If it’s not immediately obvious, that issue is that the socket should be aligned with the U1 pads. I’m at a loss for how to make this happen. I tried resetting the anchor point for the (custom) U1 footprint to be at the origin.
It occurs to me that pin 1 of the part appears to be at the anchor point of the footprint. Should I be setting the anchor point of the foot print at the center of Pin 1 of the foot print? Is there a way I could’ve known this without the 3D view? Or is there a way to change the “anchor point” of the part without changing the anchor point of the footprint? (because that was a bunch of effort to reconnect all the tracks.)
No… I need a 17.78mm (i.e. 7 holes wide on a standard board) wide 24 pos (2x12) socket, so I modified a copy of the 15.24mm one in the footprint editor.
Looking at the Mouser site, this combination of things appears to be roughly “unobtainium” so I guess I am both on my own to make the 3D model and the eventual part itself. 3D printer, here I come! I guess…
Maybe I should give up on this part. It would certainly make things a lot easier in terms of routing.
The situation is like this: The part is a Adafruit-style breakout board. I wanted to do it this way so I could interpose the part and rewire anything I got wrong. Assuming everything worked right, I also have SMT pads for the “real” chip. But if I made a mistake, I could temporarily fix it, and reorder the board, so on this first iteration, I just added both. Given that I now have to use FreeCad or Fusion or whatever to make a custom model, and knowing that the PCB assembly people aren’t likely to have a socket like this anyway, maybe it’s just not worth it…
I come from (long ago) the wire wrap world (which sucks, and I never want to do again, let alone pay for a wire-wrap gun, I figured I’d try to go this way, but I’m thinking now that I should either go for 2x SIP sockets or just screw the breakout board all together.
Oh and… for future people who may see this, yes, if I move the anchor point to the center of pin one, that solves 15% of the problem, and explains why the part in the 3D model was so far off.
In the Footprint Editor, select the part, and then File/Footprint-Properties, then click on 3D-Model tab. Look at lower left for offsets you can tweak. Rotate the 3d model so you can see the pins on the bottom to fine-tune.
Worth consideration (from someone with decades of connector experience and patents…etc):
• Engineers layout designs as (we) want and CAD and other Design software have a mix of location-origins.
• Thus, Models and Footprints downloaded can (and, do) have a mix of positions/orientations
• Kicad’s Footprint Editor facilitates User ability to repostion as desired
That said, and though nothing is wrong with using a DIP-24 15.25, just be aware that the Models in Kicad for this have Round Pin’s. This is not a typical Chip Socket Header; it may be what you want but, 99% of Chips won’t fit into it as most have Flat/Blade Pins.
See Screenshot below.
However, the Pad’s can be the same size and facilitate using either but User must be aware of that…
Works for me. I have these precision sockets and I put regular chips in them. In their previous life before I desoldered them with a heat gun they served as sockets for regular TTL and VLSI chips.
I have the depicted wiper variety too but the precision ones are higher quality and are priced higher.
Yes, I also have several that fit with the Flat pins but, Not All round sockets will fit without some crushing of the Flat pin… I was simply making User aware and to check…
(as a former Director of Engineering for a US plant of JAE, I’ve seen far too many ‘Oops!’ errors in design and production… just a caution…)
And, being a Conector designer, I am keenly aware of Contact-to-Contact interface.
If you put a Rectangular Pin into a Round Hole (Tapered, or not), only the Corner Edges make contact! And, if Tapered, it’s only worse!!
Here’s a typical listing. They are not exotic at all. They are not made for exotic ICs with round pins. They were used by commercial boards that I recovered components from. If the companies were doing the wrong thing, then they have a lot of company.
Well all I can say is machined pin sockets have been used for a long time in industry and there’s no indication anywhere that they suffer from connection problems significantly more or less than dual wipe pin sockets. You idea of contact is based on some mental visualisation of how contact is made. What actually happens with some metal deformation, oxidation on pins, and whether it actually matters for the contact may not match your visualisation.
It’s your mental concept against what has been in use for a long time.
Here’s a typical poll albeit a small one. It is more or less evenly split:
terryb’s answer is probably the most comprehensive of the ones given.
A similar but slightly different problem I hit when looking for 64-pin DIP sockets is that they are usually 1.77 mm pitch and the 2.54 mm pitch ones are rarer and expensive. A suggested solution is to make two rows of SIP sockets.
In the 80s and 90s in the UK, I was not allowed to use the dual wipe sockets in Marconi and Racal, as they were alleged to suffer from a creep out problem. We always used the round “turned pin” sockets, especially in the wire wrap era.
Also back then TI used a lead frame that went black very quickly, but the contact of the turned pin was considered gas tight and remained reliable. The flat pin sockets did not have the contact pressure and got intermittent.
Yes I have exactly the same story from the same time period, we were shown research and stuff that showed the difference and medical, aerospace industries weren’t allowed by law to use the ‘flat wipe’ sockets it was always ‘Turned Pin’ as the edges of the IC would ‘cut’ into the socket a little and make a proper metal to metal join free from any trapped oxides and general crap. I have carried this with me all my life and always used Turned Pin and probably always will.
