Hi,
Alas, I do not see a SINGLE Amplifier in this quartett which precisely doubles it's output power from 8R to 4R for 0.1% THD.
So based on your evidence so far there is no Amp that "doubles down" under the conditions under which the Citation XX is specified.
Basically another case where the vendor with the honest specs gets beaten up over them while the BullS...t merchants are lauded as paragons of performance due to their less transparent specs...
I guess if HK had rated it at 250W8R/500W4R @ <0.5% THD (which they could likely have done) all would be well...
Ciao T
Alas, I do not see a SINGLE Amplifier in this quartett which precisely doubles it's output power from 8R to 4R for 0.1% THD.
So based on your evidence so far there is no Amp that "doubles down" under the conditions under which the Citation XX is specified.
Basically another case where the vendor with the honest specs gets beaten up over them while the BullS...t merchants are lauded as paragons of performance due to their less transparent specs...
I guess if HK had rated it at 250W8R/500W4R @ <0.5% THD (which they could likely have done) all would be well...
Ciao T
Hi,
George may have been Bill (Withlock).
Given that it's the AES (and who it's leading lights are) I though proof of a double humorectomy was required before they let you in? How did you two sneak in...
Ciao T
George may have been Bill (Withlock).
Given that it's the AES (and who it's leading lights are) I though proof of a double humorectomy was required before they let you in? How did you two sneak in...
Ciao T
Hi,
All of this shows you how a devices parasitics change if voltage cross or current through the device change.
They allow you to derive the various parasitic resistance (early effect induced limited collector impedance, Beta changes, Changes of the internal emitter and base impedance as well as the parasitic capacitances), so they allow to tell you what goes wrong and where if you change voltages and/or currents (which you will as a transistor that works only static is not useful in an amplifier)...
Of course you can just blissfully ignore all the non-linear parasitics and throw a ton of looped feedback around the Amp instead...
That could be related to the fact that Motorola Semiconductors was spun off and is now called "On-Semi", just as Philips Semiconductors is now NXP and Hitatchi Semiconductors is now Renesas...
Ciao T
All of this shows you how a devices parasitics change if voltage cross or current through the device change.
They allow you to derive the various parasitic resistance (early effect induced limited collector impedance, Beta changes, Changes of the internal emitter and base impedance as well as the parasitic capacitances), so they allow to tell you what goes wrong and where if you change voltages and/or currents (which you will as a transistor that works only static is not useful in an amplifier)...
Of course you can just blissfully ignore all the non-linear parasitics and throw a ton of looped feedback around the Amp instead...
That could be related to the fact that Motorola Semiconductors was spun off and is now called "On-Semi", just as Philips Semiconductors is now NXP and Hitatchi Semiconductors is now Renesas...
Ciao T
Hi,
Your comments make me think you need a lot of power.
In this case the Plinius looks like it has the most distortion of all of them.
Maybe you just like distortion?
Ciao T
Your comments make me think you need a lot of power.
In this case the Plinius looks like it has the most distortion of all of them.
Maybe you just like distortion?
Ciao T
Hi,
George may have been Bill (Withlock).
Given that it's the AES (and who it's leading lights are) I though proof of a double humorectomy was required before they let you in? How did you two sneak in...
Ciao T
No, I know Bill. And this was no Bill
Hi,
No, this was "on line audio discussion standard reply # 1" when someone posts they preferred a Tube (High Distortion) Amp over a Solid State (Low Distortion) Amp...
Ciao T
So distortion measurements matters ...........
No, this was "on line audio discussion standard reply # 1" when someone posts they preferred a Tube (High Distortion) Amp over a Solid State (Low Distortion) Amp...
Ciao T
I've seen that in another circuit, with a footnote explaining that it's optional to use either the two diodes or the LED. Maybe that's the idea here too.
Yes, this would be a "typical" Bongiorno circuit, with a few points ironed out. A classic fully complementary topology.
Not my first, both my solid state headphones amps use a simplified version of this same topology, and, in my view, make a good job of it.
The buffer also uses the same topography, but is much simpler of course. Heck, even my voltage regulators (100% discrete, of course) could be said to use the sam topology.
While I recognize the fact that op amps CAN be made to sound very good, especially the newer generation of fast ones, I still prefer discrete. Can't help it, I'm an old fashoned guy. Op amps are what they are, while with a discrete topology, I keep full control over the sound from start to finish.
