![]() |
![]() |
IPC-7351 and SMD Pad Shapes |
Post Reply ![]() |
Page 12> |
Author | |
Nick B ![]() Admin Group ![]() Joined: 02 Jan 2012 Status: Offline Points: 1956 |
![]() ![]() ![]() ![]() ![]() Posted: 24 Sep 2024 at 12:36am |
Well said, quality and consistency is what it's all about.
|
|
![]() |
|
![]() |
|
WilliamsimC ![]() New User ![]() Joined: 19 Apr 2024 Status: Offline Points: 14 |
![]() ![]() ![]() ![]() ![]() |
Hi Susan,
In my opinion, PCB Libraries chose to make the IPC-7352 approach an option instead of the default for a few key reasons. One important factor is library consistency. They want to ensure that users are fully aware of the changes and understand the implications of opting into the IPC-7352 approach. By not imposing this change on everyone, they help prevent any confusion or errors that might arise if users aren’t prepared for the differences. Additionally, allowing users to choose the IPC-7352 option gives them flexibility. Some users may have established workflows or preferences that work well for them, and it’s important to respect that. This way, users can transition to the new approach at their own pace, ensuring they’re comfortable with it before making the switch. Overall, it’s about providing choices while maintaining clarity and consistency in the library. Thanks, WilliamsimC |
|
![]() |
|
Tom H ![]() Admin Group ![]() ![]() Joined: 05 Jan 2012 Location: San Diego, CA Status: Offline Points: 5788 |
![]() ![]() ![]() ![]() ![]() |
IPC-7351B adds Fabrication and Assembly Tolerances to calculate the resulting pad stack.
IPC-7352 removes the Fabrication and Assembly Tolerances for pad stack calculations. PCB Libraries (IPC-7351C) is only the Footprint Naming Convention. The typical package dimensions for a 1206 footprint are:
The package dimensions and tolerances calculate the pad stack size and spacing. |
|
![]() |
|
bab27 ![]() New User ![]() Joined: 01 Sep 2024 Status: Offline Points: 13 |
![]() ![]() ![]() ![]() ![]() |
Hello,
What are the main difference between IPC7351B and 7352 for land pattern dimension? Does IPC7351B use the tolerance RMS value like IPC7352 ? It could be interesting to have a standard footprint size 1206 with nominal size (_N) and tolerance and the associated footprint calculation with the IPC7351B, 7352 and PCBlibrairies. Thank you.
|
|
![]() |
|
Tom H ![]() Admin Group ![]() ![]() Joined: 05 Jan 2012 Location: San Diego, CA Status: Offline Points: 5788 |
![]() ![]() ![]() ![]() ![]() |
The Default is IPC-7351C (which was never released primarily due to Dieter Bergman's passing).
The IPC 1-13 Land Pattern Committee worked on 7351C for 6 years and approved the new IPC J-STD-001 solder joint goals and the updated Naming Convention. We had to remove references to IPC-7351C and replace it as PCB Libraries. However, we do support IPC-7351B and IPC-7352 Options. We do not default to 7351B or 7352 because the solder joint goal tables do not compare with the recommendations from IPC J-STD-001 (but IPC-7351C did). However, Footprint Expert does default to the mathematical model for pad stack calculations for pad size and placement. This model takes into consideration the min/max tolerances of the component package terminal leads from the mfr. datasheet. The downside of using the terminal tolerances is that they are sometimes too robust and not realistic. And if you study the mfr. recommended patterns from millions of datasheets, they don't use the min/max terminal tolerances. They use the Nominal package dimensions. So even though IPC went to great lengths to create this mathematical model - uploads/3/IPC-7352_Mathematical_Model.zip Component manufacturers do not use it for their recommended patterns displayed in their datasheets. |
|
![]() |
|
pcb0123 ![]() New User ![]() Joined: 03 Jan 2013 Status: Offline Points: 1 |
![]() ![]() ![]() ![]() ![]() |
Hi Tom,
We are curious, what are the factors that have led PCB Libraries to make the IPC-7352 approach an option, rather than the default? Is library consistency the main driver, where you don't want to impose the change on users (where it might slip past their attention) and you want the user to be fully aware they are opting in to the newer IPC-7352 approach? Other reason(s)? Thanks, Susan
|
|
![]() |
|
Tom H ![]() Admin Group ![]() ![]() Joined: 05 Jan 2012 Location: San Diego, CA Status: Offline Points: 5788 |
![]() ![]() ![]() ![]() ![]() |
Those 2 things were major updates in IPC-7352.
The Fabrication tolerance was originally created in the 1980's to compensate for the etching process. But we all know this was unnecessary because all fabrication shops swell the outer layer features to compensate for their etching process tolerance. Appling a fabrication tolerance on a pad stack calculation and having the fabrication shop swell the outer layers is called a "Double Tolerance". It took IPC 40 years to figure that out. The Assembly tolerance was also created in the 1980's when pick and place machines required a tolerance of 1 mil (0.025). But the pick and place machine accuracy in 2024 is 0.01 mm and it might as well be 0.00 due to the 0201 chip package. Machines used in PCB Assembly and used to manufacture components are much more accurate today than 40 years ago. |
|
![]() |
|
dramos ![]() Advanced User ![]() Joined: 18 Feb 2021 Status: Offline Points: 64 |
![]() ![]() ![]() ![]() ![]() |
Dear Tom,
Many thanks for your comments. Yes, the PCB designers are little painters, sculptors (jejeje) As the new IPC-7352 was downgraded to a guideline and we do not read anything strange about it in internet, we do not have bought it, but, is there a new mathematical model? I thought that was the same that we used in IPC-7351B, but with the recommendation of using for fabrication and placement tolerances 0.00 mm. (this is another topic that we will discuss another day) and we should change the negative values of our solder Joints to 0.00mm as well. Am I in a error? Best regards, david |
|
![]() |
|
Tom H ![]() Admin Group ![]() ![]() Joined: 05 Jan 2012 Location: San Diego, CA Status: Offline Points: 5788 |
![]() ![]() ![]() ![]() ![]() |
The corner radius was set to 25% of the pad width with a limit of 0.25 maximum radius. We also noticed that the new mathematical model for IPC-7352 that turned off the fabrication and assembly tolerance calculated some pads to be slightly smaller (depending on the terminal tolerance). Also, there is an explosion of microminiature packages where the 0.25 corner rounding exposed package terminal leads. The metal component terminals were too close to the pad corners. You go through trial and error and test things out, but the end goal is to be safe and ensure that all terminal leads have full contact with the pad and consistent quality where all rounded corners are the same size. You get the best overall functionality with symmetrical cosmetically. PCB design is an art and the goal is to create beautiful yet functional PCB Layouts, It's also good for paste mask stencils where the aperture matches the pad shape. However, the setting is User Definable to allow you to change the default setting to whatever you want. |
|
![]() |
|
dramos ![]() Advanced User ![]() Joined: 18 Feb 2021 Status: Offline Points: 64 |
![]() ![]() ![]() ![]() ![]() |
Dear Tom,
I have a question around this topic. PCBLibraries started with a Corner Radius size of 25% with a limit of 0.25mm and the new release has a limit of 0.10mm. I see it as an evolution. Besides, It is in the same way that the recommendation of some Manufacturers. My question is, what you have seen to make this evolution? What you have detected on the new components/footprints to modify this value? Which is the main reason for this change? As ever, thanks a lot for your comments. Regards, dramos
|
|
![]() |
Post Reply ![]() |
Page 12> |
Tweet |
Forum Jump | Forum Permissions ![]() You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You cannot vote in polls in this forum |