LND150 is a depletion type mosfet...will it be possible to use enhancement mode type like irf610 in its position?
Just tell me how the right EL84 gets the drive signal, when it's grid is grounded, and it sees a constant voltage at its cathode due to the LEDs.
The Mosfet is just a source follower. The upper 35k resistor doesn't do anything than dropping DC voltage. Both its ends are at AC-Ground (at the top via the PSU, the bottom via the 0.1µ)
In the schematic, this is basically a SE Amplifier with a PP-OPT with only one half of its winding used dynamically. The other EL84 is just prevents saturation of the OPT.
The Mosfet is just a source follower. The upper 35k resistor doesn't do anything than dropping DC voltage. Both its ends are at AC-Ground (at the top via the PSU, the bottom via the 0.1µ)
In the schematic, this is basically a SE Amplifier with a PP-OPT with only one half of its winding used dynamically. The other EL84 is just prevents saturation of the OPT.
It's not grounded. The triangle like thing isn't a ground symbol. Compare it to the ground symbol attached to the line across the bottom. You will realize that the two triangle like things are connected together (an on page port in Cadence speak). Then the circuit makes sense.
The two 500K resistors at the gate of the fet waste half the gain to get the gate DC voltage closer to center. You can get the gain back by putting a cap across the top resistor, or you can try eliminating both of them. I made a similar circuit for a guitar amp and there was plenty of head room to drive the output tubes to square waves before clipping the PI, so I left out the resistors.
Most any fet will work, but it is best to choose one with low capacitances that remain constant at the voltages that it will see in the circuit. The LND150 is very good in this regard, but there are others.
Last edited:
the_manta:
As tubelab said, the arrows are standard Cadence notation for on-page connectors. The way I was taught in my EE classes is that upward pointing arrows (usually) denote connections to the positive power supply and should be filled in, sideways hollow arrows denote a connection to another circuit element, and downward arrows are usually for signal or digital ground. Power Ground is usually depicted as three horizontal lines in a triangle shape pointing downward. The sideways arrows are usually labeled, but the labeling can be omitted if they all connect to each other. The reasoning is that it can remove a number of lines on the page making the circuit diagram easier to read.
0feedback:
No problem with the series connections. They run about 5% below nominal voltage, 24v vs 25.2v, but close enough.
Junm:
Any MOSFET should work, but lower capacitance is better. The target for a concertina phase splitter input voltage is about 25-33% of B+ to maximize swing, although in this situation I only need about 12v of swing each way (about 5% of B+), so the set point can probably safely go between 15% and 40%, but I'd rather keep it in the center of the usable zone. According to the 6N1P datasheet, with B+ 240v, a 30K anode load, and a 1.9v LED, I should be running at about 115v across the tube. That says the anode should be at about 117v, so the voltage divider gives me about 58 volts, which is about 25% of the B+.
In this circuit, the first stage is giving my a gain of about 30. Divided by two, the phase splitter puts out about 15x the input signal. I only need +/- 12v of drive for the 6P14Ps, so an input signal of +/- 0.8v will drive the amp to full power.
If I changed the top resistor of the voltage divider to 250K, I would have an effective gain of about 20 (needing only +/- 0.6v for full power) and a concertina set point of 80v, or around 33% of B+.
As tubelab said, the arrows are standard Cadence notation for on-page connectors. The way I was taught in my EE classes is that upward pointing arrows (usually) denote connections to the positive power supply and should be filled in, sideways hollow arrows denote a connection to another circuit element, and downward arrows are usually for signal or digital ground. Power Ground is usually depicted as three horizontal lines in a triangle shape pointing downward. The sideways arrows are usually labeled, but the labeling can be omitted if they all connect to each other. The reasoning is that it can remove a number of lines on the page making the circuit diagram easier to read.
0feedback:
No problem with the series connections. They run about 5% below nominal voltage, 24v vs 25.2v, but close enough.
Junm:
Any MOSFET should work, but lower capacitance is better. The target for a concertina phase splitter input voltage is about 25-33% of B+ to maximize swing, although in this situation I only need about 12v of swing each way (about 5% of B+), so the set point can probably safely go between 15% and 40%, but I'd rather keep it in the center of the usable zone. According to the 6N1P datasheet, with B+ 240v, a 30K anode load, and a 1.9v LED, I should be running at about 115v across the tube. That says the anode should be at about 117v, so the voltage divider gives me about 58 volts, which is about 25% of the B+.
In this circuit, the first stage is giving my a gain of about 30. Divided by two, the phase splitter puts out about 15x the input signal. I only need +/- 12v of drive for the 6P14Ps, so an input signal of +/- 0.8v will drive the amp to full power.
If I changed the top resistor of the voltage divider to 250K, I would have an effective gain of about 20 (needing only +/- 0.6v for full power) and a concertina set point of 80v, or around 33% of B+.
This LND150 vs. 12AX7 topic kept me thinking. If my memory is right the FET acts more like a pentode than like a triode. The FET and the pentode, unlike the triode, transfer input voltage to plate/drain current and the working resistor turns that back into voltage. The LND150´s curve is nice and flat. And here is my question:
Has anyone done this experiment to a hifi preamp or gainstage and how does is affect sonic performance and "tube sound". Preferably in a system without global nfb.
It is the old question that bugs me: can tube sound be achieved with one single triode in the signal path?
Has anyone done this experiment to a hifi preamp or gainstage and how does is affect sonic performance and "tube sound". Preferably in a system without global nfb.
It is the old question that bugs me: can tube sound be achieved with one single triode in the signal path?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.