Dave said:
Yes. This is a "MUST" read. Regardless I do not follow his point of view in my construction of both good sounding and measuring amp.
sajti,
Mr. Self spends most of the book discussing ways to reduce THD without increasing feedback, mainly through getting the best open loop linearity from each gain stage. This is generally regarded as pretty good design practice as far as I know in most audiophile circles.
The Blameless circuit he presents looks pretty simple, not to many parts just a well thought out and well executed design. I haven't tried one myself but I don't see any reason why it should sound bad, unless a greater level of lower order harmonics would be beneficial to the subjective performance...
The Blameless circuit actually has a few similarities with the internal schematic of the LM3875 gain clone chip.
I think the most prominent are current mirror load on the diff pair and the darlington style VAS stage. Mr Self calls this a beta enhanced VAS stage.
Mr. Self spends most of the book discussing ways to reduce THD without increasing feedback, mainly through getting the best open loop linearity from each gain stage. This is generally regarded as pretty good design practice as far as I know in most audiophile circles.
The Blameless circuit he presents looks pretty simple, not to many parts just a well thought out and well executed design. I haven't tried one myself but I don't see any reason why it should sound bad, unless a greater level of lower order harmonics would be beneficial to the subjective performance...
The Blameless circuit actually has a few similarities with the internal schematic of the LM3875 gain clone chip.
I think the most prominent are current mirror load on the diff pair and the darlington style VAS stage. Mr Self calls this a beta enhanced VAS stage.
Dave,
the current mirror, and the beta enhanced VAS increase the open loop gain, but not necessary increase the linearity. If I want to increase the linearity without large loop feedback, I will avoid the current mirror, and the emitter follower, and add some resistor between the emitter of the VAS stage, and the negative voltage rail. The second action add local feedback to the VAS, and this will gives better linearity....
Only the emitter degeneration of the input stage is the solution which increase the linearity, and reduce the open loop gain.
sajti
the current mirror, and the beta enhanced VAS increase the open loop gain, but not necessary increase the linearity. If I want to increase the linearity without large loop feedback, I will avoid the current mirror, and the emitter follower, and add some resistor between the emitter of the VAS stage, and the negative voltage rail. The second action add local feedback to the VAS, and this will gives better linearity....
Only the emitter degeneration of the input stage is the solution which increase the linearity, and reduce the open loop gain.
sajti
Interesting guys you are, some of you combative, alike warriors, strong in you ideas
I do not want to face some of you in a war...some of you really goes over speed and with strong.... i am perceiving, long time, people of Prague, also tzecs old eslovakian, passionate (full of love and passion and strong to defend you ideas)
Now chating with friends, i will give a break and finish later.
Carlos
I do not want to face some of you in a war...some of you really goes over speed and with strong.... i am perceiving, long time, people of Prague, also tzecs old eslovakian, passionate (full of love and passion and strong to defend you ideas)
Now chating with friends, i will give a break and finish later.
Carlos
Carlos,
I have visited many foreign countries from USA to Japan, and I do not feel like commenting on different nation's behaviour 😉
Cheers, Pavel
I have visited many foreign countries from USA to Japan, and I do not feel like commenting on different nation's behaviour 😉
Cheers, Pavel
Re: Interesting guys you are, some of you combative, alike warriors, strong in you id
Prague is really nice city. I drunk a lot of beer over there
🙂 I like all the people I met , during the week I spent in Prague!
sajti
destroyer X said:
Prague is really nice city. I drunk a lot of beer over there
🙂 I like all the people I met , during the week I spent in Prague!
sajti
Re: Re: Interesting guys you are, some of you combative, alike warriors, strong in you id
We are a little bit off topic, but I would like to say that so nice is Budapest.
sajti said:
We are a little bit off topic, but I would like to say that so nice is Budapest.
I think Destroyer not so angry for the offtopic😉
Pavel, I wrote, that if You are interesting, I can send the schematic of my last amplifier. I had some problem to put up here, but in email there is no problem.
