Wondering what you guys think of ESPs bootstrapped anode load as described here... scroll down to figure 5 in the "Good as it gets" section.
Preamps
Basically the circuit is CC-CF with split anode load on the CC. The CF output is feed back to the top of the lower plate load to bootstrap it. As I understand it the purpose is similar to a CCS load. His circuit uses a feedback loop also to the grid of the CC.
He is not much of a tube guy, more of a SS and opamp fancier but I wonder if this approach might not have some merit in some cases. Maybe two stages like this cascaded with a feedback loop back to the first stage cathode. If you combined with a miller cap on the first stage you might be able to minimize any negative effects of the global loop.
thoughts?
Preamps
Basically the circuit is CC-CF with split anode load on the CC. The CF output is feed back to the top of the lower plate load to bootstrap it. As I understand it the purpose is similar to a CCS load. His circuit uses a feedback loop also to the grid of the CC.
He is not much of a tube guy, more of a SS and opamp fancier but I wonder if this approach might not have some merit in some cases. Maybe two stages like this cascaded with a feedback loop back to the first stage cathode. If you combined with a miller cap on the first stage you might be able to minimize any negative effects of the global loop.
thoughts?
Mike,
The bootstrap of the anode load is a technique which has been widely used for the VAS (Voltage Amplification Stage) of SS Power Amplifiers.
Graham Maynard makes a passing reference to having used it in the 12AU7 drivers of his KT88 Ultralinear Amps it in one of his articles in Electrinics World and it was used in (I think) the Powertron Amplifiers where a ECL86 (6GW8) was used with the pentode section as the cathode follower and the triode section anode bootstrapped in this fashion. There was a thread here a few years back where it was discussed. I'll do a search and see what I can find. See edit below.
Cheers,
Ian
When one considers that the source of distortion in a triode amplifier can be basically viewed as the change in rp of the tube with signal and the level of distortion is the Delta rp compared to the anode load it is easy to see why it gives such low distortion. The anode load resistor value is effectively bootstrapped to say 8 to 10 times it's actual value, so the delta rp vs effective load is much lower (less distortion)
It effectively does the same thing as using a current source load. I would put a caveat on that - before I abandonned SS HiFi Amp design in favour of tube amps I built quite a few SS Amps and tried lots of mods. In every case where I tried a current source load in the Voltage Amplification Stage (VAS) I eventually abandonned the current source load in favour of a bootstrapped load. It just sounded so much better. That may or may not be the case for a triode stage, I haven't tried it.
Edit: I think it was in this thread
http://www.diyaudio.com/forums/tubes-valves/58352-wollcott-cross-coupled-circuit.html
Edit #2: I also recall Broskie covered the bootstrapped anode load in one of his blogs in which he concluded it has some warts. Also be carefull, this form of bootstarpping IS positive feedback and it relies on a negative feedback loop being applied around it for stability.
The bootstrap of the anode load is a technique which has been widely used for the VAS (Voltage Amplification Stage) of SS Power Amplifiers.
Graham Maynard makes a passing reference to having used it in the 12AU7 drivers of his KT88 Ultralinear Amps it in one of his articles in Electrinics World and it was used in (I think) the Powertron Amplifiers where a ECL86 (6GW8) was used with the pentode section as the cathode follower and the triode section anode bootstrapped in this fashion. There was a thread here a few years back where it was discussed. I'll do a search and see what I can find. See edit below.
Cheers,
Ian
When one considers that the source of distortion in a triode amplifier can be basically viewed as the change in rp of the tube with signal and the level of distortion is the Delta rp compared to the anode load it is easy to see why it gives such low distortion. The anode load resistor value is effectively bootstrapped to say 8 to 10 times it's actual value, so the delta rp vs effective load is much lower (less distortion)
It effectively does the same thing as using a current source load. I would put a caveat on that - before I abandonned SS HiFi Amp design in favour of tube amps I built quite a few SS Amps and tried lots of mods. In every case where I tried a current source load in the Voltage Amplification Stage (VAS) I eventually abandonned the current source load in favour of a bootstrapped load. It just sounded so much better. That may or may not be the case for a triode stage, I haven't tried it.
Edit: I think it was in this thread
http://www.diyaudio.com/forums/tubes-valves/58352-wollcott-cross-coupled-circuit.html
Edit #2: I also recall Broskie covered the bootstrapped anode load in one of his blogs in which he concluded it has some warts. Also be carefull, this form of bootstarpping IS positive feedback and it relies on a negative feedback loop being applied around it for stability.
Last edited:
Thanks Ian. Very helpful. I got to thinking about this approach when contemplating lowish B+ applications. Will have to play with the idea on my next preamp project.
mike
mike
In terms of the loading on the first stage (effective plate load) it would not be all that different from a mu follower, where it differs is DC wise as the current paths are independent.
As you mention it could be very helpful in applications where the supply voltage available is sub-optimum for use with a mu-follower.
As you mention it could be very helpful in applications where the supply voltage available is sub-optimum for use with a mu-follower.
Yes, it is logically the same as a mu-follower and can be analysed in the same way but with two minor differences:
1. the CF valve can have different (usually higher) quiescent current than the voltage amplifier so potentially better current driving capability,
2. there is a permanent AC load applied by the parallel combination of the upper anode resistor (to HT/B+) and the CF cathode resistor.
Item (1) gives the potential for improvement over a mu-follower, as the designer has an extra degree of freedom. Item (2) works against, but can be made small enough to do little harm.
I'm not sure how it compares with the mu-follower in terms of PSRR - the maths gets horrible!
Bootstrapping is a form of positive feedback, but usually fairly harmless. It was widely used in early transistor amps, the output bootstrapping the VAS. Something which is not always realised is that bootstrapping increases the output impedance of the CF stage, although by how much depends on the design details.
1. the CF valve can have different (usually higher) quiescent current than the voltage amplifier so potentially better current driving capability,
2. there is a permanent AC load applied by the parallel combination of the upper anode resistor (to HT/B+) and the CF cathode resistor.
Item (1) gives the potential for improvement over a mu-follower, as the designer has an extra degree of freedom. Item (2) works against, but can be made small enough to do little harm.
I'm not sure how it compares with the mu-follower in terms of PSRR - the maths gets horrible!
Bootstrapping is a form of positive feedback, but usually fairly harmless. It was widely used in early transistor amps, the output bootstrapping the VAS. Something which is not always realised is that bootstrapping increases the output impedance of the CF stage, although by how much depends on the design details.
I would consider something like the white cathode follower topology instead if you are contemplating driving mosfets due to their high gate capacitance. Choose a fairly high transconductance high current triode like the ECC99 or 5687.
The Aikido line stage design might be worth looking at for inspiration at least for the second stage..
- Status
- Not open for further replies.