The strange case of the BAT60 schottky diode

[dschwartz]

Hi all!!
I'm quite baffled about an issue i ran into.
I have a design that uses a 4049 inverter stage as a precision rectifier with 2 bat60 schottky diodes for octave up effect..

I used an unknown brand bat60 diodes, and worked perfectly. But now i ordered Infiniteon Bat60, and the octave effect was gone.

I measured the Vf (with a multimeter, no voltage applied) of both diodes and the one that works have 0.228V, and the infiniteon is 0.039v (wtf???)

I read the datasheet for the STM electronics and the Infiniteon and the difference is abysmal..the STM has a Vf of 0.28v at 10mA, and the Infiniteon 0.12V

And the reverse leakage current @10V on the STM is 3uA, and almost 1mA for the Infiniteon!!!

What do you think is making the rectifier to fail? The Vf or the leakage?
My feeling is that so much leakage is making the diode to act as a resistor, so the "precision rectifier" is actually working as a crappy inverting opamp instead.

Pd: i will try the STM, but I'm keeping the Infiniteons..with that leakage may be interesting to experiment with them as clippers.

[Rob Strand]

You need to look at the specific type of BAT60.  There's a whole heap with different suffixes.
Eg.

ST            BAT60J    0.28V typ.  @ 10mA
Infineon   BAT60B   0.24V typ.  @ 10mA

but,
Infineon  BAT60A    0.12V typ. @ 10mA

Yes, really annoying and yes there may be more variations.

The suffix letters can really screw you over these days, even more so when one manufacturer doesn't use the same suffix as another.

and in the end do the real parts act like the datasheet ...

[anotherjim]

Probably is the leakage
...but is it really worth using Schottky diodes? I've messed with op-amp full-wave rectifiers and only seen a marginal improvement at supersonic frequencies in the alignment of rectified half-cycles at the zero-crossing.

[Rob Strand]

Probably is the leakage
...but is it really worth using Schottky diodes? I've messed with op-amp full-wave rectifiers and only seen a marginal improvement at supersonic frequencies in the alignment of rectified half-cycles at the zero-crossing.

Typically the Schottky's with the lower forward drop will have more leakage.     So the  BAT60B is likely to be close to the BAT60J.

The rectifier speed is limited by the opamp slew-rate.  The opamp sits at -Vd then the opamp  has to make a fast transition to +V.   I've seen some improvement in speed with Schottky's.   There was an article in Electronics design in the early 90's which allowed the opamp to be biased closer to 0V  - it did make an improvement (off-hand perhaps 2x).   However a mega-fast opamp pushed the operating frequencies way up.   IIRC the fastest config was a fast opamp and a GE diode.