Collection of Class B topologies <100mA Idle and Sound closest by Class A

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Until now, it is still a challenge, to create a solid state power amplifier with <100 mA quiescence current through the power output devices and sound quality like pure class A at the same time.
Unfortunately I haven't heard such an amplifier until this day.
The same observation has also made Mr. Nelson Pass - go to
http://www.passlabs.com/pdf/articles/seclassa.pdf (third paragraph) and
http://www.passdiy.com/pdf/leaving_class_a.pdf

But below a collection of topologies, that most likely could reach the very clean sound of class-A without any harshness.
There are probably more of such circuit topologies and I want to know others (except Class D).
What about your audible experiences - what you would say, which topology is closest to Class A regarded the sound?
Thank you very much for your comments.

1) P.J. Walker Current Dumping Audio Amplifier"
http://220.232.128.43/audio/quad405/Current Dumping Audio Amplifier - PJ Walker.pdf
http://www.diyaudio.com/forums/solid-state/109771-quad-405-dcd-mod3-pcb.html
2) N.M.Visch "A Novell Class B output"
http://www.diyaudio.com/forums/solid-state/160285-class-b-w-o-crossover-distortion-1975-a.html
3) Yamaha's Hyperbolic Conversion Amplification (HCA) Circuit
http://www.diyaudio.com/forums/soli...lic-conversion-amplification-hca-circuit.html
4) NDFL = Nested Differentiating Feedback Loops
http://www.diyaudio.com/forums/solid-state/148066-new-cherry-ndfl-amp.html
http://www.diyaudio.com/forums/solid-state/96634-what-nested-feedback-how-realy-works-some-example
5) ERCO from diyaudio member "turnA_zpoint "
http://www2.plala.or.jp/puthoff/MJ.html
http://www2.plala.or.jp/puthoff/material/explanation_of_work_081030.pdf
http://www2.plala.or.jp/puthoff/material/EXP_MK2_V2.pdf
http://www.diyaudio.com/forums/soli...rly-index-1924-2009-wanted-2.html#post2104427 (Post # 11)
 
Last edited:
Current-dumping (feed-forward) applied to a conventional Class-B amplifier topology (a number of variants are possible):

http://www.diyaudio.com/forums/soli...ass-ab-self-type-ii-ef-o-p-3.html#post2102812

The darlington O/P version won't qualify as Class-A sonics, but variants that use a Class-AB Self Type-II EF output stage along with current-dumping get close to Class-A sonics.

Also, it is possible to combine current-dumping (feed-forward) with the Visch Class-B topology, which should yield sonics which are comparable to high-end Class-A.
 
Current-dumping (feed-forward) applied to a conventional Class-B amplifier topology (a number of variants are possible):

http://www.diyaudio.com/forums/soli...ass-ab-self-type-ii-ef-o-p-3.html#post2102812

The darlington O/P version won't qualify as Class-A sonics, but variants that use a Class-AB Self Type-II EF output stage along with current-dumping get close to Class-A sonics.

Also, it is possible to combine current-dumping (feed-forward) with the Visch Class-B topology, which should yield sonics which are comparable to high-end Class-A.

very interesting. Do you know commercial models uses this Vish Class-B topology?

Here various URLs with topologies, which all have one goal:
to outwit the disadvantages of Class B:

http://www.diyaudio.com/forums/solid-state/162559-new-class-super-non-switching-need-revival.html
http://www.diyaudio.com/forums/solid-state/149327-bootstrapped-classa.html
http://www.diyaudio.com/forums/solid-state/32254-new-class-whats-mean.html
http://www.diyaudio.com/forums/solid-state/922-technics-class-bias-circuits.html
http://www.diyaudio.com/forums/solid-state/136165-technics-class-biasing.html
http://www.diyaudio.com/forums/solid-state/162559-new-class-super-non-switching-need-revival.html
http://www.diyaudio.com/forums/solid-state/36789-technics-class-aa-sandman-bridge.html
http://www.diyaudio.com/forums/analog-line-level/155143-class-aa-opamp-schematic.html
http://www.diyaudio.com/forums/everything-else/48933-now-aa-down.html
http://www.diyaudio.com/forums/solid-state/47890-classes-operation-vs-sound-quality.html
http://www.diyaudio.com/forums/pass-labs/28949-pass-patent-5-343-166-a.html

For me follow topologies seems to be closest to pure class-A (have a look also to the attached PDF):
http://www.diyaudio.com/forums/solid-state/23590-audiotek-69-technics-se-a1.html
Power amplifier

Here some papers about THD measuring procedures and effects by Class-B:
ELECTRONICS & WIRELESS WORLD JULY1987 £1.95
Crossover distortion in class-B amplifiers
Detailed tests on three modes of amplifier operation, including
a non-switching class Btype, using the same basic circuit
produce a fewsurprises. E. MARGAN :
http://www-f9.ijs.si/~margan/Articles/Class_B_Dist.pdf

LM4702 driving the MOSFET output power buffer
http://www.national.com/an/AN/AN-1645.pdf
 

Attachments

  • Elektor class-B drives class A supply.pdf
    100.6 KB · Views: 479
Until now, it is still a challenge, to create a solid state power amplifier with <100 mA quiescence current through the power output devices and sound quality like pure class A at the same time.
Unfortunately I haven't heard such an amplifier until this day.
How about this one, it could qualify as it has an Iq=50mA:
http://www.diyaudio.com/forums/solid-state/185501-unigabuf-follower-cut-out-leader.html.
I have another contender, a sliding class A circuit, but the Iq is ~180mA.
 
very interesting. Do you know commercial models uses this Vish Class-B topology?

