The most critical component in a system, the power amp?

[hochopeper]

I usually buy 50K, 100k and 250k stereo versions. 

 

Why those values?

[mwhouston]

General I put my pots on the output of preamps. That's becaude they are noisy and put them I front means the noise goes through the preamp and gets amplified. If I'm making SS preamps 50K pot is fine. For my single stage tube preamps 100K to 250k is required. These preamps generally control tube power amps with upto 470K input impedance. Single stage tube preamps may sound great but have little drive power.

[Addicted to music]

Why those values?

 

you should be able to give us the answer here... :huh:   You're the SNA electronics engineer in these neck of the woods....  

 

When I was lots lots younger at your age @hochopeper  I built 2 preamps with identical parts but 9mths apart.  the 2nd 1 destroyed the 1st in SQ.  After all the parts swapping, it was a worn out volume pot.  For the next decade I tried and tested out that many pots and even tried to do a DIY elector project where recording studios knew that there equipment was let down by the humble ancient electromechanical/mechanical pot!!!!!  I came to the conclusion that all typical humble electrical/mechanical wipers no matter what you paid for them deteriorate regardless. 

 I also convinced Silicon Chip when they came out with there Ultrafedility OPA2134 based opamp preamp in designing a circuit for digital volume.  Unfortunately by the time there article was published the WM digital chip they used that was used in top of the line Marantz amps  was no longer produced and no longer available. 

So in 2012 i got on SNA and wella!   Theres a thing called a NAD M51........havent looked back!

[Addicted to music]

General I put my pots on the output of preamps. That's becaude they are noisy and put them I front means the noise goes through the preamp and gets amplified. If I'm making SS preamps 50K pot is fine. For my single stage tube preamps 100K to 250k is required. These preamps generally control tube power amps with upto 470K input impedance. Single stage tube preamps may sound great but have little drive power.

 

If you have a look at the Silicon Chip Ultrafedilty preamp thats  OPA2134 based, they used a 10K-20K ohm pot after the 1st stage controlling the 2nd OPA opamp as a buffer.  The disadvantage of this is that they used electrlytic caps on either side of the pot to eliminate DC so you dont get that srapping sound when it gets rotated.

 

To elimiate noise the shaft and housing of the pot must be grounded, also using lower values such as the SC opamp based preamp uses that in there quest to get academic noise figures.

Edited March 5, 2015 by pchan

[Malcolm]

Speakers

[davewantsmoore]

The equaliser?  

 

 

Indeed.

 

 

The speaker is the biggest source of (unintentional) frequency response non-linearity in a system.

 

Not only are the errors are orders of magnitudes higher than any other linear distortion in the system ..... but the error is different for every possible angle of radiation  (speaker don't radiate the sound equally in all directions).

 

 

Speakers are much more difficult to test both subjectively and objectively than other components in a playback system ..... and so they are often paid the least amount of "attention" by people.

 

 

 

TO go back to your comment..... If people want to put intentional frequency response changes into their system (ie.  with an equaliser).... than all power to them 100%.     It's ones people aren't aware of that they need to be worried about.

Edited March 5, 2015 by davewantsmoore

[hochopeper]

you should be able to give us the answer here...    You're the SNA electronics engineer in these neck of the woods....  

 

When I was lots lots younger at your age @hochopeper  I built 2 preamps with identical parts but 9mths apart.  the 2nd 1 destroyed the 1st in SQ.  After all the parts swapping, it was a worn out volume pot.  For the next decade I tried and tested out that many pots and even tried to do a DIY elector project where recording studios knew that there equipment was let down by the humble ancient electromechanical/mechanical pot!!!!!  I came to the conclusion that all typical humble electrical/mechanical wipers no matter what you paid for them deteriorate regardless. 

 

 

If you have a look at the Silicon Chip Ultrafedilty preamp thats  OPA2134 based, they used a 10K-20K ohm pot after the 1st stage controlling the 2nd OPA opamp as a buffer. 

 

I'm not great at analogue circuit design so that was actually an honest question ... what is the requirement of those circuits that mwhouston was building that required the use of a higher value pot ...

 

 

What I know is similar to the SC logic ... lower values are better for resistances due to reduced Johnson noise etc.

 

I've only got two builds with analogue volume controls - one is the O2 headphone amp built to the BOM and another I used a 10kohm goldpoint stepped attenuator between DAC and headphone amp.

[davewantsmoore]

the amp is, indeed, the most important.

but without the right cables, you are only half way there.

