How Loud is 1 Watt?

[EdmontonCanuck]

Ran across this interesting video on YouTube that I thought some might find interesting :

 

 

 

 

[sandyk]

I found this particularly interesting as my speakers are  87dB /1W@1M

 

Alex

[the_bat]

A very interesting, well done  experiment with (unusually for audio) some very clear results.  

 

I'm not quite so sure about his explanation of the importance of damping factor, rather than instantaneous current delivery, but that in no way invalidates the results.

[sphinxsix]

11 hours ago, sandyk said:

I found this particularly interesting as my speakers are  87dB /1W@1M

 

Alex

 

I think I might have found it even more interesting since mine are 95dB 

 

[Confused]

This reminds me of this clip:

 

 

In the above, the amount of power required is several orders of magnitude higher than is "measured" the clip in the OP.  It is certainly interesting to compare the two clips

 

OK - less efficient speakers, a larger room, different music, these are all contributing factors, but I do not think these factors alone explain the huge differences in what is "measured" here.

 

Speculation, but I suspect that a large part of the difference is how the "transient" power requirement is measured.  In the first clip, the measurement is done with a volt meter, which I suspect is averaging the reading over small periods of time.

 

In the case of the clip hear, the power is measured by the amps, which are CH Precision A1's.  

 

This from the CH Precision manual:

 

Power monitoring. Each power amplifier board is equipped with a DSP that monitors the instantaneous output voltage and current of each M1 channel. Both values are sampled at around 100 kHz, ensuring peak values are properly detected. This circuitry has several purposes: give the user a feedback of the peak power fed to the loudspeakers, and detect malfunctions such as short-circuits or amplifier damage.

 

I am thinking that because the CH Precision amps measure at 100 kHz, they show a true transient value, rather than a more averaged value as indicated by a volt meter.  So maybe the guy in the first video is actually using more watts than he might think?

 

As an aside, I tried a similar volt meter test myself once.  My results were fairly similar to that shown in the first clip, although if someone wants to lend me some CH Precision A1's for a few months, I'd be happy to revisit this.

[davide256]

I resemble that test, using the 1 watt of a Microzotl 2 to power 85db efficient Magnepan 1.7's in a 12x12 room.  I can run out of steam/watts in the low bass but the Zotl

is so much better on transients that I can tolerate that more than the loss of information  I've gotten from pairing with a 70W Prima Luna or a 100w modded Hafler amp.

[yamamoto2002]

It seems Mastech MS8217 he uses indicates average voltage. Instant power of musical signal is larger than average power.

 

I tested with my Mastech MS8264 DMM and Tafelmusik baroque orchestra Vivaldi the four seasons track 6 (Summer; Presto) and

found:

Bar 1 to 9 : meter shows 0.15V (0.0028W) to 0.23V (0.0066W) but Peak power is 0.18W

Bar 10 to 28 : 0.15V (0.0028W) to 0.18 V (0.0040W), Peak Power = 0.072W

Bar 29 to Bar 39 : 0.11V (0.0015W) to 0.20 V (0.005W), Peak Power = 0.061W

Bar 40 to Bar 54 : 0.03V (0.00011W) to 0.2 V (0.005W), Peak Power = 0.058W

Bar 55 to Bar 66 : 0.12V (0.0011W) to 0.17 V (0.0036W). Peak Power =  0.097 W

 

So, actual peak power of this recording is about 20 times larger than meter indicated value. If meter indicates 1W, amp should deliver 20W.

 

Tube amps clip softly when power is saturated and it is relatively less problematic, but when solid state amps clip, it saturates hard to rail voltage and high pitched harsh noise is heard. I experienced it on 84.5 dB/W/m speakers and 1W DIY discrete solid state audio amp.

 

The first video uses tube amps with average voltage (power) metering and the second video uses solid state amps with instant power metering.

 

[audiobomber]

Master Volume on my DAC/preamp is set for -12dB. I played the first tune from the Harbeth Video at -12dB. My Radio Shack meter never went above 87dB (set for fast response, C-weighted), which is quite OK with me, because I would not want to listen any louder, ever.

 

I have no idea how much power I was using, couldn't have been much. My monoblocks are rated for 100wpc@8ohms, 200wpc@4ohms, the speakers are 4-ohms, [email protected]/m, subs are 155W into 8 ohms, 250W into 4 ohms. I believe the sub drivers are 8ohms. I've driven the monitors with 50wpc and never heard them clip.