Here various URLs with topologies, which all have one goal:
to outwit the disadvantages of Class B:

http://www.diyaudio.com/forums/solid-state/162559-new-class-super-non-switching-need-revival.html
http://www.diyaudio.com/forums/solid-state/149327-bootstrapped-classa.html
http://www.diyaudio.com/forums/solid-state/32254-new-class-whats-mean.html
http://www.diyaudio.com/forums/solid-state/922-technics-class-bias-circuits.html
http://www.diyaudio.com/forums/solid-state/136165-technics-class-biasing.html
http://www.diyaudio.com/forums/solid-state/162559-new-class-super-non-switching-need-revival.html
http://www.diyaudio.com/forums/solid-state/36789-technics-class-aa-sandman-bridge.html
http://www.diyaudio.com/forums/analog-line-level/155143-class-aa-opamp-schematic.html
http://www.diyaudio.com/forums/everything-else/48933-now-aa-down.html
http://www.diyaudio.com/forums/solid-state/47890-classes-operation-vs-sound-quality.html
http://www.diyaudio.com/forums/pass-labs/28949-pass-patent-5-343-166-a.html

For me follow topologies seems to be closest to pure class-A (have a look also to the attached PDF):
http://www.diyaudio.com/forums/solid-state/23590-audiotek-69-technics-se-a1.html
Power amplifier

Here some papers about THD measuring procedures and effects by Class-B:
ELECTRONICS & WIRELESS WORLD JULY1987 £1.95
Crossover distortion in class-B amplifiers
Detailed tests on three modes of amplifier operation, including
a non-switching class Btype, using the same basic circuit
produce a fewsurprises. E. MARGAN :
http://www-f9.ijs.si/~margan/Articles/Class_B_Dist.pdf

LM4702 driving the MOSFET output power buffer
http://www.national.com/an/AN/AN-1645.pdf

I own (and use daily in my main system) a Technics SE-A5 wich uses the "New class A SynchroBias" circuit. I use it with horn speakers, 104dB, and I can confirm this is an amp that sounds clean and full even at very very low power (0,01watt in most of my evening listenings). Is it a result of the New class A circuit or other factors I cannot tell, but it sure sounds good. A lot of big SS amp i've tried so far were quite "grainy" and "dirty" when driving the same speakers.
 
You mean sound character is beautiful 6kHz? and up?
Exactly.
I own (and use daily in my main system) a Technics SE-A5 wich uses the "New class A SynchroBias" circuit. I use it with horn speakers, 104dB, and I can confirm this is an amp that sounds clean and full even at very very low power (0,01watt in most of my evening listenings). Is it a result of the New class A circuit or other factors I cannot tell, but it sure sounds good. A lot of big SS amp i've tried so far were quite "grainy" and "dirty" when driving the same speakers.
Thank you for this informations. I will check this, if I have a possibility.
This unity-gain follower is an audiological UFO:
It is adjustment-free, requires no thermal compensation, has no feeedback, no servo, can even work with fake or reject transistors, and yet offers low DC offset and sub-ppm linearity:
Thank you for this circuit - I will check the circuit in my CAD programm, If I have time.
 
Last edited:
ClassAB with bootstrap will have 10kHz sounds better than 6kHz. In high quality classA, 6kHz beautifully beat 10kHz and everything sounds beauty and real. I found it when play with damper characteristics.
ClassA drop characteristics has major cause. When classA using huge feedback to achieve lowest distortion, that drop characteristics is gone, and it sounds like common amplifier.

I post some picture here, may help too.
http://www.diyaudio.com/forums/solid-state/184725-alternative-topologies-3.html#post2514192
post#30
 
Class AA, VC4, Class S (Sandman) or "true current dumping" come close, if implemented correctly.

I own a SE-A100, which is already great in it's stock form, but with a little tweaking, it beats almost every Class A amp. This kind of amplifier architecture is good for distortion harmonics lower as -120dB at 20kHz.

Another option would be V-FETs (Si-SITs, not power Junction FETs), because these typically operate at Class B at currents around 200..300mA , while Bipolars and MOSFETs achieve the optimum Class B operation around 10..100mA. They also feature a zero tempco., which means, that the Class B idle current is not dependent on thermal history.

If you ever heard a Yamaha B-I or B-2 or a Victor JM-S7, you probably know, what I mean.
 