As for James Bongiorno, I have no idea what his present Ampzilla 2000 looks like topologically. But I must yet again praise his patience and goodwill in answering all questions I asked of him. I did try to square things with him by sending him a filter as a gift, I felt that was only fair. His response was to try it out, tell me: "It works." and pass it on to Bascom King for spin. I understand they are good friends. It turned out Mr King was far more interested in my loudspeakers, and he made me promise that if a pair should reach USA, he would get a spin around the block with them. Unfortunately, the company which made them suffered a heavy loss in that one of its two owners, the financial part one, suffered a horrible motorcycle accident in what was to be his last race, and all he can move today is his little finger of his left hand. He has to fed every 20 minutes because his stomach can't hold the food, he has to be wheeled around and so forth - a terrible, terrible fate. And at a time when they just started exporting it to Italy, Belgium and Holland. The company unfortunately fell apart, events like that change the human psyche more profoundly than in any Hollywood movie.
Anyway, John, a straight question - do you see anything in its topology that you would prefer done in another way, and if so, what (aside from the output transistors)? Never mind the cost of it, this is for me, not for sales.
Thank you, T., now there's some food for thought.
Oddly enough, I still have about 60 2SK170 FETs on stock, but no 2SJ74s ...
Of the FETs available to me, I have the Euro series (BC and BF) and 2SK 241/246/BJ30ATM (???), and of the P FETs only 2SJ103. As you can see, a VERY narrow selection. EXTREMELY limited.
But I am not opposed to hunting down something else if I knew what to look for.
Fair enough - what would you throw out and what would you replace it with? I'm open to all suggestions.
As for symmetrical topologies, they suffer from a logical fault - they will work very well indeed only if the transistors used are as well matched as possible. Since selecting and matching them is a lengthy and tedious process, and is consequently relatively expensive, it's not often done as well as it should be. And you hear it.
I experimented with this effect with my headphone amps. I built one by randomly pulling transistors our of a box, and then I built another by matching them down to a difference of less than 2%. Two different things. The first sample sounded loose and not quite in control, the second sample was pinpoint sharp on the dot, and offered much more ambience and spatial clues. At the cost of a whole day of just matching the transistors. Factories simply can't afford this luxury.
Yes, true , then there is 2 ohm ....
Look at the graphs again T, all 3 tell a different tale, the goldmund and the electro really fall apart after 8 ohms , unlike the plinus and I'm not surprised I found the Plinus the best sounding ...
anyway put a bid in on the XX...........
Wayne, you support a theoretical ideal. Yes, it IS important for an audio power amplifier to be as load insesitive as possible, but one should be realistic about how one determines this, as a point of a theoretical ideal, or as a point of realistic use.
NOBODY sits as thome and listens to pure sine waves at full rated power; nobody can, because even a very small transient would probably drive such an amp into clipping. You need to allow AT LEAST 6 dB headroom for real world music signal.
In other words, you buy/make a say 100 wpc amp to actually use it at 10-20 Watts of CLEAN power on any continuous basis. This also depends much on your spekers, your room, etc; my rook is small, and by about 2-3 Watts RMS, my window panes are dancing to the music. And the needles on my Marantz 170 DC power amp, imprecise as they are, are hardly even moving. At around 10 wpc, music has long become unpleasantly loud.
All of which means that in the real world, we actually use rather low power levels for real music reproduction. That's how Thorsten's arguably low powered tube amps manage to work very well indeed in real life, as opposed to theoretical models. What saves their day is their large overload margin, so a nominally say 10 wpc amp will actually distort but not clip up to say 30 wpc.
Which brings us back to higher powered amps. Yes, I do want it to be able to deliver a minimum of 28.3 Vrms into 8 and 4 Ohms nominal (or 7 and 3 Ohms actual) loads, and yes, I do want it to have sufficient headroom to negotiate even 2 Ohm loads, but that's my quirk, in reality it would be quite sufficient for it to be able to handle short term peaks into 2 Ohms. IEC standards define these short term bursts of power as lasting only 20 milliseconds.
To be bale to deliver 28.3 Vrms into 2 Ohms, it has to be able to deliver around 30 Vrms into 8 Ohms, because to the best of my knowledge, nobody has ever produced a power amplifier which can hold exactly 28.3 Vrms in to 8, 4 and 2 Ohms. Lower the load impedance and your output voltage goes down, period. The question is only by how much.