 

The 100W amps sound better though, even though both sets of amps are from the same manufacturer and use the same topology and similar parts. I attribute this to current delivery. The 100W monoblocks are rated to deliver 35A, the 50W set are rated for 10A,  so 20A when I use both. I agree with the video in the original post, it's not power that matters, it's current delivery. Something not mentioned so far is that some amps can delivery instantaneous bursts of power beyond their steady-state rating. I believe this capability is due to superior current capability.

 

This is an interesting video:

 

 

 

[Kimo]

Tomorrow, I will be receiving a Linear Tube Audio amp producing 12 watts into 4 ohms.  My current amp does 250.  My speakers are rated at 4 ohms and 87db with dips coming in at 2.8.

 

We shall see if a little power can come close to keeping up with adequate power.  I have my doubts.

[PeterSt]

35 minutes ago, Kimo said:

I have my doubts.

 

Me too. 🤪

Watch out for good control. Loud does not mean that per se (the contrary). I did not watch the whole video, but this will be about the damping (I guess).

[Kimo]

4 minutes ago, PeterSt said:

 

Me too. 🤪

Watch out for good control. Loud does not mean that per se (the contrary). I did not watch the whole video, but this will be about the damping (I guess).

Dampening with tubes and no transformers, no less.  

[fas42]

No surprise to me ... I have done plenty over the years with nominally low powered amplifiers; which have had no trouble filling the room with intense, immersive sound.

 

What it's about is whether the sound at the level where you "have to shout", to be heard, is completely 'clean' - often it's not, and that's the signpost to there being problems, 😉.

 

People have no trouble arguing about power needs all day, but once the concept of audible distortion enters the conversation, it all goes quiet 🙃 ... unfortunately, this is the Final Frontier, 😜.

[Kimo]

On 11/26/2020 at 2:46 PM, PeterSt said:

 

Me too. 🤪

Watch out for good control. Loud does not mean that per se (the contrary). I did not watch the whole video, but this will be about the damping (I guess).

Well.

 

I have not yet received the 12 watt amp, but I did receive the preamp, which also does duty as a 1 watt amp and headphone amp.

 

You have to turn the volume most of the way up, but I have not had to max it out, and  yet I am getting pretty decent volume and excellent control out of 1 ZOTL watt. 

 

I am pretty surprised by this result.  The room is not small. 

[EdmontonCanuck]

I also have the first gen MicroZOTL 1-watt amp, and am thrilled with it as a headphone amp. I always thought that this little amp would drive a pair of Klipsch Heresey's quite well (if I had them LOL).

[Kimo]

15 hours ago, EdmontonCanuck said:

I also have the first gen MicroZOTL 1-watt amp, and am thrilled with it as a headphone amp. I always thought that this little amp would drive a pair of Klipsch Heresey's quite well (if I had them LOL).

I will be hooking it up to the Z10 today.  The Z10 working straight off the DAC's volume control seems to have plenty of headroom and, again, excellent control.  The Z10 is very clean and detailed.  

[phusis]

The Pearl Acoustics Sibelius speakers are filter-less, and so their full-range driver is coupled directly to the amplifier making for a fairly easy load (in that sense one could say they're actively driven). This is not trivial when it comes to taking the most advantage of the power capabilities of the amp at hand, as conversely a complex passive filter topology with challenging phase behavior could potentially, and quite easily drain a given amp so much that it would require much more power to achieve the same SPL, with a shared sensitivity, compared to the easy-to-drive Sibelius speakers. That is to say: for the views of the OP's supplied video to have general validity it would necessitate an amp more or less impervious to load in a range of speaker contexts, which is hardly the case with most running a wide variety of passive filter-driven speakers. 

This may be a case where Pearl Acoustics are trying perhaps to "alleviate" the limited SPL potential of their no doubt very fine Sibelius speakers, and in their specific context it's quite obvious going by the video that a single watt will get you a long way SPL-wise, even with a moderate to low sensitivity here. However the practically limited Sibelius SPL envelope of 96-98dB's, while quite loud with most compressed music material, only gets you so far with recorded material that has a wider dynamic range that could easily see them small cones leave their baskets for good. Tom Danley of Danley Sound Labs played a non-compressed recording at his home with full dynamic bandwidth of some heavy fireworks through ~99dB sensitivity SH50 Synergy horn main speakers and two 97dB sensitivity TH50 tapped horns subs, all coupled to 1kw amps, and he still succeeded in making the amps clip. Look at those sensitivity and wattage numbers again for some perspective on what it takes to reproduce realistic dynamics.. 

