IPC-7352 strange evolutions

Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic

   Hello, my team recently purchased the IPC-7352 released in 2023 and I am currently in the process of studying it to decide whether we should make it our new standard for footprints creation or not.

For the moment I must admit that I am not thrilled by what I have read.
Some exemples : 
Page 4 : Figure 3-2 depicts an SOIC instead of a 1206 capacitor. I know errors can happen, but on a document of this stature, it makes me question the review process if there is already that kind of mistake on page 4.
Page 10 we are introduced to the new method to calculate "Rectangular or square end components [...] where leads are 1, 2, 3 or 5 sided", the Toe calculation for such a component with a lead widths equal or larger than 0.5mm gives me abnormal results. It is said to be "25% of the nominal height of the component, or 0.5mm, whichever is less" for B level. If I take a very standard 0402 resistor from Vishay, TNPWe3 serie (width = 0.5 +/-0.05), with a nominal height of 0.35mm, it gives me a toe of 0.0875 (rounded to 0.09). That is less that the 0.15mm toe recommended for C level. How can that be possible ? 
Round off factor for Chip components smaller in widths than 0.5mm is 0.005mm increments. Has this been discussed with a PCB manufacturer ? 5µm variation on a PCB geometry seems quite small... And it will give an absolutely crazy amount of variations for the same footprint depending on the small variations by component manufacturers.
Section 4.4.1 "Nominal Hole Diameter" describes a method for calculating drill hole diameter. It is different than the method used in IPC-2222. Which one should we use ? This method doesn't take the board level into account, and therefore, doesn't take the tolerance of the hole into account. Seems odd. For exemple, for a round terminal on a 1.6mm thick board, the hole should be "Terminal diameter max + 0.15mm". On a level A PCB with 0.2mm tolerance, assuming it is centered, it would leave only 0.05mm more that the terminal diameter max which does not seem enough.

My question : what do you all think about 7352 ?
I would be very interested in your opinion specifically, Tom H, as I know you are very much involved in the IPC talks (thanks for all your work on that by the way). Is it a good upgrade to 7351B ? Honestly I was hoping for more.
But maybe I am a bit to difficult...
Sorry if my English is not perfect, as it is not my primary language.

Author	Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic
sot23 Members Profile Send Private Message Find Members Posts Add to Buddy List Active User Joined: 30 Jun 2022 Status: Offline Points: 17	Post Options Post Reply Quote sot23 Report Post Thanks(1) Quote Reply Topic: IPC-7352 strange evolutions Posted: Yesterday at 8:23am
	Hello, my team recently purchased the IPC-7352 released in 2023 and I am currently in the process of studying it to decide whether we should make it our new standard for footprints creation or not. For the moment I must admit that I am not thrilled by what I have read. Some exemples : Page 4 : Figure 3-2 depicts an SOIC instead of a 1206 capacitor. I know errors can happen, but on a document of this stature, it makes me question the review process if there is already that kind of mistake on page 4. Page 10 we are introduced to the new method to calculate "Rectangular or square end components [...] where leads are 1, 2, 3 or 5 sided", the Toe calculation for such a component with a lead widths equal or larger than 0.5mm gives me abnormal results. It is said to be "25% of the nominal height of the component, or 0.5mm, whichever is less" for B level. If I take a very standard 0402 resistor from Vishay, TNPWe3 serie (width = 0.5 +/-0.05), with a nominal height of 0.35mm, it gives me a toe of 0.0875 (rounded to 0.09). That is less that the 0.15mm toe recommended for C level. How can that be possible ? Round off factor for Chip components smaller in widths than 0.5mm is 0.005mm increments. Has this been discussed with a PCB manufacturer ? 5µm variation on a PCB geometry seems quite small... And it will give an absolutely crazy amount of variations for the same footprint depending on the small variations by component manufacturers. Section 4.4.1 "Nominal Hole Diameter" describes a method for calculating drill hole diameter. It is different than the method used in IPC-2222. Which one should we use ? This method doesn't take the board level into account, and therefore, doesn't take the tolerance of the hole into account. Seems odd. For exemple, for a round terminal on a 1.6mm thick board, the hole should be "Terminal diameter max + 0.15mm". On a level A PCB with 0.2mm tolerance, assuming it is centered, it would leave only 0.05mm more that the terminal diameter max which does not seem enough. My question : what do you all think about 7352 ? I would be very interested in your opinion specifically, Tom H, as I know you are very much involved in the IPC talks (thanks for all your work on that by the way). Is it a good upgrade to 7351B ? Honestly I was hoping for more. But maybe I am a bit to difficult... Sorry if my English is not perfect, as it is not my primary language.



