 

[rontoolsie]

With both channels typically being supplied by a common transformer and bank of filter capacitors, it is not surprising that doubling the demand on these through running both channels would reduce the amount of wattage each channel could deliver to a point where some would chose to not specify the output for both channels driven. IMO, having a significant disparity between available power for one vs both channel driven is somewhat of a red (?pink) flag, just like it is when the power does not come close to doubling when going from an 8 ohm to a 4 ohm load (except in tube amps). And this is why monoblocks are usually a good thing, in that there is less intermodulation (i.e. NONE) between the channels.

 

[martin clark]

IMO, having a significant disparity between available power for one vs both channel driven is somewhat of a red (?pink) flag, just like it is when the power does not come close to doubling when going from an 8 ohm to a 4 ohm load

Again - it depends.

If you measure an amp one channel; vs two channels driven, 8, vs 4ohms you'll see four disctinct figures, falling-off each time (wih increased loading)

If however you market an amp as say 50W 8 ohms, 100W 4ohms, and can hit or exceed those solidly with both channels driven, you may also see more than 50W out of 8ohms one channel driven on test (voltage rising a bit unloaded)

Is this a dodgy amp or one that over-delivers ..?

 

[Stevie A]

Also many manufacturers,especially those making AVR's do'nt like you to see the "Real world" output when all channels are driven,i had to laugh at a recent review of one "Japanese" amp that was rated at 5x 110w,when actually hooked up to a real load the modest output of 5 x 37w was its true output!!!!!

 

[albireo]

Since doubling the output power only gives you a maxium of an extra 3dB of sound pressure level in the room, it's not a difference you'd hear though. Maximum output power alone is probably one of the least useful specs there is, unless you have very large room to fill and inefficient speakers.

What specs would you be looking at in an integrated if you had a medium-sized room (11x16feet) and rather inefficient standmounts (85db@4ohm)?

 

[martin clark]

I'd look at my budget and try to hear a few on the shortlist that results, drivign the same or comparable speakers. There's only one person to be please in all this - you

ETA: even 50W/ch should get you to peaks up around /just over 100dB and that should be about the right ballpark on occasion. 100w/ch would only add 3dB to that (up to about 105dB peak, if the speakers can deliver it) and 200W/ch only another 3dB again. But power output and perception of loudness are not the same thing at all.

 

[sergeauckland]

Again - it depends.

If you measure an amp one channel; vs two channels driven, 8, vs 4ohms you'll see four disctinct figures, falling-off each time (wih increased loading)

If however you market an amp as say 50W 8 ohms, 100W 4ohms, and can hit or exceed those solidly with both channels driven, you may also see more than 50W out of 8ohms one channel driven on test (voltage rising a bit unloaded)

Is this a dodgy amp or one that over-delivers ..?

This raises an interesting point:-

Some years ago, tests were done comparing amplifiers with stabilised power supplies and those with normal sagging supplies.

An amplifier with a properly stabilised supply will give its maximum power, virtually doubling into 4 ohms, on a continuous wave, but the power doesn't increase at all dynamically on music. An amplifier with a conventional sagging supply will have more power dynamically than continuous as the power supply capacitors don't get discharged on music peaks like they do on a continuous wave.

If both amplifiers were adjusted such that they both had the same dynamic power, it was found that the stabilised supply amplifier sounded better at high levels, but if the amplifiers were adjusted such that their continuous powers were the same, the sagging supply amplifiers sounded better at high levels as it had greater peak power capability.

Amplifiers with sagging supplies can have different levels of power supply stabilisation depending on the size of the reservoir capacitors and the regulation factor of the mains transformer, and so have different dynamic capabilities. That's the main reason why when comparing amplifiers, it's necessary to make sure they don't reach their maximum power capability as otherwise what's being compared is not the sound of the amplifiers, but the sound of their overload capabilities. Some will say that this is a part of an amplifier's performance and reflects how they're used at home, but my view is that if an amplifier needs to be used in its overload region, one needs a bigger amplifier.

S.

