I think I have put forward some ideas in response to Dejan asking about IGBT. With the usual caution I suspect they might be the better solution if the stability is OK and protection can be what is inherent in the devices. I would use the Exicon FET's. The first transistors might be very exotic as they will not be required to give much current. The Cgate of the FET's at circa 700 pF will be of no consequence. The gain and distortion is almost exactly that of CE output stage shown. That is circa 0.05% distortion and gain circa 93%. MOS FET CS about 20dB worse and gain of circa 78%. JLH was right and they do have merit. The Toshiba ones never happened. We don't need them.
Anyway a nice journey and I think I did find some reasonable doubts which supports Otala. Hi Fi choice took this seriously in the late 1980's. The NAD 3020 was a considerable winner on their tests. 192 watts 2R transient. Sony 170 W 8R 2 watts 2 R transient. QED. Strangely the 5 times cheaper NAD sounded better. That Sony was the most hopeless amp I ever listened to.
If you have ever read Harvey Rosenberg's account of the Futterman's amp it is the usual story of heartbreak and dead ends. What H R didn't know he asked to write the book. He stated they really required 60 dB feedback to meet Julius's hopes. One has to realize that with many tubes Rp of 500 ohms is already very good. It isn't rocket science to see even 60 dB is modest if seeking modern standards of damping factor. It oscillated nearly always said Harvey. Even transistor people shy away from 60dB. It ruins the gain if nothing else. The Futtersman's is a all NPN in valves. 800 R speaker is how Philips did it.
Anyway a nice journey and I think I did find some reasonable doubts which supports Otala. Hi Fi choice took this seriously in the late 1980's. The NAD 3020 was a considerable winner on their tests. 192 watts 2R transient. Sony 170 W 8R 2 watts 2 R transient. QED. Strangely the 5 times cheaper NAD sounded better. That Sony was the most hopeless amp I ever listened to.
If you have ever read Harvey Rosenberg's account of the Futterman's amp it is the usual story of heartbreak and dead ends. What H R didn't know he asked to write the book. He stated they really required 60 dB feedback to meet Julius's hopes. One has to realize that with many tubes Rp of 500 ohms is already very good. It isn't rocket science to see even 60 dB is modest if seeking modern standards of damping factor. It oscillated nearly always said Harvey. Even transistor people shy away from 60dB. It ruins the gain if nothing else. The Futtersman's is a all NPN in valves. 800 R speaker is how Philips did it.
Just a comment here --- engineering often Needs to go beyond the basic requirement. We dont make an amp which inherantly can only just make it to 20KHz because that is the highest we hear (or thereabouts). Same for many other parameters. Simpleton point. But.
As speaker distortion continues to slowly come down, we can detect what we couldnt before and then the parameters get changed/lowered. What was a good number 10-20 years ago drops. Slew Rate is like that also... what was fine for LP and human hearing range is not sufficient anymore for SOTA systems we can listen to now.
For myself..... I want the best we can do and not the minimum needed. Lowest THD/IM, lowest noise and highest SRate etc. Just because we can. But in the end, and if done well, it always seems to sound great.
IMO. and not that something less cant sound good. Why not do the best you can within the budget you have? Some more than others (but then we have credit cards, too ;-). The best circuit design. The best pcb layout. the best thermal management. the best protection. Why not do the best you can without imposing limits?
THx-RNMarsh
As speaker distortion continues to slowly come down, we can detect what we couldnt before and then the parameters get changed/lowered. What was a good number 10-20 years ago drops. Slew Rate is like that also... what was fine for LP and human hearing range is not sufficient anymore for SOTA systems we can listen to now.
For myself..... I want the best we can do and not the minimum needed. Lowest THD/IM, lowest noise and highest SRate etc. Just because we can. But in the end, and if done well, it always seems to sound great.
IMO. and not that something less cant sound good. Why not do the best you can within the budget you have? Some more than others (but then we have credit cards, too ;-). The best circuit design. The best pcb layout. the best thermal management. the best protection. Why not do the best you can without imposing limits?
THx-RNMarsh
Last edited:
+10
Asked the same question around here for years , its always the little train that could mentality, coupled with chase your tail designs discarded decades ago. Its all been done before , cant get any simpler than that to figure out what works or not ...
Last edited:
I surmise If we ask for the ten best pre and amp with both SS and tubes here (meaning by people who have both knowledge and experience) : nobodies agree on the rank and the trade off. Not all are agree about Ncores for ewample when it comes to speak about "musicality" (a strange concept).
There is also patents and the best of the worlds are sometimes not known or just can be understood by serious tweakers... So best of the world is for the few (you are a lot in this thread but maybe fews in relation to the whole diyer communauty: the ones who are able to assembly and set up such complicate designs.)
Sometimes as it seems there is always some trade off to choose, even with the best of the world philosophy, I mean if second hand is not more affordable than DIY sometimes ?
There is also patents and the best of the worlds are sometimes not known or just can be understood by serious tweakers... So best of the world is for the few (you are a lot in this thread but maybe fews in relation to the whole diyer communauty: the ones who are able to assembly and set up such complicate designs.)