Tom H Members Profile Send Private Message Find Members Posts Add to Buddy List Admin Group Joined: 05 Jan 2012 Location: San Diego, CA Status: Offline Points: 5967	Post Options Post Reply Quote Tom H Report Post Thanks(0) Quote Reply Posted: Yesterday at 8:56am
	IPC-7351B and IPC-7352 are identical for Surface Mount. No change except the pad stack naming convention added a double 'rr' for Rounded Rectangle pad shape. IPC-7352 introduced Through-hole technology, but most of the information was extracted from IPC-2221 & IPC-2222. The main thing that was added was the Through-hole land pattern naming convention which we created in 2008 but shelved until 2023. The IPC-735x series misses the mark in several areas. - Solder joint goals 'one size fits all' doesn't produce the best assembly attachment and it doesn't adhere to IPC J-STD-001. Also, the values between density levels is too robust. Most is too Most and Least is too Least. - The naming convention puts the 'pin qty' at the end of the footprint name. This was changed in the IPC-7351C standard that was unanimously approved by the land pattern committee but never got released. - The Zero Component Rotation differs from the standard they replaced - IPC-SM-782 Related posts: https://www.pcblibraries.com/forum/ipc7352-vs-pcb-libraries-footprint-naming-option_topic3488_post13869.html?KW=IPC%2D7352#13869 https://www.pcblibraries.com/forum/pcb-pad-footprint-orientation_topic3460_post14010.html?KW=IPC%2D7351B#14010
	Stay connected - follow us! X - LinkedIn

sot23 Members Profile Send Private Message Find Members Posts Add to Buddy List Active User Joined: 30 Jun 2022 Status: Offline Points: 17	Post Options Post Reply Quote sot23 Report Post Thanks(0) Quote Reply Posted: 4 hours 15 minutes ago at 9:05am
	Thank your for the answer. "IPC-7351B and IPC-7352 are identical for Surface Mount. No change except the pad stack naming convention added a double 'rr' for Rounded Rectangle pad shape." That is not what I see when I read both documents side by side : Table 3-3 (page 10) of the 7352 specify a Toe calculation for Square ends components with W=<0.5mm that, on the Median footprint, is dependent of the height of the component (which I think totally makes sense when comparing to J-STD-001). This is not the case for the 7351 (table 3-5, page 17). As this height dependency is only for the N footprint, it leads to cases where the N pads are smaller than the L pads, which seems strange. "IPC-7352 introduced Through-hole technology, but most of the information was extracted from IPC-2221 & IPC-2222." The Through hole calculation (4.4.1, table 4-1 and 4-2) is in direct contradiction to the calculation in IPC 2222 (Table 9-5). Or I am having big trouble understanding theses tables. Theses are mostly the points that confuses me. Thank you for the linked posts. It is very interesting to know the history behind these standards.

Tom H Members Profile Send Private Message Find Members Posts Add to Buddy List Admin Group Joined: 05 Jan 2012 Location: San Diego, CA Status: Offline Points: 5967	Post Options Post Reply Quote Tom H Report Post Thanks(0) Quote Reply Posted: 3 hours 4 minutes ago at 10:16am
	The unreleased IPC-7351C had new solder joint goal tables for Gull Wing and Rectangular or Square End Cap packages. The Square End Cap solder joint goals need to have unique Toe values for every chip size. The Gullwing terminal lead needs a different toe goal for every pin pitch. SOP/QFP Table:
	Stay connected - follow us! X - LinkedIn