Hello
Apparently many of the commercially successful amplifiers are SEPP designs.
e.g.
Sony Spontaneous twin drive
Yamaha Natural Sound
Denon High Current
...
However here seem to be few solid state SEPP circuits here on DIY Audio and also elsewhere on DIY forums. At least the search function finds little.
Why ? Can anyone point me to good solid state SEPP designs ?
Cheers
Gerry
Apparently many of the commercially successful amplifiers are SEPP designs.
e.g.
Sony Spontaneous twin drive
Yamaha Natural Sound
Denon High Current
...
However here seem to be few solid state SEPP circuits here on DIY Audio and also elsewhere on DIY forums. At least the search function finds little.
Why ? Can anyone point me to good solid state SEPP designs ?
Cheers
Gerry
I don't think these brand/ audio product descriptions are specifically SEPP topology. Most Yamaha and Denon products are ordinary, full complementary PP class AB designs whilst in solid state, SEPP is now usually a class A technique.
Single ended push-pull are a class of amplifiers which output stage uses devices of the same polarity. For their operation they need therefore one phase splitter stage, while complementary symmetry amplifiers do not need this.
In the solid state the two transistors are connected in series with respect to battery power, while normally in tube circuits the two valves are connected in parallel to the power supply.
In the case of Solid State the load sees two different impedances (a transistor comes out of emitter, the other of collector). while in the tubes, the output is better balanced (both valves come out of anode).
In my opinion it is more fair to compare the SEPP transistor configuration to socalled "Accordion" of valve circuit. In this latter topology, in fact, the two tubes are connected in series as in solid state (see the attached pdf) comparing it with a my SEPP transistor scheme .
Francesco
Thanks for the explanation francesco.
I firstly thouht that a SEPP amp had a single
LTP than a phase splitter to use a complementary push-pull VAS and OPS.
For example:
http://www.diyaudio.com/forums/solid-state/243842-gla-good-little-amp-survivor.html
I firstly thouht that a SEPP amp had a single
LTP than a phase splitter to use a complementary push-pull VAS and OPS.
For example:
http://www.diyaudio.com/forums/solid-state/243842-gla-good-little-amp-survivor.html
This design use complementary output pair in follower configuration (3 stages in cascades) named triple EF. As mentioned In this circuit the OPT stage not need of phase splitter. Infact bases of Q10 and Q11 receive the SAME signal (a cap linkes the two bases)
The double LTP are used not for phase inverter, but for higher CMRR against esternal and power supply noise.
Quasi Complementary output stage is formed by 2 stage: first stage (the driver) is realized with complementary devices that provide the two out of phase signals for the two power devices of same sex in SEPP configuration, therefore the name QUASI.
true SEPP in solid state ?
Hello Francesco,
thank you for your comments re Single-ended Push Pull. Googling brought me this Sony SEPP patent with MOSFETS. It uses dual +/- power rails
Transistorized single ended push-pull amplifier having no cross-over distortion - Sony Corporation
(dont miss the pdf link with the circuits)
Is that true SEPP in solid state in your mind ?
Here the Sony TA-F550ES which seems to employ this design idea
Sony TAF550ES Service Manual free download,schematics,datasheets,eeprom bins,pcb,repair info for test equipment and electronics
Also here the circuit of a Denon 1500, (despite the name the link is correct) .
which employs their "high current single push-pull" circuit.
Denon PMA-500AE Service Manual free download,schematics,datasheets,eeprom bins,pcb,repair info for test equipment and electronics
Are they true SEPP according to your definition? To me it looks like it.
Gerry
Hello Francesco,
thank you for your comments re Single-ended Push Pull. Googling brought me this Sony SEPP patent with MOSFETS. It uses dual +/- power rails
Transistorized single ended push-pull amplifier having no cross-over distortion - Sony Corporation
(dont miss the pdf link with the circuits)
Is that true SEPP in solid state in your mind ?
Here the Sony TA-F550ES which seems to employ this design idea
Sony TAF550ES Service Manual free download,schematics,datasheets,eeprom bins,pcb,repair info for test equipment and electronics
Also here the circuit of a Denon 1500, (despite the name the link is correct) .
which employs their "high current single push-pull" circuit.
Denon PMA-500AE Service Manual free download,schematics,datasheets,eeprom bins,pcb,repair info for test equipment and electronics
Are they true SEPP according to your definition? To me it looks like it.
Gerry
Hi,
the Sony TAF-550ES features a classical Emitter follower PushPull output stage, with the Darlingtons beeing of different Gender, while master and slave transistors of each Darlington beeing of same Gender.
The Denon PMA 500/1500 features a quasi complementary PushPull output stage, where the upper half is a Darlington with same gender master and slave transistor, while the other half is a complementary Feedback pair (CFP or compound) formed by different genders of Master and slave. The genders of the masters defining the genders of the combined transistor pairs.
As the genders of the master are different here, the circuit forms a quasi complementary output. The Denon is a bit more special as the slaves are N-channel MOSFETs.
I admit beeing not familiar with the amp circuits of the companies cited in #1.
And for sure Sony and Yamaha have had very interisting off-of-the-standard amps in their portfolio, namely the Yamaha B2 and B6, or the Sony VFET Poweramps. Iirc SEPPs were not among those.
SEPP means that upper and lower half of the stage are made from identical devices, which are connected in series between the supply lines and which are driven out of phase.
The output is taken from a midpoint between the devices.
See an example here:
SE and SEPP Buffer
The circuit to the left is a SE buffer stage. The upper half forming a follower driven by the input signal. The lower half forming a constant current source.
