I would like to agree with Richard Marsh that really great amps are not well known about here to most of our critics. We have made them, but that does not mean that we could not do even better. Usually years, even decades have passed since we last tried, so we most probably could make make something that is even more up to date. I am now listening to a 20 year old Parasound, and I would like something better. I'm pretty sure I would note the difference, even at my age.
I can't measure any distortion on any of my amps at <5W (while in A).
You need to test at >50W to see if the design has cancellation.
"not in that region" Ha ! I crank it up a lot !!
OS
Hi,
I have not tested nor heard your fine designs nor OS, either. But I have measured and heard Damir's and it is outstanding in all regards.
Two comments --- for my size room and effec speakers and spl needed for no compression, loudness needed at low freqs due to F-Munson effect etc, 50W does not make it for me.... I find about 200-250W my minimum needs...... maybe others also?
A level of THD of .1% would be detectable to me in some conditions/music. And, if source and recordings electronics etc each gear involved had .1% the accumulated level would be more easily detectable..... more unlike real sounds. So 1/10 that level is a minimum IMO. Now that means <.01% BUT if we include the recording chain as well then all equipment of record and playback should be even lower than .01% to be at an undetectable level and be like real music sounding (assuming no weirdness, like added compression etal). We can at least get to the master source sound with no detectable difference all the way to your listening room. This is where HD or HiRez mastered downloads work for me.
When you hear amps of considerably higher than 0.5 - 1v/usec/volt SR and extremely low distortion, low output Z, low reflected voltage back to Vas, etc..... the results are stunningly real. I wish we had more amps like this available to buy off the shelf.
THx-RNMarsh
Last edited:
What does this have to do with creating fine art level for audio? You can also learn what fine art is without buying it.... just appreciate it and understand it. It is Not fluff that I would like to see. Several here have taken a giant step towards audio as fine art and have truly advanced music reproduction. I hope more DIY get a chance to hear them.
THx-RNMarsh
Last edited:
That many can do much better than that. I assume you're including all transistors for a single channel? An AKSA style amp might beat that twice over at 7 transistors per channel.
Having posted that Dyna schemo, this would actually be a potentially interesting game.
But when small signal transistors are very inexpensive, unless you believe in the numerosity/complexity dogmas, use what you need. It certainly isn't about cost, and well-considered, it usually isn't about diminished reliability.
And be nice to them, and perhaps they will be cooperative. Yes it is a bit late.
The amp Richard mentioned use 10 transistors in the IPS, 12 transistors in the OPS (two pairs of drivers and four pairs of output transistors), 4 transistors used as a CCS and 2 for the bias spreader. All that to get 220 W/8 ohm before clipping, I would say that is not so many of active devices for the power and quality of the sound it can deliver.
Damir
Last edited:
The number of transistors I mentioned vs distortion is a general guideline. Note the test frequency.
OS, what's the HF loop gain on your two amps? Are they the same?
It's interesting how the harmonic structure is different. With VAS loading, you can tune the balance between the seconds and thirds.
OS, what's the HF loop gain on your two amps? Are they the same?
It's interesting how the harmonic structure is different. With VAS loading, you can tune the balance between the seconds and thirds.
Richard,
I mentioned 50 W as a ref power for the distortion figures. I have a low powered amp (class A sx-Amp) and a two big amps at 180 and 280 W per channel. I agree that bigger amps sound more realistic on some material.
On the 0.1% thing I mentioned, it is in consideration of all the other distortion sources in the signal chain. If you change yor speakers, or the room layout and the sound changes (now that is EASY to detect), why beat yourself up of a hardly detectable level of distortion in an amp? Better to work on your speakers and secondly convince your wife to go with your acoustically optimized layout for the living room IMV.
DBT is only way to verify audibly claimed differences in amplifier sound. The ear is a treacherous thing.
You may here a difference (however, remember the Carver test), but in the DBT, I'd like to see how many people would identify the 0.1% distortion amplifier correctly and secondly, how many would say 'OMG that sounds absolutely awful!'
