Cambridge CXA81 MKII Amplifier Review

[LTig]

But it n class B, AB and D you do not drive it in linear range.

If you drive the amplifier in its linear range. The individual components do not matter as long as the amplifier itself is linear.

 

[Yevhen]

Luckily the frequency range of audio signals is far below radio frequencies.

Again, I'm not an expert in semiconductor technology for audio devices, but not in radio frequency transistors the trapping effect/gate lag can be quite long, in a millisecond range.

https://www.semanticscholar.org/paper/Investigation-of-Trap-Induced-Gate-Lag-Phenomenon-De-Dutta/97b968cbc58d0c7212d49024ebbbfb68f7e5c1ed

 

[LTig]

I believe the transient response of the passive 2 way coax speaker will still be affected by delay in the crossover

The best coax driver currently known (Genelec One series) is driven actively by a DSP crossover.

 

[Yevhen]

If you drive the amplifier in its linear range. The individual components do not matter as long as the amplifier itself is linear.

The amplifier will be linear as long as you can correct it with the feedback loop. During the gate lag period you cannot effectively control what is happening with the drain.

It is not seen in the Audio Precision reports as you measure performance of PA with the continuous wave. So you only see the THD of the predictable continuous wave

 

[LTig]

Again, I'm not an expert in semiconductor technology for audio devices, but not in radio frequency transistors the trapping effect/gate lag can be quite long, in a millisecond range.

https://www.semanticscholar.org/paper/Investigation-of-Trap-Induced-Gate-Lag-Phenomenon-De-Dutta/97b968cbc58d0c7212d49024ebbbfb68f7e5c1ed

This paper is behind a pay wall. But using transistor types designed for RF in audio would be a bad choice anyway ...

 

[LTig]

The amplifier will be linear as long as you can correct it with the feedback loop. During the gate lag period you cannot effectively control what is happening with the drain.

It is not seen when in the Aodio Precision reports as you measure performance of PA with the continuous wave. So you only see the THD of the predictable continuous wave

As I wrote you can use a null test with music of your choice and see if this phenomenon exists in audio power amps.

 

[Yevhen]

If the simulations in this article is correct, then the thermal drift signal at the output is just a small wandering DC offset (corrected by large negative feedback, as most power amps have a DC gain of 1) while the audio signal itself is still corrected by negative feedback

This simulation is not going far enough. You need to see how the imperfections in the semiconductor monocrystal create the traps for electrons and holes. These traps indeed can be considered as DC offset but only at a short period of time. And this DC offset NOW depends on what music your PA played in the past and whan it will play in the FUTURE (this one I don't understand myself , just see it in the articles and workshops from semiconductor suppliers). Because your feedback loop doesn't know what will happen in the future, it cannot compensate the trapping effect.

 

[LTig]

This simulation is not going far enough. You need to see how the imperfections in the semiconductor monocrystal create the traps for electrons and holes. These traps indeed can be considered as DC offset but only at a short period of time. And this DC offset NOW depends on what music your PA played in the past and whan it will play in the FUTURE (this one I don't understand myself , just see it in the articles and workshops from semiconductor suppliers). Because your feedback loop doesn't know what will happen in the future, it cannot compensate the trapping effect.

The feedback loop corrects the current signal, as long as the amplifier has a loop gain higher than 1. I don't see the effect of the DC offset drift to reduce the loop gain significantly, at least not using proper suited transistors in a well designed circuit.

 

[Yevhen]

As I wrote you can use a null test with music of your choice and see if this phenomenon exists in audio power amps.

@amirm would you be interested in organising such a test comparing, for example, tube and transistor amplifiers? We need to solve the mystery why musicians prefer tube amps

 

[LTig]

@amirm would you be interested in organising such a test comparing, for example, tube and transistor amplifiers? We need to solve the mystery why musicians prefer tube amps

Do you have some evidence that the majority of musicians really prefer tube amps for audio reproduction (not guitar amps)? I heard that musicians often don't care much about sound quality of stereo systems at all.

 

[amirm]

@amirm would you be interested in organising such a test comparing, for example, tube and transistor amplifiers? We need to solve the mystery why musicians prefer tube amps

Musicians don't prefer such. 99.999999% of the world music is created with solid state technology and feedback. Take those away and there will essentially be no music in the world!

