# QFN 5mm x 5mm Footprint Dimensions?

I've been looking at this part for an hour, and I can't for the life of me figure from the datasheet two relationships. I have relationship A? and relationship B?, which I am struggling with. I don't have the part in hand (it's an obscure Asian IC... of course...).

I'm not a master at building footprints, but I have laid plenty of QFNs and usually a datasheet provides a few more of the center to center relationships.

I feel like Asian ICs are the worst offenders with these kinds of under-detailed documents.

This one is crazy because it is not symmetrical. Is it possible to figure out A? and B?

I feel like there must be some assumptions about how the pads are centered to each other or something else here... I don't know want to use an assumption to layout a board though.

1)A?

What am I missing in this relationship of distance of the two pads in A? -- ?

You are provided with K, which I take as the edge of the relief pad to the edge of the pad. We know the pad size (.4mm by .120mm).

So how could you determine the distance from the row of 14 pads on the top, to top of the first pad in the 10 pad row (I'm calling that distance A?).

2)B?

For B?, is the assumption that the pads are symmetrical to the 5mm package length? So, you could take the 14 pads + pitch, an and infer there is an equal offset of space on each end?

• A is going to be ((E-(e*9+b1))-2L)/2 or 0.465mm B is going to be (D-(e*13+b1))/2 or 0.165mm
– Axis
Oct 7, 2017 at 22:26
• I think a very good guess would be that the pads are centered around the centerline of the part, with spacing 'e'. Oct 7, 2017 at 22:35
• 14 pads on the north and south side, 10 pads on the east and west side. Oct 7, 2017 at 22:35
• This makes your A and B dimensions irrelevant since the part origin should be centered anyway. Oct 7, 2017 at 22:36
• You're right though, this drawing is under-specified and I keep finding that more and more, especially outside the major major manufacturers (and sometimes even they do it too) Oct 7, 2017 at 22:38

The reference for this part is at the corner where planes A B and C intersect. The device package has a tolerance in size of +/-0.05mm.

You can assume the pads are nominally centered on the nominal 5mm dimensions.

If you are trying to figure out the tolerances it gets a bit more complex, but I don't think you care that much about the location of the package relative to the pads. If you were to check the dimensions you would use those planes by putting the part in contact with precision surfaces at those planes.

It looks to me as if the pins are guaranteed to be within 0.1mm of the nominal positions relative to the A B C corner and within 0.05mm of the bottom plane.

That is a pretty fine-pitch part- 0.35mm. Ouch.

There is a drafting principle that redundant dimensions should not be included.

• What tells you that the reference is at ABC? What does that really mean? Yeah it's tiny... I get the redundancy issue, but even this same manufacturer has a newer part with a few centered lines that intersect. End of the day I just want a working part on the first board spin... Oct 8, 2017 at 4:51
• I think i have a good footprint, assuming centers on the 14 pads to 5mm and centers on the 10 pads to relief pad. I guess that is the design intent? Oct 8, 2017 at 4:53
• The reference planes A/B are shown in the top left view and the C plane in the bottom view. They are referred to in the GD&T block for the pads (M = maximum material condition). Oct 8, 2017 at 5:04
• And yes, I think it's missing the two center lines. Oct 8, 2017 at 22:17