 

[albireo]

Thank you all for your comments. I'm particularly interested in this bit by sergeauckland:

-------------------------
That's the main reason why when comparing amplifiers, it's necessary to make sure they don't reach their maximum power capability as otherwise what's being compared is not the sound of the amplifiers, but the sound of their overload capabilities. Some will say that this is a part of an amplifier's performance and reflects how they're used at home, but my view is that if an amplifier needs to be used in its overload region, one needs a bigger amplifier.
-----------------------------------

Now I'm aware of the necessity to demo, try with one's own ears, purchase according to budget, etc. However I was wondering if there's any particular technical spec I could evaluate in order to restrict my "amp demo shortlist" only to amps that are well-specified for my speakers - in the sense explained by serge. Thanks!

 

[sergeauckland]

I'd look at my budget and try to hear a few on the shortlist that results, drivign the same or comparable speakers. There's only one person to be please in all this - you

ETA: even 50W/ch should get you to peaks up around /just over 100dB and that should be about the right ballpark on occasion. 100w/ch would only add 3dB to that (up to about 105dB peak, if the speakers can deliver it) and 200W/ch only another 3dB again. But power output and perception of loudness are not the same thing at all.

Sensible advice.

With your 4 ohm rated loudspeakers, you need to find out what the minimum impedance is, and at what frequency. If it's at, say 80Hz, that will require a lot more grunt (sorry for the technical jargon) than if it's at 8kHz.

Bear in mind also that in all probablility your 85dB/4 ohm sensitivity is based on the 8 ohm watt, i.e. at a voltage level of 2.83v. This means that your loudspeakers are in reality only 82dB/ watt based on 4 ohm watts.

Nevertheless, an amplifier that is happy driving, say, 100 watts into 4 ohm and is specified also into 2 ohms will be happy driving your 'speakers and will raise over 100dB SPL in-room.

S.

 

[albireo]

Hi Serge thanks - Impedance for these reaches a minimum of 4.33 ohms at 216 Hz and a phase angle of --34.92 degrees at 111 Hz. After that they tend to stay around 6ohm pretty much all the time. The speakers are Dynaudio Focus 110 by the way. And yes, the sensitivity is based on 2.83v level. Based on what you say, it sounds like they might be quite demanding speakers.

 

[sergeauckland]

Hi Serge thanks - Impedance for these reaches a minimum of 4.33 ohms at 216 Hz and a phase angle of --34.92 degrees at 111 Hz. After that they tend to stay around 6ohm pretty much all the time. The speakers are Dynaudio Focus 110 by the way. And yes, the sensitivity is based on 2.83v level. Based on what you say, it sounds like they might be quite demanding speakers.

4.33 ohms and 35 degrees phase angle isn't too severe at all, so any amplifier that can provide 100 watts into 4 ohms will be fine. I'm quite impressed with my Behringer A500s for something new that's cheaper than anything used!

S.

 

[sojourner]

With both channels typically being supplied by a common transformer and bank of filter capacitors, it is not surprising that doubling the demand on these through running both channels would reduce the amount of wattage each channel could deliver to a point where some would chose to not specify the output for both channels driven. IMO, having a significant disparity between available power for one vs both channel driven is somewhat of a red (?pink) flag, just like it is when the power does not come close to doubling when going from an 8 ohm to a 4 ohm load (except in tube amps). And this is why monoblocks are usually a good thing, in that there is less intermodulation (i.e. NONE) between the channels.

This is very much an 'it depends' situation. I'm currently building an amp with a regulated front end supply that is more than capable of proiding the 20 to 30ma required for both channels. Each power amp board has decoupling capacitors on board as well. I am confident that there will be little or no intermodulation between channels at the front end.

The output stage has good psrr so whilst ultimate 2 channel into 4 ohms power will suffer in comparison to 1 or 2 channels into 8 ohms, in normal non-clipping use there shouldn't really be an issue with intermodulation between the channels as a result of psu design.

To be fair, a lot of commercial designs seem to be quite compromised, presumably in order to save on the costs of components.