 

I find it difficult to believe you have A vs B tested different speakers, and found that there was less difference than when A vs B testing different amplifiers on the same speaker.

 

Or... if you did......   there was something wrong with your method... or you were using an amplifier which was "faulty", or you were using speakers which were very very similar.

 

 

If none of these caveats were truly at play.... then I'd be very interested to hear more about your experience....  as it might inform speaker design.

[joz]

IMO go from back to front. (speakers first)

Yes GIGO,

But assuming everything is better than junk, I'll stick to my rationale.

Edited March 5, 2015 by joz

[David.M]

Speakers

+ room

[davewantsmoore]

Power amplifiers that work well in conjunction with the chosen speakers and can drive them adequately, are an important link in the chain.

 

Sure, it can be important.   However.

 

If you test 5 different amps on a certain speaker.... you may or may not get different sound ....  it depends primarily on the design of the speaker.

 

... but if you test 5 different speakers, all using perfectly matched power-amps .....  You will almost always get a much larger difference in performance between each than you would in the former scenario.

[davewantsmoore]

It's the source and the way it was recorded and engineered

 

Most definitely.

 

... I was not considering it as a "component" of the system.

[Batty]

I believe in the chain approach, I bought amp and speakers at the same time, waaaaaaaaaaay back in 1995, I spent quite a few hours instore listening and swapping until I chose the Castle Howards and Copland CSA-14 that are in my system to this day.

[Guest Eggcup The Daft]

In the 80s, I did the whole source first thing, with Kef Coda IIs, and I have to admit that it did work (the local burglars never stole speakers!) Since then, I've come to the belief that source first was about getting "a prize every time" rather than the source being the most important component.

What we were really doing in those days was holding the speaker/room interface constant. We put the budget speakers in a room and they more or less worked, partly because of no deep bass. Then upgrading the source brings a small improvement each time. Then upgrading the amp, again a prize every time. And we had a good system when it came to upgrading the speakers, which was always the biggest problem as I remember it - because the room/speaker interface had to be got right all over again.

So the answer to the question is that the most important thing is the room/speaker interface. Get that right and you'll have a good system.

[David.M]

All wrong.

 

It's the source and the way it was recorded and engineered.

 

I don't care what sort of system you have or how you like to store and reproduce it (vinyl, CD, DVD, Bluray etc).

 

If this link is crap, the rest will be crap.

 

End of story.

In general I would disagree.  If you have the system/room right I think you can make all but the very worst masterings/recordings quite listenable and enjoyable, the bad ones you just can't play as loud (say averaging 90db).  I don't have a great collection of loudness war CDs but played at moderate levels they still sound quite good, turned up the relentless volume, high pitch vocals & high freq harshness gives me a headache very quickly.  Average quality recordings such as VU/Layla (LPs) I can play as loud as any 'audiophile' recording without getting fatigued, they certainly don't sound like audiophile records but still sound great.

[b.d]

Sure, it can be important.   However.

 

If you test 5 different amps on a certain speaker.... you may or may not get different sound ....  it depends primarily on the design of the speaker.

 

... but if you test 5 different speakers, all using perfectly matched power-amps .....  You will almost always get a much larger difference in performance between each than you would in the former scenario.

Ya I agree. I suspect if the OP finds speakers sounding fairly similar then it's possible he's comparing a number of similarly sized 2-way dynamic speakers. Compare some panels to some horns and line arrays and single drives etc and it'd be impossible to hear greater differences by swapping power amps (at least the vast majority that do have suitable output/dampening/impedance matching)

[Guest yamaha_man]

Those amps should have more respect, and not transported by dividing by a piece of rag and pack with a plasic drink bottle!   

3 beach towels stuffed between them champ, there is no way they were going to touch each other. 

[mwhouston]

Personally I would have used large bath towels. But as audio is a subjective hobby, each to their own.

Edited March 5, 2015 by mwhouston

[Audiobugged]

The person listening. By a zillion miles. Can't understand the shite some people listen to these days!

[mwhouston]

The person listening. By a zillion miles. Can't understand the shite some people listen to these days!

As with the towels; one man's shite is another man's gold. So what will it be, beach towels or bath towels?

Edited March 5, 2015 by mwhouston

[davewantsmoore]

Ya I agree. I suspect if the OP finds speakers sounding fairly similar then it's possible he's comparing a number of similarly sized 2-way dynamic speakers. Compare some panels to some horns and line arrays and single drives etc and it'd be impossible to hear greater differences by swapping power amps (at least the vast majority that do have suitable output/dampening/impedance matching)

 

Yes.