Sajti
Pavel, I wrote, that if You are interesting, I can send the schematic of my last amplifier. I had some problem to put up here, but in email there is no problem.
Sajti
Nor angry, i am hungry, big a fat, some elefant a think
But not nervous as you can think, we are not combative, we need to be more..... our country is peacefull... too much, we must show our muscles too.... do you know... we have muscles too.
As in the past, i will like to see the courage to put schematics on forum.... i can see too much theory discussions...no one put one idea to turn this one better, or more real.
Sam i will check if i assemble the 11.4.... a can remenber, circuit was destroyed, a double barreled shot direct in the middle, so big the frustration
hehe, not so peacefull too.... just a little! hohô!
LM3875 sounded better in my home.... i have a special test now.... i called my maid.... completely out of our knowledge..my wife, my children and the neighboor...if they say same words as i can say.... i do not go ahead anymore.... if human ears said no... i decided to finish... i cannot discover the epson delta integrated determination variation numbers... so my meters is human ears.
Scope sometimes, but as a reference to talk with you... could see awfull waveforms and good sounds!... so...... the scope!
Thank you Dave, your attention and words
I already receive my "non sense" rocket portion today... as a medicine hahahaha!
PMA my bonds shake when you appear in forum, alike till, i have an interesting idea related your actions.... i remember the movie patton....General Patton... and a heavy war tank goes direct to the screen, the heroe (of course...me!) go deep in his hole, and the tank passed over..... do you know who is the war tank...maybe till....maybe you!...hahaha!
Others friends already told me afraid of you.... you strong!... and go deep... decided.... better to open the way... if not..my god!
Do not dislike you..... problem is that you have something i need and have not.... to be more strong defending my ideas.... with passion..... we, brazilian... to much peacefull. and USA is pulling oranges (market problems) into our ----.
I like of topic.... because world is not only electronics, i like girls too...you, of course also like.... dont imagine things, with the tank with engines running!.... and turn you cannon other side hihihihi, please!
Cheers Satji...if you have inside one war...please, tell me, you Satji and PMA.... because if inside a war i will be firing only the skies, maybe, an accident i can hurt a bird...if you there...i am sure i will go backward... and fast... could you perceive that!...of course not....but i am not the only that perceive you strong defending ideas....hummmmmm have slow speed?.
Before Daved, Planet 10 warns me to be more delicated.
I am sorry, in advance, if my words are not the best, if i hurt someone was an accident, i am only trying to protect myself, because i can see (OK! delirating...all right!) people with rockets and grenades... some of them already with grenade pins between their teeths....... this is not the place.... you are forgived.... and not guilty... i am delirating, isnt it?
Carlos
But not nervous as you can think, we are not combative, we need to be more..... our country is peacefull... too much, we must show our muscles too.... do you know... we have muscles too.
As in the past, i will like to see the courage to put schematics on forum.... i can see too much theory discussions...no one put one idea to turn this one better, or more real.
Sam i will check if i assemble the 11.4.... a can remenber, circuit was destroyed, a double barreled shot direct in the middle, so big the frustration
hehe, not so peacefull too.... just a little! hohô!
LM3875 sounded better in my home.... i have a special test now.... i called my maid.... completely out of our knowledge..my wife, my children and the neighboor...if they say same words as i can say.... i do not go ahead anymore.... if human ears said no... i decided to finish... i cannot discover the epson delta integrated determination variation numbers... so my meters is human ears.
Scope sometimes, but as a reference to talk with you... could see awfull waveforms and good sounds!... so...... the scope!
Thank you Dave, your attention and words
I already receive my "non sense" rocket portion today... as a medicine hahahaha!
PMA my bonds shake when you appear in forum, alike till, i have an interesting idea related your actions.... i remember the movie patton....General Patton... and a heavy war tank goes direct to the screen, the heroe (of course...me!) go deep in his hole, and the tank passed over..... do you know who is the war tank...maybe till....maybe you!...hahaha!
Others friends already told me afraid of you.... you strong!... and go deep... decided.... better to open the way... if not..my god!
