Hi George,
I thinked about your PowerDrive circuit... I really like it, I will add it to my amp.
My current favorite topology (for today, at least) is a differential pair (12AX7) with a CCS in the tail that will drive your PowerDrive, and, then, the output tubes.
I'm thinking also to add some local feedback, directly from the output tube's anode to the inverting input of the input longtail, just to linearize a little the amp response (if I have to build an amp, let's do as best as I can!!).
An idea was born in my mind: we are using CCS in almost everywhere in the amp, why do not use a CCS in the PowerDrive too?
If we replace the drain resistor with a CCS the PowerDrive should become more linear and more powerful, since its AC impedance should decrease to few Ohms, instead of being 10K.
This is what I was thinking of (this is a preliminary version, of course).
Is it useful, or is it a useless waste of time?
Ciao,
Giovanni
I thinked about your PowerDrive circuit... I really like it, I will add it to my amp.
My current favorite topology (for today, at least) is a differential pair (12AX7) with a CCS in the tail that will drive your PowerDrive, and, then, the output tubes.
I'm thinking also to add some local feedback, directly from the output tube's anode to the inverting input of the input longtail, just to linearize a little the amp response (if I have to build an amp, let's do as best as I can!!).
An idea was born in my mind: we are using CCS in almost everywhere in the amp, why do not use a CCS in the PowerDrive too?
If we replace the drain resistor with a CCS the PowerDrive should become more linear and more powerful, since its AC impedance should decrease to few Ohms, instead of being 10K.
This is what I was thinking of (this is a preliminary version, of course).
Is it useful, or is it a useless waste of time?
Ciao,
Giovanni
Power-Drive
Croccodillo-
Im using George's Power-drive in a P-P 811A amp--Its quite amazing, Such a difference in performance!
The 811A is a 'power-grid' type tube, and running at 450V plate, needs grid-current to flow--My grids on the 811A stand at around 20V positive for 60mA Ip, The power-drive delivers plenty of current to drive these beasts, so should perform wonderfully in your amp.
Ive found that once set, the bias doesnt need adjusting either--Its very stable, over many weeks of operation.......
I did modify the power-drive to include a 12V Zener diode at the gate-junction as I had an earlier failure due to my incompetance changing a driver valve, without shutting down.......
Croccodillo-
Im using George's Power-drive in a P-P 811A amp--Its quite amazing, Such a difference in performance!
The 811A is a 'power-grid' type tube, and running at 450V plate, needs grid-current to flow--My grids on the 811A stand at around 20V positive for 60mA Ip, The power-drive delivers plenty of current to drive these beasts, so should perform wonderfully in your amp.
Ive found that once set, the bias doesnt need adjusting either--Its very stable, over many weeks of operation.......
I did modify the power-drive to include a 12V Zener diode at the gate-junction as I had an earlier failure due to my incompetance changing a driver valve, without shutting down.......
You mentioned adding PowerDrive to an driver that already has a CCS in the cathode. Be careful trying to use two CCS's in the same circuit. If they are electrically in series and don't source exactly the same current, one will rail out. It may work in your case though, since they are not in series, and not set to the same current. I saw an amp with a CCS in the cathode, and another CCS in the plate circuit of the same triode, this won't work.
Feedback, is not always a good thing in a tube amp, especially a SE triode amp. It would be a good thing to use a pot for variable feedback. A little feedback tightens up the bass (lowering of output impedance) but too much removes the life from the music. Best electrical performance does not always correspond to best sound. If you have adjustable feedback, you will find that you will use different settings for different music. You need excess gain to account for the high feedback settings.
The CCS in the PowerDrive might be a good idea, I never tried it. If you try it, let us know what happens.
Alastair:
The 12 volt zener is probably a good thing, and I have added it to the schematic for the next revision of the web page. I have abused PowerDrive circuits many times without blowing a fet, but it wouldn't hurt. When I manage to blow up a PowerDrive, the CCS fails. This usually happens by poking around in a live circuit with a scope or voltmeter probe, and sparking something out.
Feedback, is not always a good thing in a tube amp, especially a SE triode amp. It would be a good thing to use a pot for variable feedback. A little feedback tightens up the bass (lowering of output impedance) but too much removes the life from the music. Best electrical performance does not always correspond to best sound. If you have adjustable feedback, you will find that you will use different settings for different music. You need excess gain to account for the high feedback settings.
The CCS in the PowerDrive might be a good idea, I never tried it. If you try it, let us know what happens.
Alastair:
The 12 volt zener is probably a good thing, and I have added it to the schematic for the next revision of the web page. I have abused PowerDrive circuits many times without blowing a fet, but it wouldn't hurt. When I manage to blow up a PowerDrive, the CCS fails. This usually happens by poking around in a live circuit with a scope or voltmeter probe, and sparking something out.
Hi,
Very good choice, I used the EL84 as Pentode as Driver for a 300B SE Amplifier. Enough gain to go from CD to 300B grid with plenty of reserve and when driving a triode output the penthode driver gives better harmonic distribution and a more delicate, more harmonically rich tone.
This shows a SV83 (and is hosted without my permission by Aquablue BTW), but an EL84 drops in with no changes.
