Back when I didn't give much thought about the cost of electricity I built a Son of Zen class A 2x10W. Sounded really nice on my Fostex FE206 horns. It didn't sound too well when not on constantly though always taking an hour at least to settle itself, so I left it on all the time like so many people do.
That was until I heard a Tripath TA2020 amp, the amp6b, and boy was I "wow"ed. I switched back to the SoZ once, and then sold it. It had lost it's magic being out-performed by this tiny little 2x10W class D amp.
I quickly discovered that the amp6 sounded best on battery, so I got one of those and a 14W solar panel. In 5 years, I have had to recharge the battery once due to a mistake. So my system is completely off-grid.
5 years later, today, I just calculated that my new system has saved me about $3000 in that time. Electricity here costs $0.35/KWh. And the SoZ uses about 200W regardless if it's playing music or not. Adding that up becomes $600 per year.
I just may have to build a small class-A to see for myself. I am looking at a basic design like the PassDIY "Camp Amp" which looks small enough and is powered with a 19 v SMPS from a laptop. The big aluminum heatsinks and industrial rackmount cases have always been a detractor due to high cost. I will have to be creative with a heatsink. Needs a high efficiency speaker to go with though...
What were referring to as an outmoded cliche? Japanese vs Italian supercars?
Maybe I should have said Toyota (class D) vs Porsche/BMW/etc? Did you know Lotus uses essentially a stock Camry V6 in their $100k Evora sports car? The British gave up on making high reliability motors and concentrate on suspension and driving dynamics which they are very good at...
That's neat to hear first hand experience from someone who has used a fabled class A amp and now thinks a $20 TA2020 beats it!
The amp6 is $50 in kit form. Don't think it really compares with any of the "ebay" TA202x designs. Solar panel was about $150 and battery (20Ah LiFePO4) around $300. So about $600 in total if we include various parts that have been replaced and the 1F cap bank on the batteries.
(Note: Thin Film solar panels and LiFePO4 batteries have dropped a lot in price since then so it could probably be had for much less today).
Last edited:
thank you for comforting me...
I agree with the inefficiency of the speaker, in recent years we have solved the strength, weight (as neodymium), and little efficiency. I think it is inherent in the concept which we can not fly... expect a new invention hehe!
Best regards
Roberto
You may check out Jean Hiraga's Le Monstre that is what users tend to power on by car batteries (actually it is wrong belief that car battery outperforms Black Gates or that class A amp needs to be constantly on unless it's not a part of operational anti-missile defense
).
A more relevant comparison to cars would be that of a Ford GT40 (class A) and an Audi R18 TDI (class D). In it's day a Ford GT40 outperformed anything at Le Mans but it wouldn't stand a chance today. It's torque, BHP and top speed would be significantly lower but the weakest point would be the 7 liter engine that uses over 6 times as much fuel. In todays Le Mans fuel tank restrictions it would basically have to pit on every other lap instead of every 12-14th as the top cars do.
Last edited:
First of all, (sorry at all for this new argument by MrPush)
beg you a pardon? this is precisely related to the ongoing discussion, it's not sidetracking it.
first pic refer to musical transient in any time segment,as x-time show
what is the meaning of "in any time segment"? it doesn't convey any meaning in this context.
and not a sin input
definitely, but a 20kHz, full-scale sin input would be even faster.
as him said, the difference is visible in relation when it is fast (transient) , we can call "phase error" but is not this the problem.
this? the phase error, or what?
see angle of peak, structure of envelope changed, then musical instrument are modified.
angle of peak changed because of insufficient sample rate and linear interpolation. now, had you used sinc interpolation instead of linear... Nyquist turns in his grave.
and musical instruments are modified by an amp having non-linear distortion, only in a different way.
obvius that we listen mp3 also and?
obvious to someone who fancies believing it.
study is not important becouse we listen even a chip 1€?
IMO, no real meaning conveyed by this sentence.
This measures refer to just at start of study
did the study end, how?
i mean is not recent.
is the study obsolete?
A lot of job to realizze today a better integrated amplifier with special smps capable of deliver performances as audiophile expected.
hand waving.
again. can you prove that the problems seen in the pics I posted are not visible in the typical measurements (standard Stereophile measurement set)?
it fails me why you are avoiding the answers. I'm not asking for trade secrets.
beg you a pardon? this is precisely related to the ongoing discussion, it's not sidetracking it.
first pic refer to musical transient in any time segment,as x-time show
what is the meaning of "in any time segment"? it doesn't convey any meaning in this context.
and not a sin input
definitely, but a 20kHz, full-scale sin input would be even faster.
as him said, the difference is visible in relation when it is fast (transient) , we can call "phase error" but is not this the problem.
this? the phase error, or what?
see angle of peak, structure of envelope changed, then musical instrument are modified.
angle of peak changed because of insufficient sample rate and linear interpolation. now, had you used sinc interpolation instead of linear... Nyquist turns in his grave.
and musical instruments are modified by an amp having non-linear distortion, only in a different way.
obvius that we listen mp3 also and?
obvious to someone who fancies believing it.
study is not important becouse we listen even a chip 1€?
