Hello,
Live instruments have way more high frequency energy than hi-fi , hi-fi has high frequency noise, this noise is what becomes offensive , too bright.
Simulations just confirm that this is a trial/error based mod.
In real world, depending on some components tolerance , it will either
be stable due to low standing current of the VAS , or if this latter
has high current enough , it will be instable , as showed by the simulations.
Indeed, when i hear of 1000V/uS slew rates , it always make
me quite pessimistic about the care that was devoted to checking stability....
Well, this crescendo amp was well known and build by thousand of DIY in the world from decades (including the VAS stage). Apart the CR mod, some polarisation currents has been slightly *increased* in this mod to optimize bandwich., and, right now, i'm listening music with it. In real world.
And, at this time, this mod has been done by several people with success, if i believe my mailbox and PMs.
Perfectly stable with no overload on square waves, despite all you can imagine in your stranges nightmares.
I would add that this mod (his schematic is very classical) is not different from the various SSA amps, including the one of this thread, and, as long as i believe what i read, they are all working ? Are-we all crazy ?
I don't understand what you are trying to do, or says. And i thing LC or Bigun do not understand neither.
If you do not understand - or know- current feed-back topology, learn and ask (Make a search about Mark Alexander, from Analog Devices, as an example).
Such slewrates and CFB are classical from decades.
End about that for me, sorry, but i have no time to loose arguing about nothing and out of topic.
Please, if you want to talk about the crescendo Mod, or if you have a problem to realize-it, open a topic about this subject.
Last edited:
Some will surely simulate it and find the same flaws...
Indeed , i wont numerate the miscalculations , they are basic ,
and any average designer can spot them even without simulator assistance.
Anyway, you should be more cautious before promoting such heavy
mods , as it can lead some followers to experiment quite a nightmare if they
havent the necessary understanding of how the circuit works.
As for being off topic , well , you seems to deliberately ignore
the fact that it s you that did brought this crescendo mod
in this very thread...
Indeed , i wont numerate the miscalculations , they are basic ,
and any average designer can spot them even without simulator assistance.
Anyway, you should be more cautious before promoting such heavy
mods , as it can lead some followers to experiment quite a nightmare if they
havent the necessary understanding of how the circuit works.
As for being off topic , well , you seems to deliberately ignore
the fact that it s you that did brought this crescendo mod
in this very thread...
Member
Joined 2009
Paid Member
Those will never built anything, just simulate, simulate, simulate, ... constantly seeking potential problems and going further more complex, since previous version showed minor flaw ... in the meantime others will enjoy fruits of their work.
Wahab, in contrary I greatly admire Esperado and his work, he surely made marvelous job, high performance Ferrari like amplifier.
Look, people do drive expensive speakers with their DIY amps, meaning nobody is so stupid to upgrade an amplifier without fine tuning calibrations (stability, bias, etc.), to be at least minimally sure their amp won't be unstable and destroy the speakers. The fixed values resistors etc. described in Esperado's upgrade were presented specifically for his case. If he would plan to make an upgrade brochure for universal upgrade for all Crescendo owners in the world, I'm sure he would certainly specify some values as trimmable and so on.
Well I fixed the bulk of the speaker hum, moved the input signal cables away from the mains trafo
I'm posting the spice file, partly because I'll probably lose it and the forum is a great repository for this kind of stuff !
The bottom one is the ''de-tuned'' feedback paths version for raising the second harmonic? The one not finally adopted?
Member
Joined 2009
Paid Member
I would really like to know how you are working, LC. Your methods.
You are so accurate, clever, elegant, and more than that ,"sensible" in your proposals.
I can read, when i look to your schematics, that you "hear" the parts you choose and feel the range where you decide to make them work.
I can feel, looking to you printed circuits designs that, in your 'reptilian' brain, you feel each printed line as the complex circuit it is at HF.
And it is like music. you know, all those different instruments making harmonious music *together*; helping each others...