Sometimes as it seems there is always some trade off to choose, even with the best of the world philosophy, I mean if second hand is not more affordable than DIY sometimes ?
Last edited:
Hi Demian,
Your question appears practical. However, whose room size and speaker efficiency and spl are we going to use for the standard or limitation?
Lowest speaker Z that is realistic and calculate everything backwards.... gives design to worse case Amp needs. What is That number? What about peak levels... crest factoring?
What about higher number of paralleled output devices which give lower distortion, higher SOA and other factors... It is only a side issue that maybe it appears higher than needed output Amp capacity.
-RM
Your question appears practical. However, whose room size and speaker efficiency and spl are we going to use for the standard or limitation?
Lowest speaker Z that is realistic and calculate everything backwards.... gives design to worse case Amp needs. What is That number? What about peak levels... crest factoring?
What about higher number of paralleled output devices which give lower distortion, higher SOA and other factors... It is only a side issue that maybe it appears higher than needed output Amp capacity.
-RM
Last edited:
really don't know, some are driving 93 DB speakers with 7 w amp in a cathedral as over 250 W with 85 db in 35 squaremeter as me with a Chord amp SPM 1200B??? I heard about a concept of current capacity than more Watts itselves. can't understand anymore the difference between Karan concept and Papa's 7 W here ???? easy to be lost also when some said just tubes can dive you into music....
Let say around 90 DB for 25 squaremeter which seem to be a majority here in Europe, don't know about USA...
But myself want to go around futur 94 DB speakers just because the best amp seem to have low wattage : tubes, Papas's...Hiragas', and so on.
Edit : oups RN.... Thanks, doesn't see all your answer because of corrections... With all these factors, we can be lost ! there is in this world no two bigger differents things than two speakers or amps ! As different as a human being and an insect with a same basic requirement to define both : they live ! Hifi seem more and more strange to me !
Let say around 90 DB for 25 squaremeter which seem to be a majority here in Europe, don't know about USA...
But myself want to go around futur 94 DB speakers just because the best amp seem to have low wattage : tubes, Papas's...Hiragas', and so on.
Edit : oups RN.... Thanks, doesn't see all your answer because of corrections... With all these factors, we can be lost ! there is in this world no two bigger differents things than two speakers or amps ! As different as a human being and an insect with a same basic requirement to define both : they live ! Hifi seem more and more strange to me !
Last edited:
even if you had magic slewrate amp, IMHO still does not exist such speaker to faithfully reproduce square signal (inertia,mass,...)
or been invented some audiophile magic grease for speaker cones, which improves this ?
Peter, all this talk of slew rate is reated to the amp only, making sure it really, really never trips up on this. Working out what is the least satisfactory requirement.
In a previous discussion on this subject, the least satisfactory limit was 1 V/uS for every peak volt of output. That would be 40 V/uS for a nominal output for a 100W/8 Ohms power level.
As you can see for yourself, in the meanwhile our appetites seem to have grown, now Richard is not sure even 100 V/uS is quite enough.
Regarding posts by Richard and Eldam, I'd just like to expand on them a bit.
On a global level, we seem to have two general categories - home audio and professional audio.
Home adio can further be regarded as ROUGHLY three tier - Low Fi (Cheap'n'Cheerful), Mid-Fi (starting to become serious, but at a higher cost) and High End, where the sky's the limit for both performance and price.
Pro audio can also be regarded as at least two tier - sound reinforcement and studio gear. In Pro audio, reliability is God, everything else takes a back seat to it, because for a lot of the time, that gear will be pushed hard, way harder than 99.9% of home gear.
I mention this only because different modes of use dictate priorities. Home High End echews most protection circuitry in favor of nominal sound purity, while in Pro sudio if your protection is not really top notch, you don't sell because your amps tend to burn out. Unfortunately, this makes comparisons rather unreliable and not really fair, since each approach has its pros and cons.
It's a bit like Wayne and me. He has extremely demanding speakers, while I have extremely undemanding speakers, so what migh be just great for me may not cut it with him. And, just to spite ol' Wayne, anthing that works with his speakers will probably work even better with mine.
As Eldam quite rightly points out, one man's meat is another man's poison. What I might just love for what I perceive as its nezrality, someone else might consider as without color, dead and boring - or vice versa. I have been through that loop many a time, after a while people get used to their sound at home and become less and less tolerant of any differences from it.
On a global level, we seem to have two general categories - home audio and professional audio.
Home adio can further be regarded as ROUGHLY three tier - Low Fi (Cheap'n'Cheerful), Mid-Fi (starting to become serious, but at a higher cost) and High End, where the sky's the limit for both performance and price.
Pro audio can also be regarded as at least two tier - sound reinforcement and studio gear. In Pro audio, reliability is God, everything else takes a back seat to it, because for a lot of the time, that gear will be pushed hard, way harder than 99.9% of home gear.
