8
\$\begingroup\$

This comes from an LT application note, but I found the topology in other circuits too.

The rest of the schematic isn't really relevant, this is the section I'm interested in:

interesting part

It seems to me some kind of bidirectional clamp. The resistors connecting to the ±15 V supply bias the Zener diode to a good current point and the bridge steers the output voltage from the circuit before.

I guess that this would clamp the output to 2.5 V + 2 diodes drops, am I right? Why the whole bridge instead of two Zener diodes?

Could it be that being the zener biased the operating point is stable so it wouldn't affect the signal (due to leakage or knee) before the clamp?

\$\endgroup\$
2
  • 3
    \$\begingroup\$ "The rest of the schematic isn't really relevant,": as always, it is :) No subcircuit without purpose (unless something is badly engineered). "I guess that this would clamp the output to 2.5V + 2 diodes drops, am I right?", yes, as explicitly described in the application note you're citing, this clamps to 3.7 V. The thing is explained, in context! \$\endgroup\$ Commented Nov 9 at 12:27
  • 5
    \$\begingroup\$ "Why the bridge instead of two zeners" is the essence of the Q here. Worth answering imo \$\endgroup\$ Commented Nov 9 at 12:31

1 Answer 1

Reset to default
9
\$\begingroup\$

It seems to me some kind of bidirectional clamp.

Yes, it is.

The nice thing about it is that the clamp voltage is perfectly symmetrical. Had they used two Zener diodes in anti-series, the clamp would be slightly asymmetrical because the two Zener diodes would not be perfectly identical.

(There is still a slight variation due to the plain diodes' drop-out voltage. But that drop-out voltage is significantly less than the clamp voltage, so the small difference in the diodes' on voltage is even smaller.)

Additionally, the LT1009 is an voltage reference IC, not a Zener diode. It's more expensive than a Zener diode. The bridge saves money because you only need a single IC.

\$\endgroup\$
2
  • 4
    \$\begingroup\$ Thanks for the clarification regarding the rectifier diodes! Yeah, the LT1009 symbol is misleading (it actually hides the ADJ terminal altogether, and I'm not sure why the LT1021-10 gets a generic "block" IC symbol in the very same schematic, whereas the LT1009, as much a precision reference as that, gets a Zener diode symbol without an indication that it's in fact not a Zener diode at all; then again, the TL431 are typically also depicted as Zener diodes, but at least not hiding the third terminal.) \$\endgroup\$ Commented Nov 9 at 14:06
  • \$\begingroup\$ AN 42 in the same series has a whole series on references... didn't actually really notice it was reference! The LT1021 probably has a box since it's a series reference and not a shunt. Looking at the circuit the symmetry is needed since it's used for integrating a bidirectional ramp \$\endgroup\$ Commented Nov 10 at 7:00

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.