IMO, no real meaning conveyed by this sentence.
This measures refer to just at start of study
did the study end, how?
i mean is not recent.
is the study obsolete?
A lot of job to realizze today a better integrated amplifier with special smps capable of deliver performances as audiophile expected.
hand waving.
again. can you prove that the problems seen in the pics I posted are not visible in the typical measurements (standard Stereophile measurement set)?
it fails me why you are avoiding the answers. I'm not asking for trade secrets.
Last edited:
Hi,
I accept from you just that unfortunately, I do not have published throughout the study done entirely, so it can be confusing (especially for you).
everyone knows that traditional measures audio only serve to show what you want and do not say anything about the dynamic capability of an amplifier. only you persist with SIN signal.
When amplifier play a sound enveloope,need a very complex work, i'm sure you not know.
regards
I accept from you just that unfortunately, I do not have published throughout the study done entirely, so it can be confusing (especially for you).
everyone knows that traditional measures audio only serve to show what you want and do not say anything about the dynamic capability of an amplifier. only you persist with SIN signal.
When amplifier play a sound enveloope,need a very complex work, i'm sure you not know.
regards
no, it is confusing for any person who looks at it from a neutral stand point, and for people who are aware of the concept of "critical thinking".
no, I'm not aware that everyone knows that for a fact. obviously, a mere THD figure doesn't say much but OTOH I believe a more comprehensive "regular" set of test does.
yes, I do and you could prove me wrong. and I state it again: the errors that are visible in the measurements above should be visible in the regular tests + a Bode plot (as regular as it gets too but phase is rarely displayed, maybe because it can be derived from the magnitude response, as signal/control theory tells us).
what I do know is that you keep repeating stuff. maybe it works with certain people, maybe it happens that you're used to such people. I like to think I'm not one of them.
are you sure you know what the measurements are? you do know they represent an amplifier loaded with a real speaker right? the sort of thing that is known widely to still have readily measurable distortion when it comes to following the dynamic/transient input signal. unlike when loaded with a purely resistive load.
or are you really stating that speakers are driven without distortion? you really have not seen this distortion? distortion caused by the electro-mechanical and reactive load and the inability of the amplifier to follow the input + gain, due to imperfect control of the transducer.
a fast transient may contain rise time that is more consistent with a much higher speed. music is made up of a series of sines, but some of those will contain faster signals than the tonal content would suggest. I hope that makes sense.
or are you really stating that speakers are driven without distortion? you really have not seen this distortion? distortion caused by the electro-mechanical and reactive load and the inability of the amplifier to follow the input + gain, due to imperfect control of the transducer.
a fast transient may contain rise time that is more consistent with a much higher speed. music is made up of a series of sines, but some of those will contain faster signals than the tonal content would suggest. I hope that makes sense.
Last edited:
this is absolutely wrong. I have seen it repeated many, many times and for a while I believed it because, well, critical thinking is not something one is born with. this is why I'm not surprised.
band-limited signals (what exits any CD, DVD or SACD player) never contain those mythical sharp transients. never. and I mean never ever.
there is an interview with Nelson Pass in Stereophile where he says that very rapid slew rates are never present in real music and I think he mentions less than 1V/us.
still measuring the amplifier output, I dont mean measuring the speaker output. in order to extend the feedback to somewhat include the transducer it needs to have the speed to react to the deformations caused by ringing/overshoot caused by the speaker itself and any capacitance and parasitics and these corrective measures may contain much faster harmonic content than the audio.
its really NOT wrong
its really NOT wrong
I believe that tendency is visible in the frequency response plot as a small bump post 20kHz.
and I would add that what you're describing is not what is visible in the plots above. and for all we know that may be some disastrous 2N3055 implementation from the 70s.
AP2 could have something going on for him if he measured a few amps, both in the "regular" way and his time-domain way in order to show a lack of correlation between the results. but he didn't and he doesn't, instead just posts slogans.
and I could very well be wrong. but slogans don't equal proof.
Last edited:
20khz? erm, no.
I dont believe he can here on the forum, or at least its difficult as a manufacturer. some would rightly be considered promotion , some may not be 100% his IP to post publicly and some bad results may not be permissible if they are other company's product, certainly it would need to remain nameless what the amplifier was.
I believe it is exactly what hes talking about. Overall I feel your blow by blow nitpicking and comments on language are not terribly mature and not in any way productive. If I had done the work to investigate this, it certainly wouldnt motivate me to do some more work so you may also understand, the way you are acting.
