[James-uk]

http://www.innerfidelity.com/images/130620_AmpMeasurements_PDF_Booklet.pdf

These headphone amp measurements show nicely. Until clipping something like the O2 stays under audible distortion levels.

 

[SunshineReggae]

  Can someone please advise me on something. I've always believed that driving an amplifier at higher levels meant that distortion would increase, even below clipping.
 
I've also been told by many audiophiles and salesman that it is far better to drive an amp that has power in reserve, so that it is not stressed than one that is running close to its limits. Hence a larger amp would be better, because it can handle the demands easier.
 
Now for the advise part. What I've recently gathered from other places is that distortion does not increase linearly with power output.  Is this true? So if I use an amp that is rated to deliver 100W, and I use 70W, the distortion between say 30W-70W won't be all that different?
 
If I'm wrong about this then it means that amp headroom doesn't really do anything useful, which goes contrary to what I've learned for years.

 
Yeah, I know that story about headroom. "You can hear the amp's power below the surface, just waiting to be used."
 
But that is not how amps work. Headroom is basically just power that's not used. There is no point in headroom in and of itself; it doesn't guarantee less distortion or better sound. All you need is sufficient power at the level you listen to at its loudest.
 
It seems like you think that if the amp has a certain amount of power that it can deliver, then its distortion will fall into a certain range, making it more or less audible. But that's complicating the issue way too much, and approaching the idea of headroom in a very convoluted and unpractical manner (and I don't mean that with any offense at all).

 

[blades]

  Can someone please advise me on something. I've always believed that driving an amplifier at higher levels meant that distortion would increase, even below clipping.
 
I've also been told by many audiophiles and salesman that it is far better to drive an amp that has power in reserve, so that it is not stressed than one that is running close to its limits. Hence a larger amp would be better, because it can handle the demands easier.
 
Now for the advise part. What I've recently gathered from other places is that distortion does not increase linearly with power output.  Is this true? So if I use an amp that is rated to deliver 100W, and I use 70W, the distortion between say 30W-70W won't be all that different?
 
If I'm wrong about this then it means that amp headroom doesn't really do anything useful, which goes contrary to what I've learned for years.

 
Yes it is generally true that distortion increases with amplifier output.  In fact power ratings are based on distortion.  Virtually every amp will perform past its full rated output but the distortion then becomes audible.
 
There is nothing wrong with having reserve power.  Virtually every amplifier has plenty of it.  But 20 watts from a 50 watt amplifier isn't any different from 20 watts from a 200 watt amplifier.  It doesn't cause more stress or heat.  it doesn't sound any different.
 
The question is not something that cannot be answered without more information.  Let's say that at 70 watts (something you aren't actually likely ever to encounter) your amp might measure a THD figure of 1/10 percent.  At 30 watts it may have 1/100 of a percent distortion.  Neither figure is audible so it isn't an issue.  At 150 watts the amp might be at 2% distortion and you could likely hear that if you concentrated on it.
 
Headroom is nothing more than unused power.  Let's say that your amplifier dissipates 1/2 watt on average driving speakers of average sensitivity.  That is a pretty typical number.  In order to have the ability to handle a peak SPL that is twice as loud as average, you would need 5 watts of power.  All the power above 1/2 watt is overhead and you would need at leats 5 watts of overhead to avoid distortion on peaks.  So your amplifier may get its average from SPL's ranging from the 5 watt dissipation down to milliwatts.  So headroom is important.  What it is not is complicated.

 

[castleofargh]

well you can dumb it down and see what you need.
-obviously having not enough "headroom" is a bad idea. we get an amp so it can drive our gears.
-just enough is a little tricky because as mentioned it's important that the amp has enough for the peaks outputs.(well it's not really tricky, you just need to be careful with the specs they are using for marketing).
-too much headroom, depending on the amp might lead to noise. troubles fine tuning your volume knob and channel imbalance, if you have too much gain and depending on the volume control, analog, "digital", with steps(pairs of resistors). and a few amps have actually more distortion when being forced to output really low voltage.
 
so some headroom yes. a shiiiiitload of it, why?

