emitter follower for synthetic inductor

I've been fiddling with using a single BC550C as an emitter follower in lieu of an op amp for a gyrator (in lieu of an inductor...).

It seems to simulate just fine in LTspice for all the test conditions I can think of throwing at it, but something seems too good to be true here.  The only failing I can see is a loss of ~1dB on the HF side of the notch.

Is there a killer flaw I'm missing here?

I'm too old to use sims... how does it work if you build one?

I have seen simple darlington followers used in active filters.

JR

I like to design first in the sim and build second. Saves me a lot of silly errors.

I figured I’d ask folks with more experience if there was something I was missing.

I added a second transistor to make a Sziklai pair and this fixed the dB loss on the hf side of the notch. Thanks for that idea.

Note that there's also a gyrator circuit which is generally considered to perform a little better. Google "elliott sound gyrator" for a nice page on these circuits including differences between op amp and discrete versions.

Personally I would use a real inductor because noise performance is better (as long as the inductor is not picking up EMI). I have a pair of really clean UREI 527s that are super quiet (95dB SNR).

But if you're trying to make something as small / cheap as possible, then yeah, a discrete circuit will work fine.

dogears said:

Is there a killer flaw I'm missing here?

Click to expand...

This arrangement is commonly used in guitar pedals. The main issue is the limited headroom.
Since they are equivalent to a LC series circuit, a little input voltage can induce larger voltages at the node, more so when the Q is high.
Since you're doing sims, I suggest you do a .tran analysis of the complete circuit and check voltages and currents in presence of real signals.