Here is where fully regulated power supplies do their best. They make sure that this difference in output voltage stays very low, if done right. They also make sure the supply lines are VERY stiff, meaning they react next to nothing in high power transients - ideally, a regulated power supply should exhibit no more than 0.1% od voltage difference between no signal and full signal on states. In the real world, this is not easy to achieve, but it is possible.
Such very stable supply lines make for a very stable amplifier. It's not unusual for amps with regulated supplies to sound bigger and more powerful than they actually are. Naim had a 30 wpc model in the 80-ies with full regulation and it was very hard to believe that it had only 30 wpc.
On the other hand, SAE had a nominally 200 wpc into 8 Ohms amp which, according to what's written on its backside, could manage only HALF that power into 4 Ohms, at a not at all insignificant price.
Bigger is not always better.
Of course he made his TIM (also hardly that new an idea) into a virtual industry[snip] But the combination of getting you into trouble with h/k AND ripping your designs off is unconscionable
[snip].
Personally, I feel that the practise of publicly badmouthing persons that cannot reply to give their view of the story, unconscionable
jan
I would like to set the record straight, so put this on record.
In my life, I have published 326 texts in magazines, I have published a book which sold out in two editions (the third being aborted by local wars), I have authored and anchored my own TV show for three seasons, I have sunp off another TV show from my original one, I have hed my own radio show regarding audio for four seasons, and I have implemented some of my ideas with people who were willing to put the time and effort into it - such as my own loudspeakers, for example.
And over the years, I have seen whole parts of my texts incorporated into other peoples' texts totally in violation of copyrights, I have even seen entire texts of mine being reprinted without permission.
With all due respect to one and all, but I know far better what it is to be ripped off than most people here. Furthermore, most of you live in ordered societeies in which laws are at least partially observed, wheras I live in country in which laws mean literally nothing "if you know the man". Out here, politics is everything, and if you're connected to any party in the ruling coalition, you need not worry about legality - anything goes.
To say that I personally am deeply ashamed because of this is the least you could say, in which case I would not, under any circumstances, rip anyone off. The instant I posted the schematic and asked my question, the moniker describibng the authorship of that amp increased no end. Before posting, it read just as "By DVV"; the instant somebody responded to may questions, uit changed to: "By DVV, with input from Thorsten Loesch, .... etc, on DIYaudio.com site". Heck, I make my friend of 20+ years who does the PCB artwork for me sign himself on those boards.
I am extremely sensitive to the issue of authorship because I've been burnt many times. And all it took was just a short note asking for permission, I am always inclined to give it. I feel all knowledge (I credit myself with some small parts of it) cannot reach its full potential until it is shared, and I cannot know how you feel about a project until I share it.
I fully understand John's position, I've been there, I've done that. Hence, no insisting, it would be in most poor taste. Contribute only what you feel like contributing, or nothing at all, if that's how your feelings take you. Everyone who does contribute is logged in for the "..., with input from ..." by default, and will not be mentioned only at his explicit request not to be mentioned.
I'm not in this for the money, but fair is fair and has nothing to do with money. My ability to look anyone in the face with a clear conscience has no price. Remember that and feel free to question me on this any time you feel like it.
Also, once completed, I will ask if anyone is interested in the PCBs for their bare manufacturing price plus shipping. You may not like the prices, I certainly don't, but in Serbia, everything is more expensive than where you people live. A fact of life I can do nothing about.
In my life, I have published 326 texts in magazines, I have published a book which sold out in two editions (the third being aborted by local wars), I have authored and anchored my own TV show for three seasons, I have sunp off another TV show from my original one, I have hed my own radio show regarding audio for four seasons, and I have implemented some of my ideas with people who were willing to put the time and effort into it - such as my own loudspeakers, for example.
And over the years, I have seen whole parts of my texts incorporated into other peoples' texts totally in violation of copyrights, I have even seen entire texts of mine being reprinted without permission.
With all due respect to one and all, but I know far better what it is to be ripped off than most people here. Furthermore, most of you live in ordered societeies in which laws are at least partially observed, wheras I live in country in which laws mean literally nothing "if you know the man". Out here, politics is everything, and if you're connected to any party in the ruling coalition, you need not worry about legality - anything goes.