Do not dislike you..... problem is that you have something i need and have not.... to be more strong defending my ideas.... with passion..... we, brazilian... to much peacefull. and USA is pulling oranges (market problems) into our ----.
I like of topic.... because world is not only electronics, i like girls too...you, of course also like.... dont imagine things, with the tank with engines running!.... and turn you cannon other side hihihihi, please!
Cheers Satji...if you have inside one war...please, tell me, you Satji and PMA.... because if inside a war i will be firing only the skies, maybe, an accident i can hurt a bird...if you there...i am sure i will go backward... and fast... could you perceive that!...of course not....but i am not the only that perceive you strong defending ideas....hummmmmm have slow speed?.
Before Daved, Planet 10 warns me to be more delicated.
I am sorry, in advance, if my words are not the best, if i hurt someone was an accident, i am only trying to protect myself, because i can see (OK! delirating...all right!) people with rockets and grenades... some of them already with grenade pins between their teeths....... this is not the place.... you are forgived.... and not guilty... i am delirating, isnt it?
Carlos
Hi Carlos,
nobody makes war here! I think, that this site is for the good discussion. But of course during it, everybody protects his own point of view. Somebody harder, somebody softer, but this is no problem.
sajti
nobody makes war here! I think, that this site is for the good discussion. But of course during it, everybody protects his own point of view. Somebody harder, somebody softer, but this is no problem.
sajti
Hi Destroyer,
There are two simple checks you can set up to test for amplifier induced amplifier-loudspeaker interface distortion, which D Self stated had not been proven at the time he wrote his articles and books.
(1) Monitor the amplifier output from another room using 10:1 voltage divider fed high quality headphones.
The difference in headphone sound as you switch between real-world loudspeakers and genuinely non-inductive 8 ohm test loads does separately clarify loudspeaker induced amplifier distortion.
(2) Wrap a known good two or three-way loudspeaker with a blanket so that only the tweeter output can be directly and clearly heard. A good amplifier will produce clean sounding treble, but an amplifier that suffers from amplifier-loudspeaker interface distortion will generate tweeter products which arise from low+mid frequency loudspeaker excited back EMF generation.
If D Self had run these tests on his own designs, and not relied so much upon instrumentation, he might not have chosen the term Blameless.
Your listening trials make most enjoyable reading, keep them coming.
Cheers ............... Graham.
There are two simple checks you can set up to test for amplifier induced amplifier-loudspeaker interface distortion, which D Self stated had not been proven at the time he wrote his articles and books.
(1) Monitor the amplifier output from another room using 10:1 voltage divider fed high quality headphones.
The difference in headphone sound as you switch between real-world loudspeakers and genuinely non-inductive 8 ohm test loads does separately clarify loudspeaker induced amplifier distortion.
(2) Wrap a known good two or three-way loudspeaker with a blanket so that only the tweeter output can be directly and clearly heard. A good amplifier will produce clean sounding treble, but an amplifier that suffers from amplifier-loudspeaker interface distortion will generate tweeter products which arise from low+mid frequency loudspeaker excited back EMF generation.
If D Self had run these tests on his own designs, and not relied so much upon instrumentation, he might not have chosen the term Blameless.
Your listening trials make most enjoyable reading, keep them coming.
Cheers ............... Graham.
thank you too Maynard
i will do some listening tests... but will delay a little, waiting some parts.
Carlos
i will do some listening tests... but will delay a little, waiting some parts.
Carlos
Graham,
Does the speaker/amp interface distortion mean that it’s a good idea to use different topologies for active and passive systems?
What should one consider for the different systems, and how should the differences be implemented in the finished design?
I know of cause you need a larger PSU for the bass comparing to the PSU for the tweeter section, but speaking only about the amplifier topology what should be done differently?
\Jens
Does the speaker/amp interface distortion mean that it’s a good idea to use different topologies for active and passive systems?
What should one consider for the different systems, and how should the differences be implemented in the finished design?
I know of cause you need a larger PSU for the bass comparing to the PSU for the tweeter section, but speaking only about the amplifier topology what should be done differently?
