Hi Aksa,AKSA said:
I am saying that Glen would criticise these current values as bad design and his reply virtually said so. I accept your values as examples and they do indeed show that VAS base current forces the mirror to supply unequal currents to the two halves of the LTP.
Even halving the VAS current and doubling the VAS hFE still results in a 2% imbalance in the LTP currents. The mirror will also have an imbalance of around another 0.5%.
Glen obviously is aware of this problem but addresses it by what he calls good design, implying that any other method is bad design.
You are in the resistor loading camp, Glen appears to be arguing that the mirror camp is better.
Hmmm, Andrew, thanks, I can wear that. We should take Glen's comments lightly, he is just trying to educate! 
I use LTP resistor loading (with particular attention to LTP balance) on two of my amps; on another three I use a modded CM. Some years ago I had noticed that resistive loading sounded better (when balanced) than CM, but recently discovered a mod to the CM (effectively a Widlar) which restored the sonics and gave the twin benefits of higher fb factor and automatic offset control. This eliminates a trimpot, which saves money and time.
Glen makes the point that charge/discharge of the Miller cap on the VAS is effected alternately from the LTP and the CM. Both circuit paths are collectors; but I would not expect their source impedances to be identical in any event, so slight variations in currents would make little difference except at the very highest frequencies, presumably approaching the HF pole.
While I don't generally condemn topologies as good/bad design, I can see Glen's point; everything is a compromise, and in light of the fact that with some work almost any configuration can be made to function, I try to choose topologies which give the best sonics, at least in my view. All of these circuits are capable of vanishingly low distortion, so rather than agonise over distortions that are 140dB down in sims, I try to road test each new circuit where subjective differences are quite apparent. It is very easy to be seduced by the intellectual purity of a topology, only to discover on building it that it does not sound good.
Focussing on the empirical is time consuming, but I'm more than prepared to listen to the experts here. From time to time, they have some very good ideas. A classic case of brilliant design is, in my view, Mihai Rauta's 'Roender' amp.
Cheers,
Hugh
I use LTP resistor loading (with particular attention to LTP balance) on two of my amps; on another three I use a modded CM. Some years ago I had noticed that resistive loading sounded better (when balanced) than CM, but recently discovered a mod to the CM (effectively a Widlar) which restored the sonics and gave the twin benefits of higher fb factor and automatic offset control. This eliminates a trimpot, which saves money and time.
Glen makes the point that charge/discharge of the Miller cap on the VAS is effected alternately from the LTP and the CM. Both circuit paths are collectors; but I would not expect their source impedances to be identical in any event, so slight variations in currents would make little difference except at the very highest frequencies, presumably approaching the HF pole.
While I don't generally condemn topologies as good/bad design, I can see Glen's point; everything is a compromise, and in light of the fact that with some work almost any configuration can be made to function, I try to choose topologies which give the best sonics, at least in my view. All of these circuits are capable of vanishingly low distortion, so rather than agonise over distortions that are 140dB down in sims, I try to road test each new circuit where subjective differences are quite apparent. It is very easy to be seduced by the intellectual purity of a topology, only to discover on building it that it does not sound good.
Focussing on the empirical is time consuming, but I'm more than prepared to listen to the experts here. From time to time, they have some very good ideas. A classic case of brilliant design is, in my view, Mihai Rauta's 'Roender' amp.
Cheers,
Hugh
AndrewT said:
You brought up the a few issues that effect the current balance of the LTP when using a current mirror, ignoring the fact that without the current mirror the situation is generally far worse. I explained how to address a few of these issues and how not to address them. All you've done since is complain. This is getting boring.
AKSA said:
About be paranoic obsession to have
the voltage across each LTP pair Transistor absolutely equal.
This is what we often see beginners do.
Not experienced amplifier designers, so much.
The only thing you wull acheive with this
is a tiny bit lower gain, in theinput as a whole.
And this is no improvement. Is it?
(One exception note is in place here.
When I design as low voltage amps as only 2 - 5 Volt supply
I will sometimes try to equal the voltage across the both transistors in diffrential input.
But for voltages above 5 Volt or as much as 20 Volt C-E,
we can forget about robbing one transistor of voltage, for no use.)
Again, AKSA explains one case where is no free lunch.
Using CCS, instead of resistor, will change the overall parameters of the amplifier
at such a degreee that measures may have to be taken.