I concern myself only that class B cross smoothly at something significantly less than full turn-on drop of a Schottky diode. Don't worry too much what exact current that might be at any given temp. TO220 Schottkys don't run hot anyway, just please don't bolt em' to same sink as hot output devices...

I'm just sayin' that trying to control precise ClassB quiescent doesn't pay off as much as enforcing a smooth crossing, and let quiescence do as it please.
 
Last edited:
The URL about Harmans T-Circuit: "An Ultra-Low Distortion Direct-Current Amplifier" from the author
BART N. LOCANTHI (James B. Lansing Sound, Inc.)
www_harman.com/EN-US/OurCompany/Technologyleadership/Documents/Scientific%20Publications/1091.pdf
is death. Here the currently URL:
http://www.aes.org/tmpFiles/elib/20110325/1091.pdf

Concerning class AA I have found this PCB devices as line preamp:
http://cgi.ebay.co.uk/NE5534-NE5532...emQQptZLH_DefaultDomain_0?hash=item1c093942eb
is it possible to prepare this PCB's in such kind, that I can add power buffer stages?

Class AA, VC4, Class S (Sandman) or "true current dumping" come close, if implemented correctly.
which currently amplifier models uses one of this topologies ??

the statements about
http://www.passlabs.com/pdfs/articles/seclassa.pdf
seems to be still valid.

BTW - the head line of this thread should actually be
Collection of Class B topologies <100mA Idle and Sound closest by Single Ended Class A
instead
"Collection of Class B topologies <100mA Idle and Sound closest by Class A"
 
Last edited:
but with a little tweaking, it beats almost every Class A amp.

May I ask? what is that little tweak? and what the differences after tweaking?
Thanks alot before.

tiefbassuebertr said:
BTW - the head line of this thread should actually be
Collection of Class B topologies <100mA Idle and Sound closest by Single Ended Class A
instead
"Collection of Class B topologies <100mA Idle and Sound closest by Class A"

And without JFET input. I feel that using JFET is cheating.
 
in the meantime the AES links
http://www.aes.org/tmpFiles/elib/20110426/1091.pdf
http://www.aes.org/tmpFiles/elib/20110325/1091.pdf
don't goes open again. Therefore I have make splitting that pdf file for an attachement here.
have a look also by
Bart Locanthi

I am still looking for currently available new commercial amp products (except Class-D) with low heating but similar sound character than pure class A.
The model that I know which match closest this profile is the "Gaincard" (4706) from 47lab - go to
http://www.sakurasystems.com/products/47amp.html
http://www.6moons.com/audioreviews/47labs5/gaincard_2.html
But there must actually be much more in the meantime.
 

Attachments

  • An Ultra-Low Distortion Direct-Current Amplifier pg 290.pdf
    63.3 KB · Views: 320
  • An Ultra-Low Distortion Direct-Current Amplifier pg 291.pdf
    89.7 KB · Views: 223
  • An Ultra-Low Distortion Direct-Current Amplifier pg 292.pdf
    62.6 KB · Views: 228
  • An Ultra-Low Distortion Direct-Current Amplifier pg 293.pdf
    58.8 KB · Views: 192
  • An Ultra-Low Distortion Direct-Current Amplifier pg 294.pdf
    121.7 KB · Views: 223
Last edited:
Technics SE-A100, Yamaha MX-10000 and Sony TA-N7B

May I ask? what is that little tweak? and what the differences after tweaking?
Thanks alot before.

The Technics SE-A100 uses not very good OPamps for driving the power stage. With more recent ones like the LME series, you get better performance. The Class AA circuit also suffers from a kind of "thinking error", which let it operate only at sub optimal performance. The reason of this i don't know, probably patent infringement issues or whatsoever.

Nevertheless my SE-A100 does well against the mighty Yamaha MX-10000, which solves the feed forward problem with a brute force 7A standing current in the error correction stage, which means, that the MX is a room heater, while the SE-A100 is not.

Regarding the idle power, one could say, that Class AA is a kind of advanced Class B, while the HCA/Feedforward circuit of the MX-10000 is a kind of advanced Class A.

If the difference in sound (the MX is indeed a little better) really judges the approx. 400W idle power everybody has to decide on its own ...

For me the Class AA is one of the smartest approches to get the maximum performance/power with the least idle power, anyhow as many other advanced topologies the high complexity does not suggest DIY ...

There are certainly other solutions which are less complex, like e.g. the Class B power cascode (debuted 1976 with Bipolars/V-FETs in the Sony TA-N7B).

What are the advantages of such a topology ?

- Lower switching distortions due to low voltage high speed Bipolars (100 ... 200MHz Ft feasible)..
- Better linearity due to minimization of influence of Miller/Early effect.
- Less thermally induced crossover distortions, due to more thermally stable bias point.
 
Allow me to tease a little......
Regards
 

Attachments

  • 9121CMQUA-SCH-606-A1.gif
    9121CMQUA-SCH-606-A1.gif
    113.3 KB · Views: 709
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.