zeus amp old DIY
Further thougths, I have a pair of jordans and I would dearly love to construct this amp. At the moment I am listening to mark fenlon's 4inch units full range driven by moolys amp the idea being the neg feedback, clamps the coil.incidentaly, mark fenlon is an english engineer who worked with ted jordan for a while, he is on the internet and his speakers are available from mouser.
Further thougths, I have a pair of jordans and I would dearly love to construct this amp. At the moment I am listening to mark fenlon's 4inch units full range driven by moolys amp the idea being the neg feedback, clamps the coil.incidentaly, mark fenlon is an english engineer who worked with ted jordan for a while, he is on the internet and his speakers are available from mouser.
Input transformer. Plain. You can balance. Secondary with a middle outlet. 2 MOSFET with a common drain (push-pool follower). Low voltage. Large current. Class AB bias circuit without shutdown. Output transformer with special technology. With middle bend. Secondary winding for loudspeaker. Low distortion (0.18%). Wide band 20Hz-100kHz. High power. Short signal path.
The peculiarity is in the self-production of the output transformer. Susan explains in detail. The input transformer can be designed to operate from a DAC or a sound card (2v 600 ohm). Balanced or unbalanced.
zeus amp
OK GUYS Letts follow susan, bias is set at afew hundred milliamps ok, q1 andq2 are used as gate followers which alter the voltage on, and thus the current flowing through the output transformers primaries, nothing new here.
the center common is to ground. of course it will not work otherwise. any input signal on t1 is magnified by the turns ratio and causes the q1 andq2 to follow the voltage on their gates, one device rising the other falling.of course this is how amps work, so nothing new here.I do not understand what susan means when she says the output tx generates the negative rail surely the power supply does that? this not meant to be a criticisme of susan as she is far more educated than me
OK GUYS Letts follow susan, bias is set at afew hundred milliamps ok, q1 andq2 are used as gate followers which alter the voltage on, and thus the current flowing through the output transformers primaries, nothing new here.
the center common is to ground. of course it will not work otherwise. any input signal on t1 is magnified by the turns ratio and causes the q1 andq2 to follow the voltage on their gates, one device rising the other falling.of course this is how amps work, so nothing new here.I do not understand what susan means when she says the output tx generates the negative rail surely the power supply does that? this not meant to be a criticisme of susan as she is far more educated than me
This amplifier is simple and not stupid, but smart. It's like seeing a final equation after been simplified. You can go around and around and then you will realize why Susan concluded her design in this way.
In this amplifier the active parts provide no gain, actually the gain is less than 1 because it's based on gate followers design. The gain is done by the interstage transformer (or input transformer) that transforms the low voltage low impedance signal from the line stage to a higher voltage higher impedance sufficient for driving the gates of the two mosfets that btw are not complimentary. The final version of Zeus also uses depletion mode mosfets that get biased by the voltage drop through the output transformer internal DC resistance. So the bias does not require active parts.
And yes the B+ of 37V 700mA (single supply; there is no -37V) applied on the 2 drains flows to the grounded centre tapping of the transformer and when the voltage at the source of the two mosfets start swinging it generates positive and negative voltage on the secondary. For example assuming 1:1 ratio you will have in theory -37V to +37V swing. There is no need of positive and negative power supply rails because the transformer generates them.
The distortion of this amplifier reaches a certain value and it remains nearly constant with the increase of output power.
In this amplifier the active parts provide no gain, actually the gain is less than 1 because it's based on gate followers design. The gain is done by the interstage transformer (or input transformer) that transforms the low voltage low impedance signal from the line stage to a higher voltage higher impedance sufficient for driving the gates of the two mosfets that btw are not complimentary. The final version of Zeus also uses depletion mode mosfets that get biased by the voltage drop through the output transformer internal DC resistance. So the bias does not require active parts.
And yes the B+ of 37V 700mA (single supply; there is no -37V) applied on the 2 drains flows to the grounded centre tapping of the transformer and when the voltage at the source of the two mosfets start swinging it generates positive and negative voltage on the secondary. For example assuming 1:1 ratio you will have in theory -37V to +37V swing. There is no need of positive and negative power supply rails because the transformer generates them.
The distortion of this amplifier reaches a certain value and it remains nearly constant with the increase of output power.
having no voltage gain, and no global feedback, this amp has no hangups...this what makes this amp unique...
b+of 37 volts dc, makes an ac swing of 74 volts peak to peak, assuming no losses, peak of 37v and 26vac rms....not into 8 ohms this is 84 watts sine power...in real world, this is less..
Many thanks for these posts...
Dear "apelizzo", "OldDIY", "BJosephs", and others,
Appreciate the time you have taken to answer the questions raised by "AMPMAN".
I hope everyone remains safe and well.
Best wishes to all,
Susan.
Dear "apelizzo", "OldDIY", "BJosephs", and others,
Appreciate the time you have taken to answer the questions raised by "AMPMAN".
I hope everyone remains safe and well.
Best wishes to all,
Susan.
Member
Joined 2009
Paid Member
I suspect the scholar of valve amplifiers would not struggle to see the elegance of this design, they would be particularly comfortable around transformers and inductors.
I cannot make sense of the term "gate follower". To me this is an ordinary push-pull topology.
The voltage at the source follow the voltage at the gate.
Gate follower.
Better name then common drain.
Gate follower.
Better name then common drain.
The Zener diode on the gate is a protection device, it is NOT doing anything in normal operation.
Without the Zener diode, MOSFETs go BOOM!!!
Do not use without a gate-protection Zener, the Zener is your friend 🙂
Without the Zener diode, MOSFETs go BOOM!!!
Do not use without a gate-protection Zener, the Zener is your friend 🙂
However, most scholars of valve amplifiers see a solid-state device and immediately discount the idea because sand = anathema.
well, times have changed, and sand can actually bring out the best in tubes....
for example, no tube can ever come close to the gm of a mosfet..so that for followers, mosfet rules...
for example, no tube can ever come close to the gm of a mosfet..so that for followers, mosfet rules...
Susan,
Gate should preferably be referenced ohmically to source (here by a rather low value resistor).
Zener diodes are notoriously slow devices and provide poor protection.
The Hitachi LFETs come with more reliable built-in protection diodes.
Zener diodes are not nice devices.
Your posts emanate an aura of serenity.
Gate should preferably be referenced ohmically to source (here by a rather low value resistor).
Zener diodes are notoriously slow devices and provide poor protection.
The Hitachi LFETs come with more reliable built-in protection diodes.
Zener diodes are not nice devices.
Your posts emanate an aura of serenity.
And yet, they are ubiquitous for clamping the gate-source voltage on mosfet followers exactly as shown here. What protection does a resistor of a "rather low value" provide?
The input transformer has an HF frequency response of the order of 200KHz for a sine wave, which is within a typical Zener response, see:
https://www.onsemi.cn/PowerSolutions/document/HBD854-D.PDF
Page 27... Figure 27. "Zener Impedance versus Frequency"
In normal operation, the Zener is mute.