\Jens
Hi Jens,
I cannot say that the basic Blameless (transistor) topology is especially bad.
It is the Miller connected C.dom and output choke that ruin it, because the input stage cannot supply enough current to dynamically charge the VAS gain amplified C.dom, and output chokes are bad news anyway.
That is why I asked Destroyer if he had just built one, in case he would like to try some experimental modifications.
Cheers ............ Graham.
I cannot say that the basic Blameless (transistor) topology is especially bad.
It is the Miller connected C.dom and output choke that ruin it, because the input stage cannot supply enough current to dynamically charge the VAS gain amplified C.dom, and output chokes are bad news anyway.
That is why I asked Destroyer if he had just built one, in case he would like to try some experimental modifications.
Cheers ............ Graham.
Graham,
this amp http://www.bvaudio.sk/Pa300.htm is in fact a blameless amp, and I can assure you that it is not any bad.
Pavel
this amp http://www.bvaudio.sk/Pa300.htm is in fact a blameless amp, and I can assure you that it is not any bad.
Pavel
Hi Graham!
In general I agree to your concerns of the Cdom and TIM distorsion.
But how do you get a gain of 3000 for the VAS ?
I am refering to the schematic as given in the early link:
http://www.dself.dsl.pipex.com/ampins/dipa/dipa.htm
Even if TR4 would operate to a perfect current source (which TR5 is trying to provide) and no output load from the output stage would be assumed.. even then the VAS would probably not make more gain than 200 due to the output resistance of T4..... I think.... or do I mess up things...?
@ All:
Coming from R2=2.2k and Cdom = 47pF we can assume a max. slew rate around 5.8V/us for slewing of VAS towards positive direction.
(2.2k can deliver max. around 0.27mA while keeping TR4 off at the same time)
For slewing toward the other direction this design would probably be 4...5 times faster as here TR3 can deliver around 1.4mA (and in that condition about 100uA...200uA of the 1.4mA required for basis current of TR4).
Please note that the tail currents of TR2 and TR3 are massively unsymmetrical at zero signal condition.....
I could think that especially at higher output levels and high frequencies the unsymmetric slew rate might be unfortunate....
Carlos,
you could try to reduce R2 and R3 down to about 1k.
This would balance TR2 and TR3. Also possible slew rate for
pos and neg direction should come more symmetrical and above
10V/us.
May be this helps .... may be not...
For low DC offset you should then change R4 to 22K (if R2 and R3 are set to 1k).
Bye
Markus
P.S.
If I am refering to another schematic than you, please let me know.
In general I agree to your concerns of the Cdom and TIM distorsion.
But how do you get a gain of 3000 for the VAS ?
I am refering to the schematic as given in the early link:
http://www.dself.dsl.pipex.com/ampins/dipa/dipa.htm
Even if TR4 would operate to a perfect current source (which TR5 is trying to provide) and no output load from the output stage would be assumed.. even then the VAS would probably not make more gain than 200 due to the output resistance of T4..... I think.... or do I mess up things...?
@ All:
Coming from R2=2.2k and Cdom = 47pF we can assume a max. slew rate around 5.8V/us for slewing of VAS towards positive direction.
(2.2k can deliver max. around 0.27mA while keeping TR4 off at the same time)
For slewing toward the other direction this design would probably be 4...5 times faster as here TR3 can deliver around 1.4mA (and in that condition about 100uA...200uA of the 1.4mA required for basis current of TR4).
Please note that the tail currents of TR2 and TR3 are massively unsymmetrical at zero signal condition.....
I could think that especially at higher output levels and high frequencies the unsymmetric slew rate might be unfortunate....
Carlos,
you could try to reduce R2 and R3 down to about 1k.
This would balance TR2 and TR3. Also possible slew rate for
pos and neg direction should come more symmetrical and above
10V/us.
May be this helps .... may be not...
For low DC offset you should then change R4 to 22K (if R2 and R3 are set to 1k).
Bye
Markus
P.S.
If I am refering to another schematic than you, please let me know.
- Status
- Not open for further replies.