the importance of the first watt and below

[Alex2507]

This is a race I can afford. Already got me a bunch stuff that will absolutely pump out watt 1, but if they all sound the same driven within their design parameters, what difference could much else make?

Is this like an audio paradox? Is an amp's ability up high at all characteristic of how it behaves down low? Is impedance at all an issue down low on the amp section? I assume that you could just forget about the power supply not being up to snuff at that level?

Just to yank the rug out from under anyone wanting to inform me that I got amp issues, I already know I got amp issues. I'm the one farming them.

 

[Adam]

The first watt is important to me. That's about all that I use.

Btw, the degradation of photo quality on this forum has made its way from Australia, to Texas, and now to the east coast.

 

[fuzz092888]

I think the first watt is incredibly important. I'm not claiming to hear major differences, but I think the Threshold sounds better than anything else I've got. I've also read as much as I could about what Nelson Pass has to say on the matter, which is a lot considering he started a company called First Watt and is all about low power amps. Really interesting stuff.


Sent from my iPhone using Tapatalk 2

 

[Alex2507]

the degradation of photo quality on this forum has made its way from Australia, to Texas, and now to the east coast.

Congrats on setting the new record for fastest derailment ever. Very impressive.

 

[Adam]

Congrats on setting the new record for fastest derailment ever. Very impressive.

Hey, I addressed your actual question first. Kind of. Plus, your sub-par photo derailed it...I just pointed it out.

Besides, I think that everyone here knows that you could improve your first-watt performance by buying my IPS-1. Granted, one would normally need to use more than one watt to overcome the sound of it humming, but the beauty of your setup is that your amps can be on the other side of a wall from your main system.

 

[psbfan9]

Have you tried tube amps? Same results? Or unable to tell? Or haven't tried? Nelson Pass' amps are often compared to tube amps. But why? Tubes are typically rated with much less wattage, but have been used, and preferred by some, to drive low impedance speakers. Is the first watt of a tube amp more stable than solid state? Tube amps seem to be a love them or hate them thing, no middle ground. They're not discussed here much for what ever reason.

*Disclaimer* I have no idea what I just said, or what you were asking.

 

[fuzz092888]

Have you tried tube amps? Same results? Or unable to tell? Or haven't tried? Nelson Pass' amps are often compared to tube amps. But why? Tubes are typically rated with much less wattage, but have been used, and preferred by some, to drive low impedance speakers. Is the first watt of a tube amp more stable than solid state? Tube amps seem to be a love them or hate them thing, no middle ground. They're not discussed here much for what ever reason.

*Disclaimer* I have no idea what I just said, or what you were asking.

I know that both tubes and Pass amps push a s*it ton if current and that topology plays a big part in it. That's kind of the big issue with low impedance loads, they require a huge amount of current. Watts aren't the only important spec IMHO when figuring out how powerful an amp is.


Then again, I have no idea what I'm talking about either. I just know that all the amps I have were billed as "high current" and I think they're awesome

 

[Irvrobinson]

The quality of the first watt can be influenced in several ways, but IMO one important factor is the quality of the input driver stage. I think these stages are best done with a very high quality, low-noise ICs, and not discrete components, but that's just an opinion. ICs typically have better performance at low power levels, due to much shorter signal paths and consistent thermal characteristics. Another factor is how many amplification stages there are. Every additional stage represents additive noise and distortion with a different signature. The purist guys try to limit the number of stages. There's power supply design and layout, layout on the circuit boards, and topology of the final output stage, which is where there's a lot of controversy. On some amps when you examine the boards for the driver stages it looks like mostly power supply regulation and smoothing with a little bit of gain stage.

For powerful amps some argue that quad-differential output stages (four amplification paths doing the work of one) are best, because they have the highest common mode rejection ratios, but in purist, low-power amps some argue that all of that extra circuitry adds distortion and noise, an argument which taken to absurdity results in those single-ended thingies. I'm of a mind that execution quality trumps theoretical topology advantages, so I think the only way to assess an amp's first watt quality is to measure it.

I've always been attracted to Nelson Pass's products because they are so innovative. I'm not sure they sound any better, but different thinking, at least the kind with good reasoning behind it, attracts me. I also like the fact that his products, whether they were Threshold or Pass Labs, have always come out fully baked, unlike Peter Walker's Quad stuff, which for their time were arguably more innovative, but had an ongoing reputation for being works in progress.

I should emphasize that while I believe the quality of first watt performance is critical, I don't believe special amplifier architectures are necessary to achieve it, just great design and execution. I also don't believe such products need to be grossly expensive either.

 

[TLS Guy]

Amps have published specs at full power. However a lot of class A/B amps increase distortion at low power, and the specs don't show this. This was one of the driving forces behind the research of Peter Walker and Nelson Pass.

I agree about the Quad 405 series being works in progress until the later iterations of the 405 II. The 606 and later are in no way works in progress.

