Only N-Channel MOSFETs (NMOS); better Audio from non complements by Audio Power?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
tiefbassuebertr

tiefbassuebertr

Member
Joined 2005
tiefbassuebertragung.de
Different Versions of only N-Channel MOSFETs in Audio Power Output Stages

I have read the comments by URL
http://www.diyaudio.com/forums/showthread.php?s=&threadid=130463 and
http://www.diyaudio.com/forums/showthread.php?s=&threadid=40065&perpage=25&pagenumber=1
and my idea was to start a new thread, where I begin an URL overview of different versions concerning power output stages with only N-CH MOSFETs. I was amazed about the wide range of circuit topologies regarded this kind of output power stages for audio amplifiers.

Until this day I have found the follow topologies

1) driving through differential driver stages
2) driving through phase splitting transistor (Totem-Pole like Bengt Ollson, Futterman or MOSFET-DOZ)
3) driving through phase splitting driver transformers
4) driving through current mirror phase shifter
5) Quasi complementary output stages
6) Version orf German's company "Protovision"
7) Version of NAD integrated amp "306"
8) driving through two op amps
9) Circlotron (lowest distortion)

Here my URL collections in this case

1) driving through differential driver stages like Hiragas Kaneda
http://www.amplimos.it/images/2SK1056AMP.bmp (tube diff amp driver for 2x2SK1056 output)
http://www.amplimos.it/images/2sk77...AMAHA B-1.gif Yamaha B1, diff-cascode driver stage
http://peufeu.free.fr/audio/schemas/Kaneda_Mosfet.jpg Kaneda, diff-cascode driver stage
http://www002.upp.so-net.ne.jp/VivaMari/audio_life/UHC_MOS_Amp.html similar to Caneda and B1
http://www.patentstorm.us/patents/5631608/claims.html (diff amp driver for N-channel VFET output)
http://www.wikipatents.com/5631608.html (diff amp driver for N-channel VFET output)
United States Patent 5783970

2) driving through phase splitting transistor (Totem-Pole like Bengt Olson, Futterman or MOSFET version of Elliott's DOZ)
http://www.tubecad.com/april_may2001/page8.html ("The Symmetrical and the Non-Symmetrical")
www.tubecad.com/articles_2001/totem...utput_Stage.pdf (split load unity gain circuits)
http://sound.westhost.com/jll_hood.htm (basic description Totem Pole)
http://www.tubecad.com/april_may2001/page17.html (Totem Pole Power Buffer Gain <1)
http://www.tcaas.btinternet.co.uk/index-1.htm The JLH Class-A Amplifier, transistor phase splitting
http://www.amplimos.it/images/N-CH1.JPG (Olsson "Only N-Channel", transistor phase splitting driver stage)
Original article: Better Audio from non complements? from Bengt Olsson (B. Olson?)
Electronics World + Wireless World, December 1994 page 988
http://www.aussieamplifiers.com/ only N-Channel PCB power amplifier unit's
http://www.diyaudio.com/forums/show...p?threadid=4476 (Ollson scheme 07-20-2004 killerops zip)
http://sound.westhost.com/project36.htm (R.Elliot DoZ, Death of Zen, transistor phase splitting)
www.6moons.com/industryfeatures/zen/plh.pdf (PLH, Pass Linsley-Hood, switchable from SE to Totem pole)
http://www.diyaudio.com/forums/show...?threadid=14320 (phase splitting like Creek)

3) driving through phase splitting driver transformers like zeuz
http://www.diyaudio.com/forums/show...?threadid=24744 (driver transformer phase splitting)
http://www.susan-parker.co.uk/zeus.htm (CSPP? - transformer for input and output)
http://www.susan-parker.co.uk/zeus.htm (SEPP-source follower with input and output transformer)
http://www.tubecad.com/2004/blog0006.htm (Simple push pull)

4) driving through current mirror phase shifter
www.vegalab.ru/forum/attachment.php...57&d=1102664366 Current Mirror Driver + SERVO BIAS