This amplifier schematic was designed to be used with a serious valve regulated High Tension powersupply (2 X GZ34 Rectifier, 2 X 6AS7 Series Regulator, EF86 Error Amplifier and ZZ1000 Neon Reference) for HT and regultated (LT1084) 300B Heater supplies.
To convert to use of normal high tension supplies add a 2K2 & 100uF decoupling RC circuit in the supply to the EL84 circuit and delete the upper (10uF) Capacitor on the screengrid.
The original Circuit used a fairly smart High Tension Noise cancellation principle, which with the originally desiged supply (when set up correctly) manages << 0.1mV unweighted RMS output noise (or around -90db referenced to 1W Output or -99db referenced to nominal full power of 8W).
The EL84 circuit by itself has a gain of nearly 100, can swing well over 200V peak-peak at low distortion and has a reasonably low output impedance (around 10K).
Sayonara
croccodillo said:
Very good choice, I used the EL84 as Pentode as Driver for a 300B SE Amplifier. Enough gain to go from CD to 300B grid with plenty of reserve and when driving a triode output the penthode driver gives better harmonic distribution and a more delicate, more harmonically rich tone.
This shows a SV83 (and is hosted without my permission by Aquablue BTW), but an EL84 drops in with no changes.
This amplifier schematic was designed to be used with a serious valve regulated High Tension powersupply (2 X GZ34 Rectifier, 2 X 6AS7 Series Regulator, EF86 Error Amplifier and ZZ1000 Neon Reference) for HT and regultated (LT1084) 300B Heater supplies.
To convert to use of normal high tension supplies add a 2K2 & 100uF decoupling RC circuit in the supply to the EL84 circuit and delete the upper (10uF) Capacitor on the screengrid.
The original Circuit used a fairly smart High Tension Noise cancellation principle, which with the originally desiged supply (when set up correctly) manages << 0.1mV unweighted RMS output noise (or around -90db referenced to 1W Output or -99db referenced to nominal full power of 8W).
The EL84 circuit by itself has a gain of nearly 100, can swing well over 200V peak-peak at low distortion and has a reasonably low output impedance (around 10K).
Sayonara
This is a really good possibility.
I could use two ECL84 (I have them) to drive my four KT88 (2 paralleled KT88 for each ECL84).
I read something about automatic bias: is it a good idea?
I mean, with automatic bias I could be able to guarantee that the current flowing in the two tubes is identical (I'm thinking at something about a differential amplifier...).
Imagine a circuit that will balance the current of the two tubes (through two cathode resistors), while referencing the bias voltage to a fixed voltage reference.
In this way the two currents are the same, and the two bias voltage will be compensated near a defined voltage (let's say -40V).
Ciao,
Giovanni
I could use two ECL84 (I have them) to drive my four KT88 (2 paralleled KT88 for each ECL84).
I read something about automatic bias: is it a good idea?
I mean, with automatic bias I could be able to guarantee that the current flowing in the two tubes is identical (I'm thinking at something about a differential amplifier...).
Imagine a circuit that will balance the current of the two tubes (through two cathode resistors), while referencing the bias voltage to a fixed voltage reference.
In this way the two currents are the same, and the two bias voltage will be compensated near a defined voltage (let's say -40V).
Ciao,
Giovanni
Ciao,
Another idea for my paralleled KT88:
what about to use a modified PowerDrive that will self-balance the two current?
This is the schematic I have in mind to realise.
There are two different PowerDrive: the one on the left is a "standard" one, the second one on the right, driving the second tube, is dynamically controlled by the op-amp that compares the two tube currents and keep them equals.
Could it work?
Ciao,
Giovanni
Another idea for my paralleled KT88:
what about to use a modified PowerDrive that will self-balance the two current?
This is the schematic I have in mind to realise.
There are two different PowerDrive: the one on the left is a "standard" one, the second one on the right, driving the second tube, is dynamically controlled by the op-amp that compares the two tube currents and keep them equals.
Could it work?
Ciao,
Giovanni
I don't see any reason why it wouldn't work. If you try it, let us know how it works. The audio signal would be treated as a common mode signal, and ignored by the opamp.
You are always giving me new ideas to try out.
Since the audio signal is common mode, the time constant could be shortened so the opamp equalizes the two tubes through the entire audio range. If that works then it may be possible to equalize two dissimilar tubes. Maybe a large high power tube like a KT-88 could be used to augment a smaller good sounding tube like a 45. I'll put it on my list of things to try.
You are always giving me new ideas to try out.
Since the audio signal is common mode, the time constant could be shortened so the opamp equalizes the two tubes through the entire audio range. If that works then it may be possible to equalize two dissimilar tubes. Maybe a large high power tube like a KT-88 could be used to augment a smaller good sounding tube like a 45. I'll put it on my list of things to try.
What is the benefit of having tube/ss hybrid full bridge rectifier which is found in the schematic of the link above?
For your reference, it is found in here: http://www.pmillett.com/images/SRPP_EL34.PDF
I am finishing up my Simple SE amp from Tubelab, and the power supply stage can be easily converted to the hybrid full bridge.
Doug
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.