Bonsai's view:-
1. Don't make things more complex than they need to be.
2. Don't chase vanishingly low levels of distortion when they don't make an amplifier audibly superior to one with 2 or 3 magnitudes higher distortion
3. Get decent speakers
4. Sort your room acoustic out
I'm still working on these things.
1. Don't make things more complex than they need to be.
2. Don't chase vanishingly low levels of distortion when they don't make an amplifier audibly superior to one with 2 or 3 magnitudes higher distortion
3. Get decent speakers
4. Sort your room acoustic out
I'm still working on these things.
I completely agree with 3 and 4.
1. If you are selling your amps then you want material and assembling cost as little as possible. In DIY how you decide what is more complex than it's needed?
2. Again if you are selling it's OK, but in DIY where is intellectual and technical challenge in that?
Getting the absolute best from the fewest components is a worthy challenge. Although you run the risk of realising that D.Self has been right all these years!
This is not the way of a logical or ordered mind.
Well you seem to spend a lot of time using completely the wrong terms and incorrect analogies. Saying 'it's ok people get me' is a cop out.
I am sure Bruno really won't lose any sleep because you poo poo his designs because I recommended them as a starter for discussion. I won't either.
My 2p worth is that there may well be a connection between the pinnacle of collection as perceived by 'society' and the quest for high end audio. In anthropological terms, and the meaning of such things as tokens within society or the collectors social network. I mean the motivation, not whether such things are actually any good (which is taken as read).
I think it crops up on this thread for example, in conversation topics such as fine wine, or fine food, elite cars etc
I don't think that is my motivation with audio though, but it means rubbing shoulders with those whose it seems to be: in discussion forums, for example. Some people genuinely simply love audio and look to achieve good results for their own satisfaction, as well and cheaply as possible - DIY seems a good magnet for this, because DIY has little social value beyond a community where genuine performance for the sake of it is the goal. One of the reasons I enjoy this forum.
Agree - its the intellectual/technical challenge that is the interesting part.
I have a fundamental problem with the blameless topology - its the non-symmetrical drive to the OPS and the non-symmetrical slew rates - which I note DS somewhere claimed it was actually necessasy (maybe I misunderstood this point, someone please correct me if I am wrong).
When I read how he tried to compensate for this in his book, I quickly ditched Lin topology and went for symmetrical - they are a bit more complex, but I see some good advantages. I would think he should have just called these two points out clearly in his book and said use a symmetrical design if its important to you. Instead we got the merry dance around the mulberry bush as he fudged it trying to push the topology where its not supposed to go.
I suppose a middle ground would be LTP input feeding balanced VAS stage a la Cordell. Again, Self finds fault with this approach, when it is in fact a pretty elegant solution if you are wedded to single ended LTP input structures.
By all accounts, the simplest decent sounding amp - albeit low power - has to be JLH's 10 watter - but in the split rail version (Richard, this is less than the idle power on your amps . . .
Even NP was impressed . . .
NB Anybody got a plot of the Mola Mola power amp square wave output driving a capacitive load at 10 kHz with a 1 us rise time source?
When I read how he tried to compensate for this in his book, I quickly ditched Lin topology and went for symmetrical - they are a bit more complex, but I see some good advantages. I would think he should have just called these two points out clearly in his book and said use a symmetrical design if its important to you. Instead we got the merry dance around the mulberry bush as he fudged it trying to push the topology where its not supposed to go.
I suppose a middle ground would be LTP input feeding balanced VAS stage a la Cordell. Again, Self finds fault with this approach, when it is in fact a pretty elegant solution if you are wedded to single ended LTP input structures.
By all accounts, the simplest decent sounding amp - albeit low power - has to be JLH's 10 watter - but in the split rail version (Richard, this is less than the idle power on your amps . . .
Even NP was impressed . . .
NB Anybody got a plot of the Mola Mola power amp square wave output driving a capacitive load at 10 kHz with a 1 us rise time source?
Last edited:
- Status
- Not open for further replies.