5) Quasi complementary output stages like Zeta and old Pass A40
http://users.adam.com.au/cgpap/QuasiWeb/index.htm
Olive audio: lo longer online, please go to webarchive
http://web.archive.org/web/*/www.oliveaudio.com
http://web.archive.org/web/20050415070238/www.oliveaudio.com/index.php?page=2
(Home of the ZETA experimental N-Channel power amplifer, Lars Clausen
Enghavevej 12, DK7500 Holstebro, Denmark temporarely unfortunately only Class D
http://www.newclassd.com/index.php?page=96

6) Version orf German's company "Protovision", type "Reality 284"
http://www.singhkang.de/proto.html (Owner's Manual)
http://img530.imageshack.us/i/2841bb.jpg/ (Image top view)
for schematic go to my PDF schematics at bottom

7) Version of NAD integrated amp "306" go to my pdf schematic at bottom - topology not to find on the web (quasi complementary or not?

8) driving through second op amp for phase shifting
http://www.tubecad.com/2005/January/blog0034.htm "Taylor source follower"
http://www.datasheetcatalog.com/datasheets_pdf/T/D/A/7/TDA7294.shtml (integrated circuit TDA7294V)

9) Circlotron (lowest distortion aff all mentioned versions)
for looking URL collection go to http://www.diyaudio.com/forums/showthread.php?s=&threadid=146540

ok, the circlotron is best of all, but what about the eight other versions, that I have found? Thank you very much for your comments
 

Attachments

  • protovision nad306 .ckt.pdf
    44.8 KB · Views: 3,435
tiefbassuebertr

tiefbassuebertr

Member
Joined 2005
tiefbassuebertragung.de
some URLs are maimed/distort above:
To Number 1
http://www.amplimos.it/images/2sk77 amp YAMAHA B-1.gif Yamaha B1, diff-cascode driver stage
To Number 2
http://www.tubecad.com/articles_2001/totem-pole_output_stage_psrr/
http://www.diyaudio.com/forums/showthread.php?threadid=4476 (Ollson scheme 07-20-2004 killerops zip)
http://www.diyaudio.com/forums/showthread.php?threadid=14320 (Creek)
To Number 3
http://www.diyaudio.com/forums/showthread.php?threadid=24744 (driver transformer phase splitting)
To Number 4
www.vegalab.ru/forum/attachment.php?attachmentid=257&d=1102664366 N-CHANNEL AMP + SERVO BIAS (Current Mirror Driver + SERVO BIAS)

Unfortunately, I do not know what I've done wrong because the addresses look like exactly as before
 
Miles Prower

Miles Prower

Member
Joined 2005
dolphin-hsl.com
"Only N-Channel MOSFETs (NMOS); better Audio from non complements by Audio Power? "

Definitely yes. "Complimentary" N-Channel/P-Channel MOSFETs are a good deal less "complimentary" than NPN/PNP complimentaries. Stick to one or the other, but not both in the same circuit. MOSFETs aren't BJTs, and topologies originated for BJTs just won't work as well. Of course, the Big Box manufacturers like those OTL BJT topologies since no OPT is cheaper than the cheapest OPT.
 
qsa

qsa

Member
Joined 2007
qsaelectronic.com
I think NMOS output stages working more stable without faults compare to BJT against SOA

Have sold 2 years ago 50 pcs. Mosfet PA amplifier.... since today 1 channel defective by clients, similar Class D amplifier faults less 1 %

but BJT amplifier/Toshiba 2SA1943 und 2 SC5200 inside, .... headache about 10 % faults/defective output devices,...amps from different professional supplier
 
tiefbassuebertr

tiefbassuebertr

Member
Joined 2005
tiefbassuebertragung.de
AndrewT said:
is it compulsory to read through all those links before we are allowed to post a comment or question?
Oops, just broke that rule!
Click to expand...

That's no true. I have make the subdivisions 1) until 9) for better overview


KSTR said:
You used literal copies (cut&paste from screen), which when abreviated with "..." in between obviously give wrong URL's ;)
Sorry for another metacomment.
Click to expand...

