4
\$\begingroup\$

I have a question regarding the DESD1LIN2WSQ TVS diode by Diodes Inc. which I want to use to protect some 24V IOs of a microcontroller board.

Until now, the TVS diodes I worked with always had the same characteristic in both directions, but this one doesn't.

What is unclear to me is, how I should communicate the desired polarity to our pcb assembler.

This image from the datasheet clearly shows that polarity matters:

I-V graph

But I can't figure out which pin is pin 1 and which pin is pin 2. Those images from the datasheet don't show this clearly:

device schematic

product marking

I have not seen this device in real, because our assembler does the sourcing. Assuming that the device has a "cathode" marking, I'd place the diode this way by intuition:

schematic

simulate this circuit – Schematic created using CircuitLab

It might be obvious to an experienced engineer, but I don't want to decide "by intuition", which could waste a whole batch of PCBs. So, is the above orientation correct?

Edit 1: I found this document with general information about Diodes' SMD packaging. There is this entry for the relevant diode:

enter image description here

Does this help to determine the correct orientation (e.g. is there a convention like "pin 1 has to be oriented upwards in a reel")?

\$\endgroup\$
13
  • 3
    \$\begingroup\$ Oh wow. Someone dropped the ball on that datasheet. You should message them. \$\endgroup\$
    – DKNguyen
    Commented Sep 1, 2021 at 16:31
  • 1
    \$\begingroup\$ I'd check the tape & reel info from the datasheet but there's none. Pin-1 or pin-2 should've been marked according to the feed direction, at least. One possible option could be that the assembler should check the marking to determine the placement direction. Hope all the components are placed in the same orientation (Sigh). \$\endgroup\$ Commented Sep 1, 2021 at 16:36
  • \$\begingroup\$ Thanks for the posting. I'll watch out if any of my engineers try to use that weird thing. It has Vbr1 = 20.3V, Vbr2=30.3. That's gonna be difficult now and trouble in the future. \$\endgroup\$
    – jay
    Commented Sep 1, 2021 at 16:42
  • 1
    \$\begingroup\$ @SimSon yes, that's the info I'm referring to. This should've been given in the datasheet. Anyway, the feed direction for P&P machine, according to the image, is left to right. So, initially (i.e. at zero degree of rotation) the mark should look towards the left-hand side. Your component has a mark on it, so I'd expect placement acc. to the top detail. But your component is bidirectional, so let's hope they sticked to the top detail anyways. \$\endgroup\$ Commented Sep 1, 2021 at 18:42
  • 1
    \$\begingroup\$ @SimSon So that means that my assumption (pin 1 is "cathode") is correct? looks like so. \$\endgroup\$ Commented Sep 1, 2021 at 19:11

3 Answers 3

4
\$\begingroup\$

Looking at a couple other bidirectional/asymmetrical TVS from Nexperia, Vishay, and ST, the stripe indicates pin 1. I'd test to double check, but it looks like a pretty safe bet that your device is similar.

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Thanks for looking up those diodes! It's becoming more and more likely to me that the way I would assemble the diode is correct. I hope to get a response from Diodes, though. \$\endgroup\$
    – Sim Son
    Commented Sep 1, 2021 at 17:11
4
\$\begingroup\$

I got an answer from Diodes: Pin 1 is marked with the bar.

I also had my assembler do a measurement and the correct orientation for my application (i.e. steady 24VDC are expected accross the diode) is:

  • the side marked with the bar has to be connected to GND
  • the side without marking has to be connected to 24V
\$\endgroup\$
3
\$\begingroup\$

By convention, pin 1 would be marked with the bar. This is supported by the bar being on the left on one drawing, and pin 1 being on the left on the other. It's not a guarantee, though; if you have the chance, you could verify it by applying a voltage (~36V) through a series resistor (~27 ohms) and measuring it each way.

\$\endgroup\$
1
  • \$\begingroup\$ I would normally test it as you said, but the thing is that I don't have such a diode at hand as our assembler did the sourcing... I got in touch with Diodes, hope they will help. \$\endgroup\$
    – Sim Son
    Commented Sep 1, 2021 at 17:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.