 

[bigshot]

Headroom is only important if a peak comes along that requires it. More headroom isn't better. All you need is enough.

 

[Hijodelbrx]

I understand that theoretically one doesn't need tons of power. But I've always subscribed to that saying; "better to have it and not need it, than to need it and not have it"

 

[ab initio]

I understand that theoretically one doesn't need tons of power. But I've always subscribed to that saying; "better to have it and not need it, than to need it and not have it"

 
One could say the same thing about low noise electronics. In this case it reverses whether or not you want 95% unused power for headroom.
 
 
Cheers

 

[jcx]

you would rather have noise and not need it?

 

[Hijodelbrx]

I have a 100W Music Reference RM9 tube amp, an Audio Research LS15 tube preamp, and Spendor SP100's. The music sounds glorious! The 'noise' you guys are harping about is inaudible. If you're ever in Palm Beach Florida you're welcome to come over and hear for yourself!

 

[bigshot]

Not all tube amps have audible noise or distortion. Some tube amps sound as good as solid state amps.

 

[blades]

  Not all tube amps have audible noise or distortion. Some tube amps sound as good as solid state amps.

 
Yes, there are a few.  I used to have an Audio Research tube amp that scored no audible difference when compared to a solid state amp.  But there aren't many.

 

[ab initio]

  you would rather have noise and not need it?

 
That's basically what we are debating with 90% headroom amps, no? 

 
Cheers

 

[Mezzo]

The best way to match levels is to measure the voltage across the speaker terminals while playing a test tone of 1000 hz or so.  If you are using headphones, then you would measure the voltage at the cable connector.  Set the meter to AC volts and put the probes on the connector.  Very simple.

 
It would be great if someone could actually show how this is done, not by words, but with a video. Posted on Youtube, for instance, showing how it's all done.
 
I would love to be able to level-match my gear, but I wouldn't know where to start. If someone could show how it is done in real time it would help immensely. Problem is, I don't see any videos posted anywhere for this kind of thing, so I imagine most people (layman) will never know how its done.
 
It's one thing to explain with words, but actions speak louder, if you know what I mean.

 

[stv014]

I do not have a YouTube video, but for measuring the voltage on headphones, if you do not want to solder, you need:
- a digital multimeter that is capable of measuring AC voltage, and is at least somewhat usable at ~1 kHz (it is not a problem if it e.g. only shows half voltage at that frequency, as long as it does so consistently)
- a splitter with one female and two male TRS connectors, like this one
You should then use the splitter as an extension cord (headphones to female connector, one of the male connectors to the amplifier's headphone jack), and the voltage can be measured on the other male connector. Or it can be plugged into a sound card line input, if you are aware of certain caveats regarding the use of sound cards as audio analyzers, and their work-arounds (see here, for example). But for simple level matching, just measure the voltage on the connector with the multimeter:
- the tip is the left channel
- the ring (middle part) is the right channel
- the sleeve is ground (common for both channels)
As a test signal, you can use a 1 kHz sine wave, which can be generated with a variety of freely available software. If you are matching two amplifiers, adjust the volume knobs until both output the same voltage within 1% error (~0.1 dB), and the volume is also suitable for normal listening. You may also want to check for channel imbalance by measuring both the left and right channels: If it is too different between the amplifiers (this can happen especially at low volume settings), then they could sound different even if the overall level is accurately matched.

 

[Mezzo]

Sorry, I should have been more clear. I am talking about speakers, not headphones. But I think I understand what you are saying and the 1 kHz test signal makes sense. Thanks for the post!
 
However this :
 

If it is too different between the amplifiers (this can happen especially at low volume settings), then they could sound different even if the overall level is accurately matched.

So if one measures the left and right channels and there is imbalance then basically there is no way to level-match properly? So the test is basically worthless?
 
I want to know how to level-match equipment for my own benefit and also if I want to know if I can hear the difference between electronics. I don't have any background or experience in designing listening tests, so I have no idea how to go about this, but I would rather start somewhere and not at all.
 
Thanks again for the feedback.