 

[ski2xblack]

Hmm. I fully agree with Irv's final couple lines, pretty spot on. My woo radar goes off when I hear someone go on about the precious first watt.

psb- Tube amps, in particularl the super simple SETs often favored by the "precious first watt" folks, have a whole host of issues compounding the matter, so it's hard to differentiate all the weird things they do from the good performance they may or may not have underneath it all. They add stuff that's not part of the original recording, even at low output, so yes, the first watt sounds more, well, more something with such devices. Distortion and all, they often sound hyper-real and crystal clear. So I guess the question is, is that from the quality of the first watt, or high output impedance mediated phenomena and gobs of low order harmonics?

There is even a lot of designed in idiosyncacies in the First Watt stuff, it's all over the board as far as output impedance and harmonic coloration from model to model. Some are cleaner, while some are decidedly voiced. Mr. Pass caters to his market very well, and the sound, while enjoyable as can be, is in choose your illusion territory.

 

[PENG]

All I want to see is low distortion from 0 to 1W, my amps mostly cruise between 0.05 to 0.2W with peaks to 1 to 2W. I never thought it is a big deal any more for amps to have low distortion at low output levels but I do know there are the odd ones that don't do so well at 1W and below. Fortunately all my amps, espeically the Bryston, do quite okay at below 1W. Crossover distortion could be higher at lower level but as long as THD+N is well below 0.1% I am quite sure it would have no effects on me.

 

[Steve81]

My thoughts on performance <1W:

Here's the power versus distortion of the mid-level Onkyo TX-SR706 receiver into 8 ohms and 4 ohms:
http://www.hometheaterhifi.com/images/stories/2009/march-2009/onkyo-tx-sr706-receiver-thd-n-vs-power-8-ohms-large.gif
http://www.hometheaterhifi.com/images/stories/2009/march-2009/onkyo-tx-sr706-receiver-thd-n-vs-power-4-ohms-large.gif

As most will note, THD+N is steadily on the rise by the time you're at the 0.1W mark (the lowest charted). On the 4 ohm chart, distortion reaches 0.2% THD+N at the 0.1W mark, or ~54dB below the fundamental signal.

Now my speakers are 4 ohm nominal (realistically, not as rated by the mfr), and have a sensitivity rating of 89dB w/ 2.83V or 86dB w/ 1 watt, and are at a distance of 8 feet from the listening position. Assuming an additional 4dB gain for being in room versus anechoic conditions and plugging into an SPL calculator, that 0.1W will net me ~72.5dB at the listening position, with THD+N at <20dB, which is well below the noise floor of my room. Food for thought, and of course different conditions may yield different results.

 

[Alex2507]

I've also read as much as I could about what Nelson Pass has to say on the matter ...

Thanks, I just read the wiki page on him. All of the stuff he was involved in leaves the $1/watt approach with four flat tires but I'm just trying to confuse Harald.

Have you tried tube amps? Is the first watt of a tube amp more stable than solid state? They're not discussed here much for what ever reason.

I've never heard a tube amp but that's a good question about their stability down low. I've heard it said that tube amps are not reviewed for much the same reason that Car&Driver does not review buggy whips. Adam on the other hand has an Implements of Bondage and Humiliation manifesto with a chapter or two on a variety of whips.

Crossover distortion could be higher ...

What's that? Is that like cross talk? I don't understand that either but if I guess right the Marantz monoblocks would be immune, right? How's the Bryston for heat output? Seems like that's of as much interest to me as all this watt1 stuff.

that 0.1W will net me ~72.5dB at the listening position

The graphs do show more distortion at low output driving a low impedance load. What's of interest to me is that I actually listen right at that level ... mostly. When I do crank it to obscene+1 dB, I look at the speakers and subs sort of waiting for the first cone to fly out as a signal to turn it down in hopes of saving the rest of the drivers. I would love to see Graphs like that for the various H/K rec'rs I've used.


Somewhere along the line I seem to remember hearing that distortion levels as high as 1% fall into the range of questionable human audibility. Speaker distortion of 5% and for subs, 10% are pretty common ... leaving all that 0.1 and 0.2 stuff in the mostly unimportant category

I didn't understand anything Irv said. Can we get a translator over here?
I just need to read that over a few more times.

I wonder if there might be amps out there that put out low heat and are well executed. IIRC YAA (fuzz) said his Threshold gets hot. Both my rec'rs run pretty hot even with a good deal of the work being picked up by cool running amplification. I wonder how those Quads do with heat. If it wasn't for TLS, I wouldn't be aware of them at all but just like a few of the other brands mentioned here, I remember them being kind of steep in the $/watt arena which isn't too surprising but the model I looked at only put out like 90 watts. Anyway, affordability and max output are different discussions.

 

[Steve81]

Somewhere along the line I seem to remember hearing that distortion levels as high as 1% fall into the range of questionable human audibility. Speaker distortion of 5% and for subs, 10% are pretty common ... leaving all that 0.1 and 0.2 stuff in the mostly unimportant category

The other thing to keep in mind with those graphs is that it's THD+Noise, and that at those low levels, the graph is more than likely going to be dominated by the noise floor of the system rather than objectionable high order harmonic distortion.