Getting only half the story is frustrating, no doubt about that, but I really dont think Roberto Owes us anything, hes not selling us something here, sure he makes commercial amps and PSUs, but nothing has been specifically promoted, so I believe the only person he needs to satisfy for now is himself.
Maybe this is all smoke and mirrors designed to sell...erm...?, who knows, but some of the problems being discussed are very valid targets, so I encourage any study undertaken to understand it, whether its shared with us for freely and openly or not.
at the same time, Roberto, its pretty clear that you will not be permitted to theorize about this, actual measurements or nothing at all seems to be whats expected. I can understand that to a point, but it certainly cuts out a lot of quite interesting and fruitful conversation if thats required for everything a poster says, before they say anything.
I dont believe he can here on the forum, or at least its difficult as a manufacturer. some would rightly be considered promotion , some may not be 100% his IP to post publicly and some bad results may not be permissible if they are other company's product, certainly it would need to remain nameless what the amplifier was.
I believe it is exactly what hes talking about. Overall I feel your blow by blow nitpicking and comments on language are not terribly mature and not in any way productive. If I had done the work to investigate this, it certainly wouldnt motivate me to do some more work so you may also understand, the way you are acting.
Getting only half the story is frustrating, no doubt about that, but I really dont think Roberto Owes us anything, hes not selling us something here, sure he makes commercial amps and PSUs, but nothing has been specifically promoted, so I believe the only person he needs to satisfy for now is himself.
Maybe this is all smoke and mirrors designed to sell...erm...?, who knows, but some of the problems being discussed are very valid targets, so I encourage any study undertaken to understand it, whether its shared with us for freely and openly or not.
at the same time, Roberto, its pretty clear that you will not be permitted to theorize about this, actual measurements or nothing at all seems to be whats expected. I can understand that to a point, but it certainly cuts out a lot of quite interesting and fruitful conversation if thats required for everything a poster says, before they say anything.
Last edited:
qusp,
I think that you are cluing into a factor that is overlooked much to much in the amplifier speaker interface. All of the analysis of an amplifier and the actions of the feedback circuits are looked at in a vacuum without a real load hanging as a pole on the outside of the closed amplifier circuit. The roll of EMF from the speaker and the reactive component is just tossed out the window in the amplifier circuit analysis. This is what is screwing with the linearity that is expected from the NFB circuit, the feedback loop can not deal with this function, it is outside the loop with no real consideration of the complex interaction. In reality the speaker should be inside the feedback loop though I am not sure how you would achieve that goal. I tried in another thread just yesterday to give someone information on a conjugate LRC tank circuit attached at the speaker terminal to control the complex phase relationship that should be addressed before the crossover network is designed and it just isn't accepted that a simple Zobel network does not do the trick. The LRC tank circuit can be adjusted for Q that the simple Zobel can not, while the zobel does take care of the simple part of the impedance rise it is not a match to the real complex function of the speakers complex phase response and the back emf output that is skewing the NFB circuit of the amplifier. When something is outside of the normal discrete way we look at each section of the audio chain it is just left out or discounted. This is going to cause problems with any amplifier whether they are class A, AB, D, G, or H. The outside pole function of the loudspeaker is just looked at as a separate uncontrollable issue that is beyond consideration in the amplifiers need for consideration.
I think that you are cluing into a factor that is overlooked much to much in the amplifier speaker interface. All of the analysis of an amplifier and the actions of the feedback circuits are looked at in a vacuum without a real load hanging as a pole on the outside of the closed amplifier circuit. The roll of EMF from the speaker and the reactive component is just tossed out the window in the amplifier circuit analysis. This is what is screwing with the linearity that is expected from the NFB circuit, the feedback loop can not deal with this function, it is outside the loop with no real consideration of the complex interaction. In reality the speaker should be inside the feedback loop though I am not sure how you would achieve that goal. I tried in another thread just yesterday to give someone information on a conjugate LRC tank circuit attached at the speaker terminal to control the complex phase relationship that should be addressed before the crossover network is designed and it just isn't accepted that a simple Zobel network does not do the trick. The LRC tank circuit can be adjusted for Q that the simple Zobel can not, while the zobel does take care of the simple part of the impedance rise it is not a match to the real complex function of the speakers complex phase response and the back emf output that is skewing the NFB circuit of the amplifier. When something is outside of the normal discrete way we look at each section of the audio chain it is just left out or discounted. This is going to cause problems with any amplifier whether they are class A, AB, D, G, or H. The outside pole function of the loudspeaker is just looked at as a separate uncontrollable issue that is beyond consideration in the amplifiers need for consideration.
1V/usec slew rate is like a 1 MHz risetime for the signal. Isn't that much faster than 20 kHz sine wave component?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.