All that cannot be made by "spice". ( One thing i'm pretty sure is LC listen, listen listen, and that his amazing experience comes from there).
Well spice is a fantastic tool (that i had discover recently) to explore deeply each stage and validate testings on a theoretical point of view. (Just it saves a lot of calculation's time)...
But i wonder how anybody can *design* an amp with it... if it is not yet designed by brain: ideas, habits and calculations.
Long years ago, we had not those tools. Design (original idea of architecture) , calculations, prototypes were the only way. Then in the industrial world, you learned, little by little when the SAV calls you for help on the first serial failures ;-)
DIY is not so different. Just, you can tune your things to the top... As does Bigun.
But, for sure, it is with our soldering irons that we do our job !
Build, tune and listen to the changes, and tune again.
This thread is original in a way, interesting and educational, as Bigun makes us share his experience, step by step, as if we where just behind his shoulder, smelling his soldering iron's smoke. And it is interesting to see how born, on an different working table, the LC designs.
Member
Joined 2009
Paid Member
I'm starting to wonder if I've learned something else that is useful from my TGM5. I may be somebody who does not prefer a predominant large 2nd harmonic profile. I like the sound of TGM5 but simulations suggest that H2 is at same level as H3 with the higher harmonics being much lower. In the past when I have toyed with changes to TGM1 as well as TGM5 (and others) designed to increase H2 relative to the other harmonics, the results didn't impress me.
I do believe that elevated levels of H2 and H3 is able to mask objectionable higher order distortion (TGM1) but it's not always as satisfying as avoiding the higher order distortion in the first place. TGM1 has a relaxed 1970's sound which I like, but it's clearly a 'flavour'.
I remember Nelson commented that he found about 1/3 of people prefer H2, another 1/3 prefer H3, and the rest neither/both equally (I maybe misquoting here).
I may be in the last 1/3.....
I do believe that elevated levels of H2 and H3 is able to mask objectionable higher order distortion (TGM1) but it's not always as satisfying as avoiding the higher order distortion in the first place. TGM1 has a relaxed 1970's sound which I like, but it's clearly a 'flavour'.
I remember Nelson commented that he found about 1/3 of people prefer H2, another 1/3 prefer H3, and the rest neither/both equally (I maybe misquoting here).
I may be in the last 1/3.....
Last edited:
Member
Joined 2009
Paid Member
Member
Joined 2009
Paid Member
Thanks! It was in post 87 where the schematic was. Is there a latest BOM for that schematic somewhere? What value are those resistors?
How universal is that degeneration application? Have you considered or tried a cascode on TGM5 for the VAS?
Member
Joined 2009
Paid Member
Andrey, if you mean add a contents at the first post - well I thought I can't edit old posts, only Moderators can do that, and I'm not moderate at all
I'll see what information I can add this weekend though.
Cascode on the VAS? - I did consider it. This would be the 'next step' if looking to run the amp at high rail voltages but I don't have a need for such power levels - I would prefer to use more sensitive speakers if I needed 'louder'.
This is the first and only time I've used a Cascode. For awhile I was decided to remove them from the design. Without them TGM5 would pretty much be a symmetric TGM3 which already sounded nice so I thought it would be lower risk and simpler. But in the end I decided in favour of the Cascode, partly because of the benefits of reduced power dissipation in the small SMT devices I used, partly because I was curious.
I'll see what information I can add this weekend though.
Cascode on the VAS? - I did consider it. This would be the 'next step' if looking to run the amp at high rail voltages but I don't have a need for such power levels - I would prefer to use more sensitive speakers if I needed 'louder'.
This is the first and only time I've used a Cascode. For awhile I was decided to remove them from the design. Without them TGM5 would pretty much be a symmetric TGM3 which already sounded nice so I thought it would be lower risk and simpler. But in the end I decided in favour of the Cascode, partly because of the benefits of reduced power dissipation in the small SMT devices I used, partly because I was curious.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.