With Carlos bringing Carnivale to DIY 
, and Kanwar transporting us to Bollywood with his enthusiasm for quasi-comps 
, yes, Jacco, keep your vanity (we are all 21 in our minds!), and Kanwar, a question, before Dimitri returns from his party-going 
:
Have you studied the high power 90V circuit from Quasi in South Australia, and do you like the circuit? in particular however, you want a falling FR to meet the Bode-Nyquist criteria and a cascode, by eliminating Miller effect, means you have to ramp up the lag compensation to keep it stable. I like the use of both diff pair outputs, but suspect the CCS on the VAS is not necessary. Can you give us your opinion.......??
Cheers,
Hugh
, and Kanwar transporting us to Bollywood with his enthusiasm for quasi-comps , yes, Jacco, keep your vanity (we are all 21 in our minds!), and Kanwar, a question, before Dimitri returns from his party-going :Have you studied the high power 90V circuit from Quasi in South Australia, and do you like the circuit? in particular however, you want a falling FR to meet the Bode-Nyquist criteria and a cascode, by eliminating Miller effect, means you have to ramp up the lag compensation to keep it stable. I like the use of both diff pair outputs, but suspect the CCS on the VAS is not necessary. Can you give us your opinion.......??
Cheers,
Hugh
AKSA said:With Carlos bringing Carnivale to DIY
Have you studied the high power 90V circuit from Quasi in South Australia, and do you like the circuit? in particular however, you want a falling FR to meet the Bode-Nyquist criteria and a cascode, by eliminating Miller effect, means you have to ramp up the lag compensation to keep it stable. I like the use of both diff pair outputs, but suspect the CCS on the VAS is not necessary. Can you give us your opinion.......??
Cheers,
Hugh
Hi HUGH DEAN,
Welcome to the party again!
Yes, I have studied the 90V Quasi amp.....
Yes i liked it , its good...to an extent....
In my opinion CCS is more used as Bias Controller in Quasi's amp, Setting a simple resistor wouldn't do any temp.compensation for the output..But...
Yes, it could be replaced by a simple resistor..if the biasing is controlled by CCS of First input Differential.....along with mechanism of controlling temp.compensation...
I have recently designed an amp features One differential + Emmiter Follower + Cascode BJT VAS....Just replace the Second PNP BJT differential with...NPN EF's in Quasi amp and then Drive Cascoded PNP BJT's emmiters or P-channel Mosfets sources .....then on to output N-channels....This Design gives Exceptional Wide Bandwidth upto 500KHZ Flat...closed loop gain is 36X ......Only one comp.Cap at feedback resistor + RC network at collectors of first differential works fine to stable the design...and its R2R...also
regards,
K a n w a r
Hi Kanwar,
Of course aspects that so much affect the resulting sound quality of Class AB amps such as Power Supply Rejection are not going to be cured by double diode ref'd current sources or conventional Vas comp -
finer points that the Simple Killer Amp' (great name for a thread!) improves on with it's elegantly simple 'Patent pending' topology,
so you're listening to the music and not the power supply.
Cheers,
Greg
Of course aspects that so much affect the resulting sound quality of Class AB amps such as Power Supply Rejection are not going to be cured by double diode ref'd current sources or conventional Vas comp -
finer points that the Simple Killer Amp' (great name for a thread!) improves on with it's elegantly simple 'Patent pending' topology,
so you're listening to the music and not the power supply.
Cheers,
Greg
amplifierguru said:
Hi Greg (not me the other one),
Every time is read the name "Simple Killer Amp", I think about wiring the 240VAC straight to the aluminium volume knob. Now that would be the simplest "Killer" amp. 😀 (Don't do this only a joke)
regards
Hi Greg #,
I'm sure it would sail through the Govt req'd $5k C-tick tests!
Got to be a winner. Just need some sporting hero to endorse it.
Cheers,
Greg
😀
I'm sure it would sail through the Govt req'd $5k C-tick tests!
Got to be a winner. Just need some sporting hero to endorse it.
Cheers,
Greg
😀
amplifierguru said:
My question Greg is this -
Are you saying that the PSRR on SKA is so good that there is no sonic benifit at all from all of the usual techniques to clean up the supply and using seperate supplies for o/p & i/p / follower stages ?
If this is the case, it is indeed very remarkable as I have never found any cct to date, that did not benefit from noise reduction measures in the supply.
mike
Greg Erskine said:
O.G. Simpleson,
please post before and after SKA pictures, any orientation is fine.
A shocking tale:
I just got a new test fixture with a 1" diameter hole, and a 2" square opening, exposing 7.5KV....
The safley interlock switch already welded itself to the bench top and dumped a 20A breaker.
Fun.
I just got a new test fixture with a 1" diameter hole, and a 2" square opening, exposing 7.5KV....
The safley interlock switch already welded itself to the bench top and dumped a 20A breaker.