To say that I personally am deeply ashamed because of this is the least you could say, in which case I would not, under any circumstances, rip anyone off. The instant I posted the schematic and asked my question, the moniker describibng the authorship of that amp increased no end. Before posting, it read just as "By DVV"; the instant somebody responded to may questions, uit changed to: "By DVV, with input from Thorsten Loesch, .... etc, on DIYaudio.com site". Heck, I make my friend of 20+ years who does the PCB artwork for me sign himself on those boards.
I am extremely sensitive to the issue of authorship because I've been burnt many times. And all it took was just a short note asking for permission, I am always inclined to give it. I feel all knowledge (I credit myself with some small parts of it) cannot reach its full potential until it is shared, and I cannot know how you feel about a project until I share it.
I fully understand John's position, I've been there, I've done that. Hence, no insisting, it would be in most poor taste. Contribute only what you feel like contributing, or nothing at all, if that's how your feelings take you. Everyone who does contribute is logged in for the "..., with input from ..." by default, and will not be mentioned only at his explicit request not to be mentioned.
I'm not in this for the money, but fair is fair and has nothing to do with money. My ability to look anyone in the face with a clear conscience has no price. Remember that and feel free to question me on this any time you feel like it.
Also, once completed, I will ask if anyone is interested in the PCBs for their bare manufacturing price plus shipping. You may not like the prices, I certainly don't, but in Serbia, everything is more expensive than where you people live. A fact of life I can do nothing about.
Last edited:
Hi,
K246 and J103 if matched are not a bad choice for the input stage and input cascode. You loose some transconductance but that may be a good thing. You can loose the CCS's for the tails as the input will be self-biasing and the cascode can just use resistive biasing to the tails. It does simplify things a lot and no base current no distortion from base current...
They also make a okay figure as VAS Buffer, VAS Transistors I would suggest 2SA1360/2SC3423 but you can use BD139/140 if you absolutely have to have have them.
One more trick, use the N-Channel 2SK246 to buffer the positive (PNP) VAS Transistor and the P-Channel 2SJ103 to buffer the negative (NPN) VAS Transistor. Then you can use a single resistor (set for around 3mA through the J-Fets between the two VAS base's so give the VAS Buffers operating current and you minimise the voltage appearing across the J-Fets.
The 2SK216/2SJ77 drivers are not really negotiable I'm afraid...
Ciao T
K246 and J103 if matched are not a bad choice for the input stage and input cascode. You loose some transconductance but that may be a good thing. You can loose the CCS's for the tails as the input will be self-biasing and the cascode can just use resistive biasing to the tails. It does simplify things a lot and no base current no distortion from base current...
They also make a okay figure as VAS Buffer, VAS Transistors I would suggest 2SA1360/2SC3423 but you can use BD139/140 if you absolutely have to have have them.
One more trick, use the N-Channel 2SK246 to buffer the positive (PNP) VAS Transistor and the P-Channel 2SJ103 to buffer the negative (NPN) VAS Transistor. Then you can use a single resistor (set for around 3mA through the J-Fets between the two VAS base's so give the VAS Buffers operating current and you minimise the voltage appearing across the J-Fets.
The 2SK216/2SJ77 drivers are not really negotiable I'm afraid...
Ciao T
Last edited:
Thank you Thorsten. I will investigate as soon as I can.
Two things. I do not HAVE to have anything except the output devices, and this includes BD 139/140. I do like them, they do come in handy, but I am not married to them.
The limited choice of locally available FETs is dictated by the requirements of Japanese TV sets and some audio, I think. DIY audio is small locally, too small to determine anything. The local DIY forum is all about worshipping the ground Nelson Pass walks on, and they always have group orders and buys for many FETs and MOSFETs they use. They are mostly in the trip of "if it ain't a FET of some kind, it can't play music well". Well, maybe a tube or two, but that's it.
Last edited:
That seems eminently plausible --- somehow the footnote got lost. As far as which is superior, it depends on whether you want a negative tempco of collector current or not. I recall an opamp using standard two-diode bias
which did so partly to make the transconductance of the nondegenerated input pair more constant with temperature.
To clarify: the LED bias will correspond to a low-tempco, the two diodes to a negative tempco, the latter being appropriate to some extent for stabilizing a diff pair gm.
Brad
Last edited:
- Status
- Not open for further replies.