traderbam said:
I think you'll find that most successful manufacturing companies don't waste their profit margins on parts that are superfluous.
Conjecture.
Why?
Yes, agreed.Your example does not include the SOA protection that a real Naim circuit has. I didn't bother to mention this earlier.
The Naim provides an abject lesson on why subjectivist ''design by ear'' methods should be dismissed with alacrity.
That 22K collector resistor is, in fact, entirely useless; it imbalances first stage quiescent current, and, among other vices, aggravates DC offset at the output. Absolute rubbish.
The rudimentary single slope SOA protection scheme used by Naim may indeed protect the output devices, but is likely to be invoked with all but the most benign of loads; this would sound rather unpleasant.
Kiwanuka, M., ‘Transparent V-I Protection in Audio Power Amplifiers’ Electronics World,
September 2002, pg. 47, & October 2002, pg. 50.
while I agree 22K R5 looks bad I don't really think it plays the biggest role in "quiescent current imbalance"
it does cause a ~ 10 V Vce diff between the supposedly matched input diff pair Qs, and it causes miller multiplication of Q2 Ccb which interacts with the feedback network
the Vce diff will cause differential Q1,2 operating temp and consequent V offset (greatly reduced by just gluing/clamping Q1,2 together for lower dT), as well as minor hfe diff from early effect and therefore some ib diff too
but I would guess the worse effect is the probable Q1,2 Ic mismatch from the tail current and VAS (TIS for Mike) bias arrangement which has little to do with 22K R5
the designer apparently relies on faith that the tail current set by either a led or diode string and fixed resistor R5 will be precisely 2x of the current needed to bias undegenerated Q4 Ic to match Q5 Ic – as is required to balance Q1,2 Ic
there is no reason to expect Q1,2 Ic will match to even 10% in production without some explicit trimming arrangement - current mirror load seems the "simpler" approach when you really want to balance the diff pair Ic current to minimize distortion in the input differencing operation - and of course Vos will be improved by the higher gain
it does cause a ~ 10 V Vce diff between the supposedly matched input diff pair Qs, and it causes miller multiplication of Q2 Ccb which interacts with the feedback network
the Vce diff will cause differential Q1,2 operating temp and consequent V offset (greatly reduced by just gluing/clamping Q1,2 together for lower dT), as well as minor hfe diff from early effect and therefore some ib diff too
but I would guess the worse effect is the probable Q1,2 Ic mismatch from the tail current and VAS (TIS for Mike) bias arrangement which has little to do with 22K R5
the designer apparently relies on faith that the tail current set by either a led or diode string and fixed resistor R5 will be precisely 2x of the current needed to bias undegenerated Q4 Ic to match Q5 Ic – as is required to balance Q1,2 Ic
there is no reason to expect Q1,2 Ic will match to even 10% in production without some explicit trimming arrangement - current mirror load seems the "simpler" approach when you really want to balance the diff pair Ic current to minimize distortion in the input differencing operation - and of course Vos will be improved by the higher gain
Tube_Dude said:
Put a O.2 Ohms in the output of a amp ,and give a listen...
Have you ever tried it?
No.
But 0.33 Ohm + B&W 804S + Krell KST100.
I could redo it tomorrow evening.
Made me a little bit regret to have bought the fatter cables.
OTOS the price of copper has risen since then.
Gerhard
Y'know,
All these 'appalling' designs should make Mikeks really happy!!
While we might like to gloss over the fact he does not (at least to our limited knowledge here in this forum) have a product out there competing with all this junk, their obvious incompetence makes him look terrific!!
I feel all properly socialised genii should be grateful to the idiots of this world - they make the genius really stand out.
Mike, you stand out. Why the outrage? Try a little circumspection!
Cheers,
Hugh
All these 'appalling' designs should make Mikeks really happy!!
While we might like to gloss over the fact he does not (at least to our limited knowledge here in this forum) have a product out there competing with all this junk, their obvious incompetence makes him look terrific!!
I feel all properly socialised genii should be grateful to the idiots of this world - they make the genius really stand out.
Mike, you stand out. Why the outrage? Try a little circumspection!
Cheers,
Hugh
The fact of the matter is people post rubbish on this site anyway
imbalance would cause second harmonic which is not bad
Mikeks i know you have a perfect amp in mind with min distortion
and dcoffset, all you mention is correct, but people want amps
to have character and this can only be achieved by making
amp different, some dont have the knowledge to design the
amp you have in mind anyways
lets see an example of your amp so we can understand better
what you have in mind, you also mentioned the amp not having
dcprotection whats that got to do with performance yes could
be important to limit the amp within a range where the best
performance exists
this site is more of ideas not complete well designed products
this idea is rubbish but nevertheless it does alter the circuit
somewhat which is not all that bad
the same could be said about np products wasting so much
energy in idle state
your posts are appreciated and yes are correct but to which amp
is best sonicly is another matter all together, maybe yours is or
may depend who is listening to it
john
jacco vermeulen said:
You conveniently cited just one word and dropped the other 98 %.
In fact, the answer was more than yes.
And after trying some of their circuits some 30 years ago, I never ever
felt a drive to repeat that Elektor experience. And if I did, I wouldn'd
dare to admit. I make my living on professional electronics.
There are enough good sources such as Williams, Gilbert, Widlar,
Self, Pease, Johnson/Graham, Motchenbacher, Gray&Meyer, Rhea,
Hill, Tietze/Schenk. Don't need no Elektor.
Gerhard
Hi jh6you how are you
how is your headphone amp performing, when i listen to mine
i miss the sound stage and real performance effects i also get
wearing fatigue and eardrum overload, i have come to the conclusion that loudspeaker amps is what ireally enjoy the
sonics are just breathtaking, my headphones amps were state
of art devices but when i compare it to loudspeaker thats it
time to give it away,headphone listening is a big compromise
i just cant give up real listening
secrets in bins,tell that to our #1 friend here Mikeks
john
how is your headphone amp performing, when i listen to mine
i miss the sound stage and real performance effects i also get
wearing fatigue and eardrum overload, i have come to the conclusion that loudspeaker amps is what ireally enjoy the
sonics are just breathtaking, my headphones amps were state
of art devices but when i compare it to loudspeaker thats it
time to give it away,headphone listening is a big compromise
i just cant give up real listening
secrets in bins,tell that to our #1 friend here Mikeks
john
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.