This is not meant to put down Pearl Acoustics Sibelius speakers, mind you. One point source per channel, no passive (or other) filter in the chain, a folded horn, very nice cabinet build quality and overall attention to detail seems like a recipe for lovely sound in limitedly spaced surroundings. 

[botrytis]

52 minutes ago, phusis said:

The Pearl Acoustics Sibelius speakers are filter-less, and so their full-range driver is coupled directly to the amplifier making for a fairly easy load (in that sense one could say they're actively driven). This is not trivial when it comes to taking the most advantage of the power capabilities of the amp at hand, as conversely a complex passive filter topology with challenging phase behavior could potentially, and quite easily drain a given amp so much that it would require much more power to achieve the same SPL, with a shared sensitivity, compared to the easy-to-drive Sibelius speakers. That is to say: for the views of the OP's supplied video to have general validity it would necessitate an amp more or less impervious to load in a range of speaker contexts, which is hardly the case with most running a wide variety of passive filter-driven speakers. 

This may be a case where Pearl Acoustics are trying perhaps to "alleviate" the limited SPL potential of their no doubt very fine Sibelius speakers, and in their specific context it's quite obvious going by the video that a single watt will get you a long way SPL-wise, even with a moderate to low sensitivity here. However the practically limited Sibelius SPL envelope of 96-98dB's, while quite loud with most compressed music material, only gets you so far with recorded material that has a wider dynamic range that could easily see them small cones leave their baskets for good. Tom Danley of Danley Sound Labs played a non-compressed recording at his home with full dynamic bandwidth of some heavy fireworks through ~99dB sensitivity SH50 Synergy horn main speakers and two 97dB sensitivity TH50 tapped horns subs, all coupled to 1kw amps, and he still succeeded in making the amps clip. Look at those sensitivity and wattage numbers again for some perspective on what it takes to reproduce realistic dynamics.. 

This is not meant to put down Pearl Acoustics Sibelius speakers, mind you. One point source per channel, no passive (or other) filter in the chain, a folded horn, very nice cabinet build quality and overall attention to detail seems like a recipe for lovely sound in limitedly spaced surroundings. 

 

More then likely, no - they probably have a baffle step compensation filter and some other, single drivers, do use a Zobel network to tame the overwhelming midrange produced.

[the_bat]

52 minutes ago, phusis said:

and he still succeeded in making the amps clip

Was he in the room when he did this, and if so does he have any hearing left?

 

That translates (I think) to 122 db spl which would cause immediate pain and ear injury.

[fas42]

3 hours ago, phusis said:

This may be a case where Pearl Acoustics are trying perhaps to "alleviate" the limited SPL potential of their no doubt very fine Sibelius speakers, and in their specific context it's quite obvious going by the video that a single watt will get you a long way SPL-wise, even with a moderate to low sensitivity here. However the practically limited Sibelius SPL envelope of 96-98dB's, while quite loud with most compressed music material, only gets you so far with recorded material that has a wider dynamic range that could easily see them small cones leave their baskets for good. Tom Danley of Danley Sound Labs played a non-compressed recording at his home with full dynamic bandwidth of some heavy fireworks through ~99dB sensitivity SH50 Synergy horn main speakers and two 97dB sensitivity TH50 tapped horns subs, all coupled to 1kw amps, and he still succeeded in making the amps clip. Look at those sensitivity and wattage numbers again for some perspective on what it takes to reproduce realistic dynamics.. 
 

 

There is an angle here that probably a lot of people don't appreciate - a 1kW amp quite often fails to perform as it should, and the reason is that the power supplies aren't good enough - the supply rails can drop dramatically in level, as a transient spike of loss of the design voltage - and the amp clips, momentarily. Why? Because the power supply is too primitive, and can't deliver the energy fast enough to maintain the correct numbers, the current requirements are more than the engineering allows.

 

I did some analysis many years ago using Spice circuit simulation software, and it was easy to make conventional power supplies fail to perform, purely as a model, by using closer to real world parts. And real world gear would do even worse.

 

Which is why some amps which have very simple amplifying circuits, but monster power supplies, do so well in listening tests - the unit actually works as intended, rather than being close to the edge when more volume is desired.

[phusis]

2 hours ago, botrytis said:

 

More then likely, no - they probably have a baffle step compensation filter and some other, single drivers, do use a Zobel network to tame the overwhelming midrange produced.