To avoid instability and misbehavior.
Finally.
Very good amplifier idea, Dxvideo
Thanks for one interesting Topic
Lineup
Hi Lineup,
Another old topic, but still controversial one. On a subjective note close balancing has the benefit of creating wider more precise soundstage. On a electrical note reduces distortion, both high and low order.
For beginners, I dont think so......., more for advanced design I would say.............
Read Aksa reply......... Using resistive loading with particular attention to balancing........and resistive loading sounded better when balanced......., this should tell you something or are hundreds of peope who listen to his amps wrong about its sound quality.
There is no problem using mirrors either, can sound very good depending on the design of mirror. Ordinary ones sound awful.
Minor compromises such as stability has to be looked at with mirrors, but hey look at all the other benefits, subjective and electrically.
Another old topic, but still controversial one. On a subjective note close balancing has the benefit of creating wider more precise soundstage. On a electrical note reduces distortion, both high and low order.
For beginners, I dont think so......., more for advanced design I would say.............
Read Aksa reply......... Using resistive loading with particular attention to balancing........and resistive loading sounded better when balanced......., this should tell you something or are hundreds of peope who listen to his amps wrong about its sound quality.
There is no problem using mirrors either, can sound very good depending on the design of mirror. Ordinary ones sound awful.
Minor compromises such as stability has to be looked at with mirrors, but hey look at all the other benefits, subjective and electrically.
Listen to Hugh fellows, the guys is not young, very experienced
and the most important, well succeded.... he made a name and a product sold around the world, others want to be so good as he is, and tested, proved real world, but can only talk about academic things having not a name and a position into this world...people trying to be.
Hugh already IS...he made himself working hard and knows that you can only understand the whole thing doing and trying the modifications...while others goes imagining..he knows.
Read with attention what he say....mathematics and theories are needed to calculate circuits...to make them sound good you have the second step, to learn how each circuit sounds and the combination of those things.
I motor, a machine can have turbo charger, especial overhead control to valves, incredible exaust tubes, polished input tubes, compressor, fuel injection, electronic controled injection and ignition... computer monitored exausted gazes, controled ignition point timing......BUT...if pistons have different weigth, the motor will vibrate and will not develop the full power could develop because of this detail....also the audio happens this way.
Attention once again folks..... Hugh have made a product and a name..has a factory, a brand, a respectable equipment sold around the world by happy and satisfied customers...others....well...others are very good too.....but examine what they have provided you as diy product..observe what they have made.
Hugh is a tested....he is not trying to be under the spot ligth...he is already a star.
Listen to Hugh... read what he say..... well... listen to everybody...but test them.... watch what they have done.... some speaks, speaks... and do nothing.
All i know was because of forum, and i learned the ones have consistence and the others have formulas..... because of Hugh i could give my second step up into the direction of the knowledge..others goes analising the cement that was used to construct the stair.
regards,
Carlos
and the most important, well succeded.... he made a name and a product sold around the world, others want to be so good as he is, and tested, proved real world, but can only talk about academic things having not a name and a position into this world...people trying to be.
Hugh already IS...he made himself working hard and knows that you can only understand the whole thing doing and trying the modifications...while others goes imagining..he knows.
Read with attention what he say....mathematics and theories are needed to calculate circuits...to make them sound good you have the second step, to learn how each circuit sounds and the combination of those things.
I motor, a machine can have turbo charger, especial overhead control to valves, incredible exaust tubes, polished input tubes, compressor, fuel injection, electronic controled injection and ignition... computer monitored exausted gazes, controled ignition point timing......BUT...if pistons have different weigth, the motor will vibrate and will not develop the full power could develop because of this detail....also the audio happens this way.
Attention once again folks..... Hugh have made a product and a name..has a factory, a brand, a respectable equipment sold around the world by happy and satisfied customers...others....well...others are very good too.....but examine what they have provided you as diy product..observe what they have made.
Hugh is a tested....he is not trying to be under the spot ligth...he is already a star.
Listen to Hugh... read what he say..... well... listen to everybody...but test them.... watch what they have done.... some speaks, speaks... and do nothing.
All i know was because of forum, and i learned the ones have consistence and the others have formulas..... because of Hugh i could give my second step up into the direction of the knowledge..others goes analising the cement that was used to construct the stair.
regards,
Carlos
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.