 

[Yevhen]

Musicians don't prefer such. 99.999999% of the world music is created with solid state technology and feedback. Take those away and there will essentially be no music in the world!

Sorry, I didn't make it clear. I'm talking about guitar amplifiers used live and in studios.
For example, take your favorite rock band and check out their amplifier gear here:

Equipboard

Equipboard is the world's largest community of artists and their gear. All crowd-sourced and built by you.

equipboard.com

It seems like 99% guitarists use tube amps in 99% of records we listen to daily.

 

[Yevhen]

Do you have some evidence that the majority of musicians really prefer tube amps for audio reproduction (not guitar amps)? I heard that musicians often don't care much about sound quality of stereo systems at all.

They only care about the sound that goes as the feedback to their ears, something which happens NOW. Otherwise, you cannot control your guitar, and for them, the transients are much more important than the frequency response and distortion. And I think that's why all the guitar amplifiers are tube-based and all the guitar speakers are 1-way full range.

 

[LTig]

They only care about the sound that goes as the feedback to their ears, something which happens NOW. Otherwise, you cannot control your guitar, and for them, the transients are much more important than the frequency response and distortion.

Nope. It's the distortion characteristic of the tube amp (soft clipping with lots of lower harmonics) which sounds nicer than a clipped solid state amp (hard clipping with lots of higher harmonics). For audio reproduction this behaviour plays no role as amps must not be clipped in the first place.

And I think that's why all the guitar amplifiers are tube-based and all the guitar speakers are 1-way full range.

Guitar speakers are also driven into clipping which adds a specific sound. You cannot do this with a 2-way design as it would kill the tweeter.

Lots of modern guitar amps are solid state based and use DSP based amplifier modeling, where the distortion characteristic of many tube amps and their speakers is created on the fly, at much lower SPL, which is one reason why modern stage sound is much better today.
When rock bands show lots of guitar and bass speaker cabinets on stage today you can bet that most of them are not connected or are even dummies.

 

[Yevhen]

Nope. It's the distortion characteristic of the tube amp (soft clipping with lots of lower harmonics) which sounds nicer than a clipped solid state amp (hard clipping with lots of higher harmonics).

I've spent 10+ years (as my hobby) designing transistor-based guitar pedals and trying to get close to the sound of tube. You can make any kind of soft clipping if you put different types of diodes into the feedback loop, play with the type of clipping diodes, serial-parallel connection etc. In SPICE model, they will behave identical to the ECC83 tube. Then you give it to the experienced guitarist and they all say exactly the same: the "tone" (freq response and harmonic content) is indeed identical but the "attack" (transient response when going from quiet to loud) feels somewhat different.

There are indeed tonnes of the DSP based gear, but I don't see it on the list of the most famous/experienced guitarists (equipboard.com)

 

[LTig]

I've spent 10+ years (as my hobby) designing transistor-based guitar pedals and trying to get close to the sound of tube. You can make any kind of soft clipping if you put different types of diodes into the feedback loop, play with the type of clipping diodes, serial-parallel connection etc. In SPICE model, they will behave identical to the ECC83 tube. Then you give it to the experienced guitarist and they all say exactly the same: the "tone" (freq response and harmonic content) is indeed identical but the "attack" (transient response when going from quiet to loud) feels somewhat different.

Did they know what they used, comparing SS with tube gear, or was it a blind test?

 

[Yevhen]

Did they know what they used, comparing SS with tube gear, or was it a blind test?

No, it wasn't blind. But if you are interested, you should try the tube-based guitar amp in some music store. It definitely has something. I don't know how to describe it properly

 

[LTig]

No, it wasn't blind. But if you are interested, you should try the tube-based guitar amp in some music store. It definitely has something. I don't know how to describe it properly

Sorry, I play drums only. But I know how a good tube amp sounds. I also know what modeling can do for keyboards (Leslie effect).

 

[vuittion]

Même pas, John Bachel ! Presque tous les amplificateurs de puissance haute fidélité ayant une entrée asymétrique sur RCA ont une sensibilité d'entrée qui ne posera aucun problème en sortie de ce Cambridge dont le 1,6 volt délivrera leur puissance maximale à ces amplificateurs.

Use English.

 

[Haskil]

Use English.

Hello dear gentleman who thinks he is a moderator, it was a mistake...
Is kindness an option for you?

PS : Seven pages later...