Do you agree that this diode footprint is incorrect?

[treez]

Hello,
Do you agree that the recommended footprint on page 4 of the SBRT25U80SLP diode datasheet is incorrect?
The long sides of the SBRT25U80 diode have metal protrusions sticking out of them, which emenate from the large main pad of the SBRT25U80 diode.
However, the recommended footprint does not offer a metal pad for these bits of protrusion. Worse still, if one were to use the recommended footprint shown, then there would be solder resist under those “lateral protrusions”. This solder resist would partially melt during surface mount flow soldering, and would “stick” to the lateral protrusions, and thus prevent the component from “settling” onto its pads….thus meaning improper soldering.

Do you agree, the shown recommended footprint needs to be extended to take into account the lateral protrusions to the main pad of the SBRT25U80SLP diode?


Datasheet : SBRT25U80SLP diode
**broken link removed**

 

[FvM]

I fear you are seriously over-engineering.

Designers are using this footprint since many years, nobody seem to experience a problem.

Solder resist doesn't melt at legal solder temperatures, by the way.

 

[treez]

Thanks , I don't think it will melt, but will not be as smooth a surface for the settling movement of the component as would a solder resist surface of the FR4.....As you know, the "settling" of a surface mount component onto its pads is the way that these components get properly soldered into position, on pads like these, that are very small.

 

[D.A.(Tony)Stewart]

Diodes Inc datasheet shows the link for the latest footprint which includes pad extensions for tabs. Use the latest, or as shown below.


Always read the fine print. RTfP

 

[treez]

thanks, this is bad news for us though, clearly there is a problem with the footprint in the datasheet which we have copied, or else diodes.com would not have changed it......

 

[marce]

I have used these in the past and never had any problem, I have always used a pad that is rectangular and large enough to cater for all the pads, with a specific solder paste pattern to avoid excess solder and slight tweaking of the resist pattern, the pad though is a plain rectangle. Don't forget manufacturers footprints are a suggestion, you are better off following the IPC-7351 guidelines and if you have any experience of electronic assembly (sadly lacking in many PCB designer these days) then you can make your own decisions on a footprint requirement...if it is not explicitly covered by IPC-7351.

 

[treez]

thanks, so may I please ask Marce for advisory?..., your footprint would be literally one big rectangular pad and then four small rectangular pads for the anode pins...ie, just five pure_rectangular pads?

 

[D.A.(Tony)Stewart]

thanks, this is bad news for us though, clearly there is a problem with the footprint in the datasheet which we have copied, or else diodes.com would not have changed it......

Datasheet also indicates to check for updates with a given weblink, where I found the updated footprint. Now you have more experience in how to read datasheets. YOu should have hired an expert.

 

[marce]

No I have some little squiggly bits (it allows for basic SOIC packages to be placed on the same footprint) , but then I add a copper pour for heatsinking, so the footprint ends up with the pads being solder mask defined. This enlarges the pad area overall as solder mask is globally enlarged (by 0.5-0.1mm) but again this does not affect the solderbility or the position of the component during reflow. As stated you very rarely use these components without some extra copper for heatsinking so take that into account when doing the footprint.....
I also have a lot of experience with production lines over many years from the first days of SMD becoming common place (vapour phase reflow...) and of stencil design having worked closely with the likes of DEC and Tecan amongst others over the years; so I know what will work and what wont, put it down to 30 years hands on experience and LEARNING.