1
\$\begingroup\$

Is there such a thing as a "passive" bootstrap?

C11, R21 and R22 in the current source of Douglas Self's Blameless Class-B Amplifier (Fig 33 in http://douglas-self.com/ampins/dipa/dipa.htm) looks a lot like a bootstrapped resistor circuit, but my confusion is that there's no buffer driving the capacitor C11, so no positive feedback. It's also not "just a low pass filter" because if I attach it to ground instead of between R21 and R22 it does essentially nothing. (And there's also already a 220uF capacitor from the positive rail to ground, so we wouldn't expect increasing that to 220uF+47uF would do much more). So how is it working?

Simulation shows me that C11 gives a 10dB attenuation of noise from the positive rail (a 1 volt 60Hz signal added to the positive rail causes only .14% error in the TR1 current with C11, and .42% error with C11 removed). Schematic of Douglas Self's Blameless Amplifier

Here's circuitlab simulation (https://www.circuitlab.com/circuit/pm8wssqz84uh/bootstrap-question/) that shows the impact of C11. Frequency response with and without the 47uF capacitor

\$\endgroup\$

2 Answers 2

Reset to default
3
\$\begingroup\$

The purpose of a bootstrap is to increase impedance by using positive feedback, this is not doing that.

It is basically a lowpass filter, TR1 gets it's base current through R21 and R22, the capacitor and R22 act as a filter.

Sometimes it's hard to make sense of a circuit with PNP transistors because we're more used to seeing NPN circuits. Imagine if TR1 was an NPN, R21 and R22 would be going to positive instead of ground and the capacitor would be going from their junction to ground. So an NPN version would look a little something like this (RL is the CCS load, which in the amp would be the input stage):

schematic

simulate this circuit – Schematic created using CircuitLab

This way it's a little easier to see what's going on. We're used to seeing filters with the capacitor going to ground instead of the positive supply rail, but if you flip the polarity that's what you get.

So if we redraw the PNP version it looks like this:

schematic

simulate this circuit

They work the same way, but it's more intuitive for us to see the filter capacitor grounded.

It does nothing when you attach it to ground instead of the resistor junction because then it's just across the positive supply rail and ground where 47 uf isn't going to make much difference.

\$\endgroup\$
3
  • \$\begingroup\$ Your circuit is actually better than Self's because it protects against noise on both the collector and emitter-side rails, while Self's only filters noise on the emitter-side rail. (His R22 goes to ground not the negative rail, while your R3 goes to Vcc). \$\endgroup\$ Commented Nov 18, 2024 at 4:01
  • \$\begingroup\$ @WanderingLogic No, the whole point is that it's basically the same but the polarity is reversed to more clearly show how the capacitor is a filter. See the update to my answer. \$\endgroup\$ Commented Nov 18, 2024 at 7:02
  • \$\begingroup\$ It doesn't really matter, but it really does make a difference whether R22 is attached to the same node as RL rather than a voltage half way between the two power rails \$\endgroup\$ Commented Nov 18, 2024 at 12:12
0
\$\begingroup\$

It is a low pass filter. Adding 47u to 220u rail makes nothing, but adding 47u to 10k reduce noise much significantly.
Maybe the author didn‘t think about filtering 60Hz but rather high frequency noise.
But even in 60Hz it can be said it makes an improvement.
Increasing the 47u to 470u will filter a 60Hz more obviously.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.