Fun.
amplifierguru said:
This is very much like the amp i'm using now but Ive only got half of the input stages.. 😛, It sounds good. i think I'll try to spice it up on the computer and make it double..
but i'm concerned about q5,q6,r11,r12, I think i can see that it would'nt be linear, but also ,if so that it doesnt matter.. oh well i'll let the computer worry about that part.. i dont have the bootstrapped collectorload.. hmmm that's a bit interresting(only a little).. i'll try that out..
peranders said:
Nelson Pass claims that damping factor isn't particular important.Something to think about. I don't say it's true but I'll guess there is a limit surely but where is it?
My opinion so far is that at least we should'nt be without damping factor.. hehe.. maybe lower damping is better, depending on what's causing the damping,wouldnt more output stages increase damping too?
Are the current through mosfets still 100mA? depending on voltage, my personal opinion is that it should be at least doubled.. but i realize that it can be a problem if there is more than one output pair.. I biased my irfp240's on 250mA but that's only on 28+28volt..
Hi mikelm,
....."Are you saying that the PSRR on SKA is so good that there is no sonic benifit at all from all of the usual techniques to clean up the supply and using seperate supplies for o/p & i/p / follower stages ?"
No, not at all. On simulation it shows about 40dB improvement over a conventional simple topology of diff'l, Vas,EF over much of the audio spectrum. In practice, with a conventional 2BJT current source this reduces to some 20 dB improvement, or a 10 times desensitization to power supply borne artefacts. Is there any room for improvement - probably. But what I didn't want was for a simple design to become overly labouring on one aspect chasing magical numbers with no real audible benefit, and no longer being simple.
Hi nikwal,
The Q5,6 and R11,12 are fundamental to the topology. Q5,6 are high Ft, small signal, Hfe =500 BJTs running at Vgs =3.5V and 8mA to drive the MOSFETs +/-2V . They also provide the low(er) Z to facilitate the bootstrap C's that boost first stage gain by 50, yes that's 50, dB with improved linearity through the resulting reduced Ic modulation. R11,12 could be one resistor , but, at just under 1W Pd, I split it into two 1Wers, the zero point providing a node for a number of interesting options, being modulated at Vg mod.
Another yet to be explored possibility is as an input node for error correction, with an op amp sensing in/out error and driving this point.
Output MOSFETs are running Iq about 50mA each. 3HD matches 2HD from other sources.
Cheers,
Greg
....."Are you saying that the PSRR on SKA is so good that there is no sonic benifit at all from all of the usual techniques to clean up the supply and using seperate supplies for o/p & i/p / follower stages ?"
No, not at all. On simulation it shows about 40dB improvement over a conventional simple topology of diff'l, Vas,EF over much of the audio spectrum. In practice, with a conventional 2BJT current source this reduces to some 20 dB improvement, or a 10 times desensitization to power supply borne artefacts. Is there any room for improvement - probably. But what I didn't want was for a simple design to become overly labouring on one aspect chasing magical numbers with no real audible benefit, and no longer being simple.
Hi nikwal,
The Q5,6 and R11,12 are fundamental to the topology. Q5,6 are high Ft, small signal, Hfe =500 BJTs running at Vgs =3.5V and 8mA to drive the MOSFETs +/-2V . They also provide the low(er) Z to facilitate the bootstrap C's that boost first stage gain by 50, yes that's 50, dB with improved linearity through the resulting reduced Ic modulation. R11,12 could be one resistor , but, at just under 1W Pd, I split it into two 1Wers, the zero point providing a node for a number of interesting options, being modulated at Vg mod.
Another yet to be explored possibility is as an input node for error correction, with an op amp sensing in/out error and driving this point.
Output MOSFETs are running Iq about 50mA each. 3HD matches 2HD from other sources.
Cheers,
Greg
mikelm []
Post #266
quote:
Originally posted by amplifierguru
so you're listening to the music and not the power supply.
My question Greg is this -
Are you saying that the PSRR on SKA is so good that there is no sonic benifit at all from all of the usual techniques to clean up the supply and using seperate supplies for o/p & i/p / follower stages ?
If this is the case, it is indeed very remarkable as I have never found any cct to date, that did not benefit from noise reduction measures in the supply.
mike
-----------------------------------------------------------
i agree with that too, power supply noise is not good, a form
of distortion in itself, however as my friend Douglas in his book
has identified more serious ones i dont see anything remarkable
Psrr is good though, i always aim for the quitest supplies
cheers
Post #266
quote:
Originally posted by amplifierguru
so you're listening to the music and not the power supply.
My question Greg is this -
Are you saying that the PSRR on SKA is so good that there is no sonic benifit at all from all of the usual techniques to clean up the supply and using seperate supplies for o/p & i/p / follower stages ?