Thank you for this advice. All URLs by the follow correction goes open - copies in this case I have make from the displayed address in the web address field

What means "metacomment" ?
 
tiefbassuebertr

tiefbassuebertr

Member
Joined 2005
tiefbassuebertragung.de
I see for most of electronics engineer that's very hard to find for the individual versions classify and enclose since there are in some circuits transitional forms between pure "single ended" and pure "push-pull" (In contrast to "Power Follower", i. e. Gain <1 and "Power amplification" respective "common source", i. e. Gain >1 that clearly is to differentiate).
If there are pure "single ended" version, no mode in class A/B operation is possible, only pure Class-A operation
If there are pure "push pull" version, both modes, class A and class A/B are possible

A few weeks ago I had the following two threads opened with an URL collection concerning "single ended", that means only "Pure Class A" - see
http://www.diyaudio.com/forums/showthread.php?threadid=146581 (overview of most circuits)
and http://www.diyaudio.com/forums/showthread.php?threadid=146528 (overview commercial amplifier devices)
and additional "Totem Pole" and CSPP (Single Ended related)" see
http://www.diyaudio.com/forums/showthread.php?s=&threadid=146540 (overview of most circuits)

Mr. Pass would filed all this kinds of circuit topologies to "Single-ended Class A systems, including SRPP and variants" because by operating in pure class A mode the measured differences by all this versions are relatively low.

My aim of this thread actually should be to make an overview about all known output stages with "only N-channel MOSFET" that not only can operated by pure Class A (i. e. also suitable is for push-pull operated in Class A/B with only 100 mA idle current through the MOSFETs). Unfortunately I forgot this to mention at the beginning.

I note by this circuit topologies, that I cannot say bei number 3), 4), 8) and 9) clearly the possibility of class A/B operation

1) driving through differential driver stages - class A/B is possible
2) driving through phase splitting transistor (Totem-Pole like Bengt Ollson, Futterman or MOSFET-DOZ) - class A/B is possible
3) driving through phase splitting driver transformers - only pure Class A ?
4) driving through current mirror phase shifter - only pure Class A ?
5) Quasi complementary output stages - class A/B is possible
6) Version orf German's company "Protovision" - class A/B is possible
7) Version of NAD integrated amp "306" - class A/B is possible
8) driving through two op amps - only pure Class A ?
9) Circlotron (lowest distortion) - only pure Class A ?

If Number 3), 4), 8) and 9) can use really only for class A, this versions must go out from this thread

when this is completely clear, I can go to the next step: Namely, to find out and investigate, which of these versions look the best suited for amplifiers repairs as to use for universal solution circuit. They would then have a cheap and good alternative for the "hard to find" MOSFET replace of many amplifiers, for which there is very expensive or no longer exists (most complementary MOSFETs). For example, there are no replacement for the famous Yamaha and Sony V-FETs and no replacement for the Hitachi lateral MOSFETs in the TO-3 outline (except only the very expensive BUZ900/901 from Semelab).
 
megajocke

megajocke

Member
Joined 2003
The QSC have its output FETs in tight feedback loops with one opamp each, making them contollable current sources. Parallelling can be done with only low-resistance current sensing resistors. The intricate PUPPETMSTR (I love that name :D) phase splitter balances the bridge and drives the current cells.

That Protovision circuit is quite close. If the output transistors are replaced by current cells it would be very close, though non-bridged and with a much simpler phase splitter...
 
Wavebourn

Wavebourn

Member
Joined 2006
wavebourn.com
megajocke said:
The QSC have its output FETs in tight feedback loops with one opamp each, making them contollable current sources. Parallelling can be done with only low-resistance current sensing resistors. The intricate PUPPETMSTR (I love that name :D) phase splitter balances the bridge and drives the current cells.

That Protovision circuit is quite close. If the output transistors are replaced by current cells it would be very close, though non-bridged and with a much simpler phase splitter...
Click to expand...

I did something similar with transformer output: I've converted such a way Altec transistor power amp to a hybrid (double triode: amplifier and concertina splitter). Opamps were driven by tubes. Sounded well on a bass guitar. :)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom