Glass Passivated rectifiers

[mediatechnology]

+15-20V to the cathode and a series resistor of 100K on the anode to ground. 1 nA would be equal to 100 uV.

Randy - to test those you could just reverse bias them. Use a series R to ground and a DC voltmeter to read the leakage current indirectly as a voltage across the sense resistor. +20V (or more) to the cathode, DUT ("diode" under test), anode to 100K. Connect the supply and DVM common to the other end of the 100K.

You'll need a good DVM. 100K will provide 100 uV for every 1 nA of Il. The series R could also be 1 meg to give you 1 mV/nA but you'll have some significant measurement error due to the DVM's input impedance.

The bias current of the 1510 1512 is quite large: 14 uA max. A diode with 100 nA of Il isn't going to be a deal-breaker when compared to 14 uA. But, very high Il will cause offset when it becomes a significant percentage of I bias. My hunch is that very high values of Il will also increase noise.

[mediatechnology]

Thanks Randy! I appreciate the drawing and your data.

If you look at the Vishay datasheet for the GP there's conflicting info. They're good diodes though.

http://www.vishay.com/docs/88504/1n4001gp.pdf

They show 100 nA max in the bullets then 5 uA at 25 degC. The ones I tested were excellent at around 1 nA. The worst one was 20 nA. I get mine from Allied electronics since Mouser do not stock the GP Vishay.

http://www.alliedelec.com/Search/Produc ... t=950-0207

I should probably test some DB-104 DIP (4 pin) bridges.

[Bruno2000]

Just bought 100 of these from Allied and tested them all. They were all between 2-3 nA. Nice!
Best,
Bruno2000

[mediatechnology]

Thanks for checking these. They're a little hard to find being that some distributors stock glass passivated rectifiers but mistakenly think that the "GP" suffix means general purpose. Digi-Key does that. So you don't really know what you're getting.

The Vishay from Allied are the real deal. Fairchild makes a glass-passivated version - you can find it at many Fairchild distributors - but the datasheet on the Fairchild website is very hard to find. I ended up getting a Fairchild datasheet from a datasheet "mill."