4
\$\begingroup\$

My reason for wanting to pick a Schottky diode is being lead to believe they have lower \$V_F\$ than conventional silicon or germanium ones. So I was a little taken back to look at offerings at Mouser and digi-key. Even for fairly low forward current ratings (100mA would more than suffice), the \$V_F\$ values are really no better than silicon. In fact much worse. Am I just "looking for love in all the wrong places" here?

I'm thinking now I'm missing some key parameters in my search. I may pull my post until I've been a little more thorough. Thanks though!

Here's a typical part whose forward voltage drop of 1V seems ridiculous: BAT41 datasheet

Improve this question
\$\endgroup\$
3
  • \$\begingroup\$ Can you post a datasheet to which you were looking? \$\endgroup\$ – Golaž Jul 14 '16 at 17:57
  • \$\begingroup\$ I'm thinking now I'm missing some key parameters in my search. I'm may pull my post until I've been a little more thorough. Thanks though! But here's a typical part whose forward VF of 1V seems ridiculous. mouser.com/ProductDetail/STMicroelectronics/BAT41/… \$\endgroup\$ – Randy Jul 14 '16 at 18:02
  • \$\begingroup\$ @Randy Please place this information and your current thoughts/uncertainty in the question text. \$\endgroup\$ – user2943160 Jul 14 '16 at 18:13
6
\$\begingroup\$

If you run a diode near its rated current, you don't get much advantage, but that has little to do with the process involved. Instead, the limit in such cases is the power dissipated by the package, and for the same current level that implies the forward voltage will be about the same. So the Schottky diode will be wimpier (for the same current) and will not perform much better.

Instead, look at what happens if you run at, say, 10% of rated current. Compare this 40V, 1A Schottky with a classic 50V, 1A 1N4001 but with both run at 100 mA. The Schottky will drop about 0.4 volts, the 1N4001 about 0.8 volts.

Improve this answer
\$\endgroup\$
2
  • \$\begingroup\$ Thank you. I somewhat suspected that, because even an ordinary diode can be coaxed in to a little bit lower VF at low currents. But I suppose this is a case where maybe my "hopes" vastly exceeded reality for a part. (happens to me a lot! :-)). I guess I had hoped for something better than germanium, otherwise I could just use that. \$\endgroup\$ – Randy Jul 14 '16 at 18:36
  • 1
    \$\begingroup\$ No, in general, if you use a beefy Schottky and run at low levels you'll get the 0.4/0.5 volts you've heard about. For the same sort of ratings/current a silicon will go 0.7 to 0.8. Just like the example. \$\endgroup\$ – WhatRoughBeast Jul 14 '16 at 19:29
1
\$\begingroup\$

Looking at the manufacturer datasheet for the STmicro BAT41 small-signal Schottky diodes, look at Table 4:

table 4, "Static electrical characteristics" of the STmicro BAT41 diode

The typical forward drop at 1mA (a reasonable small-signal) is 400mV, maximum 450mV, exactly as expected from a Schottky diode. However, the rating given on the Mouser page is the maximum voltage drop at 100% of the rated current. There is no typical voltage drop value given for the full 200 mA forward current. This higher voltage drop may be from both the diode dynamic resistance and the actual packaging resistance.

Improve this answer
\$\endgroup\$
1
  • \$\begingroup\$ The thing is though, it looks like if I really wanted something better then, say, germanium at 100 mA, I guess germanium really is my best bet. This is why I guess I don't "get" all the excitement about schottky's so called "low" VF. \$\endgroup\$ – Randy Jul 14 '16 at 18:38
1
\$\begingroup\$

OK folks... I believe you've helped me to better understand this! So it seems that if I want a really low VF in a schottky, what I need to do is look for a diode with a current rating significantly higher then my needs. So, since I'd like to be good for 100mA, I started looking at 3 amp or better diodes, and I'm finding much better selections. Here, for example, is a spec for a 3A diode, and SR302, at Mouser. Down around 100mA, the graph shows what looks to be about 280mV for VF, and I assume if my actual current is a bit lower, it will perform even better. Thanks again!

http://www.mouser.com/ds/2/395/SR302%20SERIES_I13-523634.pdf

Improve this answer
\$\endgroup\$
1
  • 1
    \$\begingroup\$ There is also typically an advantage in choosing a diode with a lower voltage rating. Check out Figure 2 here. The price you pay is increased reverse leakage. \$\endgroup\$ – Spehro Pefhany Jul 14 '16 at 19:39
0
\$\begingroup\$

Since you are looking for very low forward voltage drops and you happen to compare Schottky and Germanium diodes in a comment, it seems you are looking for some diode able to rectify a low amplitude signal, and not something for power applications.

Other people have already given you good answers regarding why Schottky diodes may be no better than Germanium ones, but this smells like an XY problem to me, so I'll try to give you an answer to what it seems your real problem (rectifying a low level signal).

When you need to rectify a low level signal and its frequency content is not too high and allows the use of an operational amplifier, you could just use a so-called opamp ideal rectifier (a.k.a. precision rectifier).

This circuit exploits an opamp and negative feedback to compensate for the actual rectifier voltage drop. In other words, the plain silicon rectifier is put inside the feedback loop of the opamp and this makes the circuit behave as an almost ideal rectifier (as long as the opamp and the diode are fast enough).

Improve this answer
\$\endgroup\$
1
  • \$\begingroup\$ Thanks, but the actual app is actually pure DC. I didn't want to complicate the post. There is a low power Li-Ion cell tester with 4 low current leds, intended for a single cell. I'm using it in a 3 cell system having a separate battery protection circuit. So my Li-Ion tester simply taps at the first cell, and offers a reasonable approximation of battery health. Unfortunately, when overdischarge blocking happens, an unexpected polarity reversal occurs through the battery tap pos, to the tester. To prevent that i need a diode, but anything more than a few hundred mV makes the test too inacurate \$\endgroup\$ – Randy Jul 14 '16 at 21:04

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.