 

"The absence of a crossover network almost guarantees a simple and straightforward amplifier load, and that did indeed turn out to be the case. The impedance stays comfortably above 6ohms throughout, and the quarter wave line is tuned to 48Hz. ..."

https://pearlacoustics.com/wp-content/uploads/2020/01/Sibelius396.pdf

[phusis]

3 hours ago, the_bat said:

Was he in the room when he did this, and if so does he have any hearing left?

 

That translates (I think) to 122 db spl which would cause immediate pain and ear injury.

 

Very short SPL-bursts of this magnitude (or whatever the more precise level of sound pressure must have been in this case) are, if not in any way regularly recommended, tolerable to the human ear.

 

The point remains: realistic dynamics with ample headroom requires more than one would think. Say the dynamic envelope of a symphony orchestra where levels up to or in the vicinity of ~110dB's are not uncommon, depending of course on how close you're seated to orchestra. For this level to be reproduced fairly faithfully by one's home stereo not only requires for it to be able to reproduce up to that SPL, by way more for it to be an effortless, low distortion presentation. Which, as it happens, would then be perceived less loudly the more cleanly it's reproduced.  

[phusis]

54 minutes ago, fas42 said:

 

There is an angle here that probably a lot of people don't appreciate - a 1kW amp quite often fails to perform as it should, and the reason is that the power supplies aren't good enough - the supply rails can drop dramatically in level, as a transient spike of loss of the design voltage - and the amp clips, momentarily. Why? Because the power supply is too primitive, and can't deliver the energy fast enough to maintain the correct numbers, the current requirements are more than the engineering allows.

 

I did some analysis many years ago using Spice circuit simulation software, and it was easy to make conventional power supplies fail to perform, purely as a model, by using closer to real world parts. And real world gear would do even worse.

 

Which is why some amps which have very simple amplifying circuits, but monster power supplies, do so well in listening tests - the unit actually works as intended, rather than being close to the edge when more volume is desired.

 

It's a common misconception to believe pro amps sport malnourished power supplies, or are otherwise sonically lacking. Some are better then others, naturally, but in quality installations they're used within their specs and built to last. Some of the best amps I've heard are from the pro segment, like Crown's Studio Reference series and MC². 

Tom Danley wouldn't have used anything that wasn't up to the task; it's not amp insufficiency, but a true indication of what live dynamics actually necessitates. 

[fas42]

8 minutes ago, phusis said:

 

The point remains: realistic dynamics with ample headroom requires more than one would think. Say the dynamic envelope of a symphony orchestra where levels up to or in the vicinity of ~110dB's are not uncommon, depending of course on how close you're seated to orchestra. For this level to be reproduced fairly faithfully by one's home stereo not only requires for it to be able to reproduce up to that SPL, by way more for it to be an effortless, low distortion presentation. Which, as it happens, would then be perceived less loudly the more cleanly it's reproduced.  

 

Many rigs fail to do the higher SPLs with "effortless, low distortion" - the 'strain' is quite obvious, and people immediately wind back the volume.

 

If the sound is presented at high levels very cleanly, then the adjective used is no longer "loud", but rather, "intense" - the experience mimics live music dynamics, and immersiveness - it becomes, "fully convincing".

[fas42]

5 minutes ago, phusis said:

Tom Danley wouldn't have used anything that wasn't up to the task; it's not amp insufficiency, but a true indication of what live dynamics actually necessitates. 

 

Sorry, live dynamics requires the gear to work 'correctly', rather than just to specs - I have heard what well sorted, low power amplifiers can deliver as a subjective experience; and what a mess of it very powerful amplifiers in a non-optimised rig can make, too many times to think otherwise ...

[phusis]

19 minutes ago, fas42 said:

 

Many rigs fail to do the higher SPLs with "effortless, low distortion" - the 'strain' is quite obvious, and people immediately wind back the volume.

 

Sure; most any rig, not least one driven too far from its unstrained operation, can be made to sound this way.

 

My intention for this discussion though is to highlight what it would take for a home set-up to reproduce live dynamics fairly accurately. Very few come close..

 

Quote

 

If the sound is presented at high levels very cleanly, then the adjective used is no longer "loud", but rather, "intense" - the experience mimics live music dynamics, and immersiveness - it becomes, "fully convincing".

 

I'm basically saying the same thing, so nothing to argue here.