7
\$\begingroup\$

enter image description here

Please refer to the schematic extract above which is the positive half of a bipolar voltage regulator in an amplifier. It is a mostly classic design but there is a part I don't understand. The part I'm not sure about is the arrangement of T46, R85 and R92 between the pass element T53's base and the raw DC input. I have never seen this in "basic" regulator circuits.

The general principle of a series-pass voltage regulator like this is that the controlled element (T53 here) varies its impedance with respect to the load like the upper resistor of a potential divider, the load being the lower resistor in the analogy; so the load "sees" a constant voltage.

The error amp is of course the IC3 op-amp, with a voltage regulator as reference, which drives T50 more or less conductive and in turn lowers or raises the voltage on T53's base.

From what I understand T46 is driven in phase with pass element T53, e.g. for example when T53's base voltage lowers by action of the error amp, T53 turns on more and so does T46 because of the R85/R92 voltage divider applying a higher voltage at its base.

This has the effect of reducing T46's collector/emitter voltage, thus reducing the effective emitter/base voltage of the pass transistor.

I can only conclude that this is a form of negative feedback intended to protect pass transistor T53 from being overdriven. Am I correct? And if so, wouldn't a zener diode provide the same protection? With current-limiting resistor R100 this almost looks like a constant-current regulating circuit but that makes no sense in the context of a series-pass voltage regulator, no?

I welcome any references on this particular circuit arrangement you can provide. Does it have a common name? As I said in the beginning of this post I never saw such an appendix to a pass transistor before.

\$\endgroup\$
4
  • \$\begingroup\$ T46 /R85/R92/R100 forms a current limiter. You can use a two Si diodes instead. \$\endgroup\$ Oct 29, 2023 at 22:42
  • \$\begingroup\$ It looks like T46 together with R85 and R92 limits the sum of the voltage drops across R100 and T53 (IL * R100 + VbeT53) to around 0.7V*(1 + R92/R58) = 1.1V. And this leaves around 0.4V to R100. Thus, the maximum load current will be around IL_max =0.4V/4.7R = 85mA \$\endgroup\$
    – G36
    Oct 29, 2023 at 22:42
  • \$\begingroup\$ Btw, this limiter is very clever designed because BE junctions of T46 and T53 thermaly compensates each other if bjts are thermally coupled. \$\endgroup\$ Oct 29, 2023 at 22:49
  • \$\begingroup\$ @Michal The two transistors are not thermally coupled other than T46 being next to T53's small heatsink (but not in contact with it) on the PCB. However I take good note of that possibility as I might replicate this circuit as a DIY build (It is a very sweet sounding british amp from the '90's). If I do I will find a way to couple the transistors. Would a dab of electronics grade silicone joining T46 and the heatsink do some good? It is a tweak I could probably add to the existing amp as well. \$\endgroup\$
    – Joe
    Oct 29, 2023 at 23:43

2 Answers 2

7
\$\begingroup\$

T46, R85, and R92 form a Vbe multiplier, sometimes called an active zener or a rubber diode. This is across R100. As the output current increases, so does the voltage across R100. At some point it exceeds the "zener" voltage, that circuit starts to conduct, pulling the T53 base toward its emitter, limiting the voltage across the T53 base-emitter junction, decreasing the collector current, decreasing the output voltage. It can be thought of as shunting base current around the base-emitter junction; in your case, around the base-emitter junction plus R100 combination.

Overall the rubber diode works with T53 to form a constant-current limiter, but it has many other uses. It is common in higher-power audio power amplifier designs.

https://en.wikipedia.org/wiki/Rubber_diode

Update:

Responding to mrbean's comment.

Both of those links are operationally the same as in this thread. They each use a single Vbe junction as the reference voltage for the limiting action, rather than increasing that voltage with the active zener.

The technique is less complex than a true constant current circuit because itis independent of the voltage control loop. It is difficult to adjust because 100% of the output current goes through the component that sets the threshold. In a lab or bench supply, the shunt is much lower value and goes through a gain circuit before modifying the voltage control loop.

\$\endgroup\$
1
5
\$\begingroup\$

The subcircuit in question is a voltage clamp, or an active zener. The clamp level can be adjusted with the ratio of the resistors as shown in the simulation below. This circuit forms an inexpensive current limiter which is designed for protection of the circuitry.

enter image description here

\$\endgroup\$
1
  • \$\begingroup\$ It was hard to choose one reply as an accepted answer as all replies are excellent. Thanks to all \$\endgroup\$
    – Joe
    Oct 30, 2023 at 0:38

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.