The circuit to the right is a SEPP buffer stage. The upper half forming a follower driven by the input signal. The lower half forming a modulated current source, driven by a signal generated by the voltage drop over a resistor in the upper half (R102). This generated signal is out-of-phase with the input signal, so that upper an lowr half work in antiphase or PushPull.
Variations of the above but using the same basic structure are found here:
CFP²-Hybrid-Super Buffer and Compound-Super Buffer
BTW: These examples include the unique(?) possibility to compare SE drive to SEPP drive by simply opening the connection from the modulating resistor in the upper half to the input cap of the lower half without changing anything else.
jauu
Calvin
the Sony TAF-550ES features a classical Emitter follower PushPull output stage, with the Darlingtons beeing of different Gender, while master and slave transistors of each Darlington beeing of same Gender.
The Denon PMA 500/1500 features a quasi complementary PushPull output stage, where the upper half is a Darlington with same gender master and slave transistor, while the other half is a complementary Feedback pair (CFP or compound) formed by different genders of Master and slave. The genders of the masters defining the genders of the combined transistor pairs.
As the genders of the master are different here, the circuit forms a quasi complementary output. The Denon is a bit more special as the slaves are N-channel MOSFETs.
I admit beeing not familiar with the amp circuits of the companies cited in #1.
And for sure Sony and Yamaha have had very interisting off-of-the-standard amps in their portfolio, namely the Yamaha B2 and B6, or the Sony VFET Poweramps. Iirc SEPPs were not among those.
SEPP means that upper and lower half of the stage are made from identical devices, which are connected in series between the supply lines and which are driven out of phase.
The output is taken from a midpoint between the devices.
See an example here:
SE and SEPP Buffer
The circuit to the left is a SE buffer stage. The upper half forming a follower driven by the input signal. The lower half forming a constant current source.
The circuit to the right is a SEPP buffer stage. The upper half forming a follower driven by the input signal. The lower half forming a modulated current source, driven by a signal generated by the voltage drop over a resistor in the upper half (R102). This generated signal is out-of-phase with the input signal, so that upper an lowr half work in antiphase or PushPull.
Variations of the above but using the same basic structure are found here:
CFP²-Hybrid-Super Buffer and Compound-Super Buffer
BTW: These examples include the unique(?) possibility to compare SE drive to SEPP drive by simply opening the connection from the modulating resistor in the upper half to the input cap of the lower half without changing anything else.
jauu
Calvin
Last edited:
Calvin, thanks. Very informative! Can you please also comment on the Sony TA-F590ES with Mosfets. I have attached the schematic.
Thanks a ton.
Gerry
Attachments
Single Ended Push Pull aka Quasi Complimentary is likely seen as unnecessary today. SEPP is more of a tube terminology.
Quasi Comp dates from the time where (as far as Silicon goes) PNP power transistors were inferior and more expensive to their NPN counterparts, so it was more favourable to use NPN power devices in the output stage. Nowadays, this is not a problem - PNP's are just as good as NPN's.
It is sometimes seen with MOSFETs because N and P channel devices often have quite different characteristics. One design I know of off hand is Arcam's "Alpha" amps that used the IRF540N device.
As for sound quality ? IMO, quasi or not is irrelevant to it.
Quasi Comp dates from the time where (as far as Silicon goes) PNP power transistors were inferior and more expensive to their NPN counterparts, so it was more favourable to use NPN power devices in the output stage. Nowadays, this is not a problem - PNP's are just as good as NPN's.
It is sometimes seen with MOSFETs because N and P channel devices often have quite different characteristics. One design I know of off hand is Arcam's "Alpha" amps that used the IRF540N device.
As for sound quality ? IMO, quasi or not is irrelevant to it.
Hi sony patent is related to SEPP output stage employing Vfet.
If you like Vfet visit my site:
AMPLIMOS one stage amplifiers, amplificatori audio monostadio
Sony TA-F550ES is a classic complementary out stage: one device is NPN (SANKEN 2SC2922) other device is PNP (SANKEN 2SA1216).
Denon amplifier employ a quasi-complementary output stage, which driver is realized with a couple of NPN/PNP BJT and final stage with 2 N mosfet (in sepp).
Francesco
If you want a clear picture of SEPP, SRPP, and "mu follower" output stages,
then you might want to take a look at John Brodsky's Tube CAD Journal.
http://www.tubecad.com/may2000/
then you might want to take a look at John Brodsky's Tube CAD Journal.
http://www.tubecad.com/may2000/
Last edited:
Two SEPP examples (not your textbook or garden variety), one MOS the other bipolar (can be MOS or darlington as well, both have been successfully implemented):
http://www.diyaudio.com/forums/soli...-dirty-its-circlomos.html?highlight=Circlomos
http://www.diyaudio.com/forums/soli...le-cheap-circlophone.html?highlight=Circlomos
And one SRPP bjt/MOS hybrid (minimalist, but functional):
http://www.diyaudio.com/forums/soli...amplifier.html?highlight=transistor+amplifier
http://www.diyaudio.com/forums/soli...-dirty-its-circlomos.html?highlight=Circlomos
http://www.diyaudio.com/forums/soli...le-cheap-circlophone.html?highlight=Circlomos
And one SRPP bjt/MOS hybrid (minimalist, but functional):
http://www.diyaudio.com/forums/soli...amplifier.html?highlight=transistor+amplifier
The expression Single Ended Push Pull does not make sense. Irrespective of device polarity and device technique, in the presence of two driving forces of similar strength and opposing orientation in a parallel alignment, phase shifted by half wavelength, the transfer characteristic is said to be centrosymmetric. It means a distinct functional difference and a substantial degradation from a distortion standpoint compared to the asymmetric charge distribution, such as in the Single Ended topology.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.