What's that? Is that like cross talk?

Re: Crossover distortion, this link gives some explanation:
Class-A Amplifiers explained

 

[Irvrobinson]

I didn't understand anything Irv said. Can we get a translator over here?
I just need to read that over a few more times.

I'll translate for you.

I offered the opinion that if one wants a very clean output from an amplifier at low power levels that it was better to use integrated circuits for the input driver circuitry. ICs have technical advantages at low voltage levels. Many audiophiles think ICs are anathema in their amplifiers, so many high-end amps use input stages with discrete components, and I believe that's why you see some high-end amps with worse signal-to-noise ratios than more value-priced options.

Some amplifier designers believe a different strategy should be used for low-power amps than high-power ones. For example, amplifier designers use a stair-step strategy for amplifying the input signals from the pre-amp to the voltage level required for driving the speakers. Gain is achieved in stages. A very common strategy is the three-stage approach, where the pre-amp inputs are amplified to the next voltage step, then those signals are amplified to the next step, and then those twice-amplified signals are sent to the output stage, where the signals are amplified to speaker level. Some designers feel that you can achieve a cleaner output to the speakers by using only two stages when you aren't scaling very high output heights (like 200w+). Nelson Pass uses only two stages in some of his First Watt designs. The notion is that each gain stage adds noise and distortion, so want as few as possible, so a two-stage amp might sound better than a three-stage amp.

Finally, achieving low noise and distortion when you have very high gain requirements like you do with high-power amps requires special circuitry to reduce the noise and distortion inherent in high-gain circuits. So amplifier designers use a strategy called differential signaling, which takes advantage of cancellation effects for in and out of phase signals. This is accomplished by building mirror images of the signal path. Most amps use one mirror image and only in the output stage, but some very high-power amps use a mirror image of the mirror image, or four amps doing the work of one, and some apply the differential signaling philosophy from input to output. The thinking is that the higher the gain the higher the level of cancellation you need. Purist amp designers argue that all of that extra circuitry isn't necessary for low power levels, and that all of that extra circuitry is somehow detrimental, so they play around with ways to eliminate it. The absurd conclusion are so-called single-ended amplifiers, which don't even use differential design techniques in the output stage. Some audiophiles believe single-ended amps sound inherently cleaner because they are so much simpler.

I concluded by saying I don't buy any of these arguments that more sophisticated circuitry results in worse sound at low power levels, and I believe the quality of the design and the execution in manufacturing are the biggest factors in how good an amp will perform at low power levels, and the only way to determine if an amplifier performs well is to measure it.

Is this easier to read? I stayed off the chocolate this time.

 

[Irvrobinson]

The 606 and later are in no way works in progress.

Good point. The same is true for their latest ESL-2805 speakers too.

 

[Bryceo]

Sorry Alex I clicked on this by mistake if I'd listened in electronics instead of burning people with the soldiering iron, I'd possibly have some helpful advice for you

 

[Adam]

Adam on the other hand has an Implements of Bondage and Humiliation manifesto with a chapter or two on a variety of whips.

Indeed. It is my life's masterpiece. Keep in mind, gentlemen, that the manifesto cannot be purchased...it must be earned.


Alrighty, back to amps. Dude, there's theory and then there's experience. I had a $450 Pioneer receiver, to which I added $1800 in amps (granted, about $1000 of that was an unintentional donation to the Fool-Me-Twice Fund). At lower volumes, I couldn't tell any difference between the receiver's amps or the external amps. That doesn't mean that you couldn't...but I couldn't. So, part of what I'm getting at is this - if you're happy with your amps, just enjoy them.

As for heat, there's a difference between how much heat something dissipates and how hot it gets. Something can dissipate a good deal of heat but remain relatively cool if it has good cooling, while something else that dissipates little heat can get relatively hot if it isn't designed to lose heat very efficiently or if someone doesn't allow for the necessary air flow. I mention that last part because my IPS-1 normally ran very cool to the touch, but when I was experimenting with putting it in the closet, I had it in standby mode sitting on the carpet...and it got pretty darn warm. Sure, it makes their statement to me that it dissipates something like 1W or less in standby (even though I told them that it was drawing more than that) another complete fairy tale because it would have risen less than a degree if that were true, but it's an example of adequate air flow making a big difference. Another point to this is...just because something feels relatively cool doesn't mean that it isn't drawing a decent amount of power or isn't heating up your room.

 

[ski2xblack]

Irv, you ought to love this

 

[Irvrobinson]

Irv, you ought to love this

Yeah, I've read Nelson's entire website before. I really enjoy his thinking, analysis, design skills, and writing style. I wish I could be an intern for him. I doubt I'd ever be a customer for his First Watt amps, but obviously I admire the guy. I would be a potential customer for his Pass Labs stuff, but I can't bring myself to spend that much on amps and pre-amps anymore, when the el-cheapo ATI amp that I use sounds awesome. (Ha! El-cheapo by comparison, let's say.)