I mention this only because different modes of use dictate priorities. Home High End echews most protection circuitry in favor of nominal sound purity, while in Pro sudio if your protection is not really top notch, you don't sell because your amps tend to burn out. Unfortunately, this makes comparisons rather unreliable and not really fair, since each approach has its pros and cons.
It's a bit like Wayne and me. He has extremely demanding speakers, while I have extremely undemanding speakers, so what migh be just great for me may not cut it with him. And, just to spite ol' Wayne, anthing that works with his speakers will probably work even better with mine.
As Eldam quite rightly points out, one man's meat is another man's poison. What I might just love for what I perceive as its nezrality, someone else might consider as without color, dead and boring - or vice versa. I have been through that loop many a time, after a while people get used to their sound at home and become less and less tolerant of any differences from it.
Slew Rate in Audio Amplifiers - What Does it Mean? | Audioholics
I was reading up on Douglas Self and slew rates. A very interesting piece of advice. Never test at full power at high frequencies for extended periods of time. Well I always do and as far as I can see it should make no difference. Am I lucky or what?
I was reading up on Douglas Self and slew rates. A very interesting piece of advice. Never test at full power at high frequencies for extended periods of time. Well I always do and as far as I can see it should make no difference. Am I lucky or what?
Tremendously depends on your nominal load, Nige. People get carried away and hook up what they feel are difficult loads, which often results in power stage burnout. So instead of their prowess, they demonstrate their ignorance.
And yes, you have been lucky - don't push it, old son.
I consider this PIM stuff a dead end, a waste of time that will amount to nothing. Dvv with all due respect to your design accomplishments, you are fooling yourself if you think wide OLBW alone matters.
Hi RN,
Sound like active multi amps... but question remains between 7 w amps and 30 W class A (not anymore 250 W if we go to multiple amps and active if I understand what you write above)! Think the parameters seem about the speakers the ones needed for a more accurate listening experience : at least great dynamic between the lower and the higher signal... that help also not to put the level higher and save our ears from a highDB output with a low dynamic... So 120 db peaks for Chet Baker trumpett peaks but with low middle listening level !
Even with those parameters a 845 tubes is different from an 30W class A Accuphase (for the demonstration I mean !)
I see also people who construct clone of famous fast swiss amp of the 80's... which sound horrible to me but gave maybe of the price a feeling of confort !
Last edited:
I never said it's all that matters, Scott, I mentioned it more in passing to explain why I chose relatively small global NFB factors of 20 +/-1 dB as one of my goals.
I don't see slew rate as a critical factor for sound, assuming the basic requirement of 1 V/uS per peak volt output is met. I learnt that from your field, op amps. Playing around with them, I realized that the venerable OPA 37 op amp actually sounded better than a number of op amps which had as much as ten times its nominal slew rate (it's rated at 15 V/uS and is the uncompendated version of OP27, its only possible drawback being that it requires a gain of a minimum of 5:1 for stability).
Just to reaffirm that, OPA 275 (Butler FET/BJT front end, rated at best at 22 V/uS, if memory serves) in my book usually beats a lot of other far faster op amps on good sound.
Thanks for the answer, you may be correct, it certainly doesn't sound clean.
Nonetheless it was VERY educational for me to find square-waves in Foobar. For instance, if I play with the equalizer just a little, the entire square-wave can fall to pieces.
That's why some headphones can have such 'horrendous' looking square-waves, even if they have 'perfect' square-waves, since it's simply FR deviations we're looking at which fiercly convulute the wave to the naked eye.
Last edited:
My little amp is 8 watts. I never really needed more. My recent OB speakers needs less than 1 watt to give me ear damage. I have to discipline myself as I get close. Even at +16 db 30 Hz EQ there is enough power. I spent months on that amp to ensure the maximum was usable. I have tried a number of amps with 300B and would say 5 watts to be realistic, the last 3 watts with obvious distortion. The much cheaper EL 34 can give more. To my ears it sounds better. My OB speakers use an Eminence 12Lta which is dam close to being excellent. I have Magneplanar SMGa as reference. It is hard to say which I prefer. The latter needs more power.
I sometimes use a 150 watt amp. It allows me greater choice of speakers. It might be nice to bridge it sometime for the fun of it.
The SMGa claim to be 4 ohms pure resistive. When I clip amps with protection circuits nothing drastic happens. The load is nice.
Spencer Hughes said something to me. " People say the Spendor BC1 is a nice load. If I were an amplifier I might say. I have looked down these cables and I see no speakers. All I see is capacitors and chokes. I think I will stay at home as that doesn't look nice". Not bad for a BBC man. I was 20 I guess?
I sometimes use a 150 watt amp. It allows me greater choice of speakers. It might be nice to bridge it sometime for the fun of it.
The SMGa claim to be 4 ohms pure resistive. When I clip amps with protection circuits nothing drastic happens. The load is nice.
Spencer Hughes said something to me. " People say the Spendor BC1 is a nice load. If I were an amplifier I might say. I have looked down these cables and I see no speakers. All I see is capacitors and chokes. I think I will stay at home as that doesn't look nice". Not bad for a BBC man. I was 20 I guess?
- Status
- Not open for further replies.