If this is the case, it is indeed very remarkable as I have never found any cct to date, that did not benefit from noise reduction measures in the supply.
mike
-----------------------------------------------------------
i agree with that too, power supply noise is not good, a form
of distortion in itself, however as my friend Douglas in his book
has identified more serious ones i dont see anything remarkable
Psrr is good though, i always aim for the quitest supplies
cheers
amplifierguru said:
Yes - I know exactly what you mean I also love simplicity and I also really like your cct cos the diagram looks so neat ! but....
although I may have been guilty of chasing low distortion numbers in spice ( and to some extent have benefited from this cos some of my previous designs simply had too much of it ),
regarding noise it is always my ears, rather than my intellect, that have demanded more & more measures to reduce PSU noise and subsequently celebrated each and every one.....and no matter how much noise reduction I may have done in a given cct I have never failed to notice the benifit of the next mod that reduces it just that bit more. So....
When I do get around to buying & building your simple killer amp I would be very surprised if I do end up employing every last technique I can muster just like all the other amps I have built and I will celebrate that extra few dB's of PSRR as the icing on the cake. I am looking forward to finding out if this is the case.
Also I'm not sure that good PSRR figures have much impact on RF noise that seems to go straight through most CCS ccts etc and adds a roughness to the sound which is a joy to erradicate.
well, that's my halfpenny worth.....hope some might find it constructive & beneficial to their search for glorious reproduction.
cheers
mike
Hi mikelm,
You sound very much like me in your quest. As regards the PSRR, having listened for over a month now, I feel it is fine - I am using the GB150AS power supplies and individual 160W transformers so essentially monoblocks. However, I have anticipated DIYers like yourself (and me) who NEED to go that little bit further to see (or in this case hear), so I have provided holes for extra components for filtering the CCS's when the urge calls.
As regards RF, the amp is extremely quiet and has shown zero effect of household appliance switching, so I'm inclined to believe it's not a problem.
Cheers,
Greg
You sound very much like me in your quest. As regards the PSRR, having listened for over a month now, I feel it is fine - I am using the GB150AS power supplies and individual 160W transformers so essentially monoblocks. However, I have anticipated DIYers like yourself (and me) who NEED to go that little bit further to see (or in this case hear), so I have provided holes for extra components for filtering the CCS's when the urge calls.
As regards RF, the amp is extremely quiet and has shown zero effect of household appliance switching, so I'm inclined to believe it's not a problem.
Cheers,
Greg
Has anyone tried out Guru's amplifier?
Has anyone tried out Guru's amplifier? If yes, what's the verdict?
How does it sound? I'm interested in trying a 100w rms 8ohm amp, but I'm very confused. I can't figure out which sounds better...AKSA 100, Guru's 100, Rod's P101? Can anyone make a suggestion to me?
P.S....Sorry about my poor English.
Has anyone tried out Guru's amplifier? If yes, what's the verdict?
How does it sound? I'm interested in trying a 100w rms 8ohm amp, but I'm very confused. I can't figure out which sounds better...AKSA 100, Guru's 100, Rod's P101? Can anyone make a suggestion to me?
P.S....Sorry about my poor English.
I already have my kits, but it will be some time before I can listen to them...
Some comments:
- SKA - a new proposition, unusual circuit (the only really new in decades).
- ESP 101 - very classical circuit, there's nothing wrong with it.
- AKSA - more of a subjectivist circuit, also nothing wrong with it.
I really don't think the differences among any one of them are significant. Spend your time improving your speakes for a much bigger reward!
Some comments:
- SKA - a new proposition, unusual circuit (the only really new in decades).
- ESP 101 - very classical circuit, there's nothing wrong with it.
- AKSA - more of a subjectivist circuit, also nothing wrong with it.
I really don't think the differences among any one of them are significant. Spend your time improving your speakes for a much bigger reward!
Jorge said:
I think a folded cascode is substantally different
Really - I believe you when you have made them...🙂
Re: Mikem
Well I have not heard any of these amps but, someone once told me that building a very fast amp with with high slew rate etc would not be noticable to the sound.
I found out later when I tried it that he was wrong !
If the SKA has 100dB OL gain at 10Khz as claimed it suggests to me that this may be quite a fast amp. If this is so I would expect the clarity with bells chimes etc and overall transparency to be very good. Also the version with no FB cap will have added clarity.
I would guess the the other two would sound somewhat softer and a little veiled by comparison.
Any thoughts on this Greg ?
Which is best is a matter of opinion - Many, I have noticed, like a smooth sounding amp..... I like fast & transparent...😉
Please let us know when you have built & listened
cheers
mike
msamkl said:So, Mike what's your opinion? Which one sounds better?
Well I have not heard any of these amps but, someone once told me that building a very fast amp with with high slew rate etc would not be noticable to the sound.
I found out later when I tried it that he was wrong !
If the SKA has 100dB OL gain at 10Khz as claimed it suggests to me that this may be quite a fast amp. If this is so I would expect the clarity with bells chimes etc and overall transparency to be very good. Also the version with no FB cap will have added clarity.
I would guess the the other two would sound somewhat softer and a little veiled by comparison.
Any thoughts on this Greg ?
Which is best is a matter of opinion - Many, I have noticed, like a smooth sounding amp..... I like fast & transparent...😉
Please let us know when you have built & listened
cheers
mike
- Status
- Not open for further replies.