An unclear situation with a Schottky diode.

[iet]

An unclear situation with Schottky diodes 1N5819. There is one removed from the board and a new diode. After measuring Fv, it became clear that there is a fairly significant difference between 0.153 V for the one removed from the board and 0.213 V for the new one. I also noticed that the dimensions of the new one are two times smaller than those removed from the board. Who has dealt with such diodes in their practice??
The measurements were taken with two multimeters and tweezers.

[moffy]

The data sheet is found at: https://www.onsemi.com/download/data-sheet/pdf/1n5817-d.pdf
You should be able to answer your questions from that.

[Doctorandus_P]

Just like all other electronic components Schottky diodes also come in different forms. Differences in handling current, leakage, maximum reverse voltage, forward voltage drop, size, power dissipation (correlated to size) etc.

[iet]

moffy
Thanks
The datasheet gives the maximum value for 1N5817,18,19. Fv-0.32,0.33,0.34v. The real value, as we can see, can differ significantly.

[ArdWar]

The datasheet gives the maximum value for 1N5817,18,19. Fv-0.32,0.33,0.34v. The real value, as we can see, can differ significantly.

It may differ, but assuming equal test current and temperature (and equally working part) it should not be that different. Anyway those 0.3v values are for 0.1A I_F, meters often reads somewhat lower than that simply due to the lower test current.

[iet]

It may differ, but assuming equal test current and temperature (and equally working part) it should not be that different.

This is very interesting in general in relation to Schottky diodes. Three meters give the same result on two diodes.

[iet]

Differences in handling current, leakage, maximum reverse voltage, forward voltage drop, size, power dissipation (correlated to size) etc.

I encountered this problem for the first time. What is the correct way to measure?

[dietert1]

As you can see in the datasheet, forward voltage correlates with maximum reverse voltage spec. Different schottky diodes do exist! For lower forward voltage one could get a 3 A or 5 A 20 V schottky, or two of the smaller diodes in parallel.

Regards, Dieter

[Phil1977]

1N5819 is a very common type that´s available from many sources in different housings - so it´s no big surprise to have some variation.

In case you have doubts about the quality of the new parts you probably have to do some tests under load - measure Vf at 1A and check for allowed dissipation. You can also check the Vr-reverse-breakdown voltage with a voltage source limited to e.g. 0.5mA. It´s a little more effort to check recovery time.

In case these tests pass everything should be okay. It would be no good circuit design if the application requires a 1N5819 that´s better than specified in the datasheet.

In case these tests fail then it´s probably a fake. Nothing is to cheap to be faked 

[iet]

In case you have doubts about the quality of the new parts you probably have to do some tests under load - measure Vf at 1A and check for allowed dissipation. You can also check the Vr-reverse-breakdown voltage with a voltage source limited to e.g. 0.5mA. It´s a little more effort to check recovery time.

I don't have such opportunities.

[Seekonk]

I have some 15A 15SQ045 commonly used as bypass diodes in solar panels.  I was given some higher current diodes of the same flavor. I forget the number and the data sheet had little info.   The forward voltage of the higher current diode at room temp was higher and that surprised me.  Putting both diodes in series with a thermal break I ran them up to temperature with current only about 5A.  At temperature the higher current diode then had the lower forward voltage.  Didn't expect that.

[iet]

Sometimes I wonder why Schottky invented these diodes. Just kidding.

[David Hess]

It may differ, but assuming equal test current and temperature (and equally working part) it should not be that different.

I measured about one third of that difference with a set of 9 1N5819s, but they were likely all from the same order, so 60 millivolts would not surprise me at all between two different batches or two different manufacturers.

[iet]

I measured about one third of that difference with a set of 9 1N5819s, but they were likely all from the same order, so 60 millivolts would not surprise me at all between two different batches or two different manufacturers.

At what current did you make your measurements?

[David Hess]

I measured about one third of that difference with a set of 9 1N5819s, but they were likely all from the same order, so 60 millivolts would not surprise me at all between two different batches or two different manufacturers.

At what current did you make your measurements?

I just used my multimeter diode test function, like the original poster.

[iet]

Understood. There may be a significant discrepancy in the measurement results, depending on the manufacturer and the production batch.

[Phil1977]

That´s especially true for the DMM-test with quite low current.

You can try it out by yourself: If you switch a - e.g. 1kOhm - resistor parallel to the diode then you ostensibly lower the forward voltage of the diode quite significantly. But is this diode in any way better than without the leakage resistor? for sure not!

In other words: What the DMM is showing can vary quite much and it´s not even clear if lower values are good or bad. The DMM is best suitable if you just want to know if the semiconductor is still alive and to test for polarity