0
\$\begingroup\$

I have an issue with an LTSpice simulation and I'm not sure if the issue lies in the simulation setup or somewhere more basic.

The simulation is as follows: fullsim

The system is as follows: on the left, there is a Class E amplifier. In the center is an inductive link which I've used the T-model for (so instead of a transformer, I have that LM). On the right is a full wave rectifier, voltage doubler, and "load". The part that I'm interested in is middle-top: a buffer and envelope detector.

The issue is this: the signal entering and leaving that first order active LPF are identical. I don't know how this can be the case, as the carrier is 1 Mhz, the modulation is 50 kHz, and the LPF rolloff is ~110 kHz. See image below: LPF

Can someone see how I've misconfigured the envelope detector? Thanks!

\$\endgroup\$
1
  • \$\begingroup\$ what is UGBW of the opamp? \$\endgroup\$ – analogsystemsrf Aug 7 '19 at 5:12
2
\$\begingroup\$

Here's an envelope detector:-

enter image description here

What you have is a diode on the input of an op amp which has bias current of 10fA (0.00000000000001A). At this current the diode is practically a short circuit.

To work properly the envelope detector must have a load. The capacitor needs to be much larger than the diode's parasitic capacitance. The resistor needs to draw much more current than the diode's leakage current. It also has to be low enough to discharge the capacitor during the troughs of the modulation, but high enough to follow the peaks of the carrier. These values are a compromise between carrier ripple and modulation distortion.

enter image description here

\$\endgroup\$
3
  • \$\begingroup\$ I've tried a version which doesn't include the buffer or diode, but the active LPF just rails. I avoided the passive envelope detector configuration because of the current draw (lower impedance than my amplifier at my desired RC and so my inductive current dropped). Is there a high impedance version of what I'm trying to do? Active peak detectors with reset switches may also do the job but that seems unnecessarily complex. \$\endgroup\$ – PoGaMi Aug 7 '19 at 14:52
  • \$\begingroup\$ To reduce rf loading you will need a buffer. The LM6081 is not fast enough. The LT1800 worked well for me in simulation, with RB705D Schottky diode, 100pF and 20k. Input impedance was ~80k Ohms. Your modulation filter has low input impedance (68 Ohms), high gain (10x) and gentle roll-off. Consider using a Sallen-Key filter with a buffer before it. \$\endgroup\$ – Bruce Abbott Aug 8 '19 at 4:20
  • \$\begingroup\$ Thanks, I'll try it! \$\endgroup\$ – PoGaMi Aug 12 '19 at 15:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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