 

The crux of the problem is that speakers are for most people completely impractical to test in a way which would be considered a very robust test/method.

• If you have multiple speakers setup in your room, with a switch so you can flick between them .... you have them in different locations and angles.
• If you move the speakers in and out of location .... it takes too long.
• If you don't change the amplifiers AND the speakers, you have the potential to get something which is abnormal  (some speakers are amp sensitive and vice versa)
• Speakers sound different in every different room  .... because they throw sound in 3 dimensions .... other system components don't

 

 

That's all very prescriptive and scientific, I know ......  and in no way am I ignoring that all speakers are a compromise, and that people just like what they like, and be happy and move on   (music sounds great on most things IMO) .....   my point is that if we wanted to find out what "is best" ..... then it is challenging to do.

 

 

Luckily there are lots of people who work on this problem and share their (well documented and reproducible) results.    There's a century of research, and a lot less disagreement than some people might expect.

 

As a speaker designer, I've tested the research, and found it to be correct  (and offer quite profound results IME)

Edited March 6, 2015 by davewantsmoore

[mwhouston]

Yes.

The crux of the problem is that speakers are for most people completely impractical to test in a way which would be considered a very robust test/method.

• If you have multiple speakers setup in your room, with a switch so you can flick between them .... you have them in different locations and angles.
• If you move the speakers in and out of location .... it takes too long.
• If you don't change the amplifiers AND the speakers, you have the potential to get something which is abnormal (some speakers are amp sensitive and vice versa)
• Speakers sound different in every different room .... because they throw sound in 3 dimensions .... other system components don't

That's all very prescriptive and scientific, I know ...... and in no way am I ignoring that all speakers are a compromise, and that people just like what they like, and be happy and move on (music sounds great on most things IMO) ..... my point is that if we wanted to find out what "is best" ..... then it is challenging to do.

Luckily there are lots of people who work on this problem and share their results. There's a century of research, and a lot less disagreement than some people might expect.

As a speaker designer, I've tested the research, and found it to be correct (and offer quite profound results IME)

Generally rapid ABing doesn't really work long term. Living with a component for a few weeks or a month is the real test. Are you still happy with your purchase a month down the road? Or is there something missing? I have lived with my DIY 300B for years now and are extremely happy with it. I have had the new DIY Class D in for two weeks and though I think it's a marvel how will I feel two more weeks from now or two months from now. I'm already thinking about putting back the 300B? Time will tell. I always leave new DIY gear in for months on end then go back to something else or try something else to see if it has stood the test of time. Edited March 6, 2015 by mwhouston

[davewantsmoore]

Generally rapid ABing doesn't really work long term. 

 

So do a longer term A vs B.

 

However, it is well documented that "audio memory" is quite short.... and if there is a long time between A and B, then you can't tell very well.

 

 

Generally if I am testing something, the first type of questions I want to answer are "is there a difference" .... and if so, "what is the rough magnitude of the difference".     Switching back and forth between A and B while sitting in the listening position is a good way to get a feel for that.

[mwhouston]

If a friend or acquaintance rings you and says "it's me" most of the time we immediately know who it is without them saying thier name. I think our auditory memory is better than we think it is.

We all examine gear in different ways, like our hobby, it is also subjective. I have been on the end of AB testing many times and most of the time the differences don't matter. That's becaude there is always a compromise. You gain one thing and lose another. No gear is perfect, it just suits us better for where we are now in our listening journey. Tomorrow, next week or ten years from now we start another journey. becoming tired of inplace gear usually means in the end our taste has shifted and we are now finding something else is exciting us. I guess it is both the beauty and curse of this hobby.

When demoing gear to others I never play the same track on the two pieces of equipment under test but similar tracks from the same CD, LP or whatever. Only ever three or four tracks on each CD and all very different. Generally the second time you hear the same track on another piece of equipment it will sound better. That's our auditory memory saying " I've heard that before, I know that and it makes me feel good, this time". We are being fooled by our clever minds.

It's like relationships. The first two dates are fun but after a long while with someone likes and dislikes become more obvious.

[davewantsmoore]

If a friend or acquaintance rings you and says "it's me" most of the time we immediately know who it is without them saying thier name. I think our auditory memory is better than we think it is.

 

Different thing.... a voice pattern is much more basic and fundamental ....  as opposed to analysing whether A sounded better (or even different) to B ... when A and B are quite close, and there is a time delay.

 

 

there is always a compromise.  You gain one thing and lose another. 

 

Not always true.