Creating partial QFP64 footprint?

I’d like to make the QFP64 footprint for designing a programming adapter where only a few pins have to be soldered and that’s why I’d like to make the footprint that only has the pins in the corners and the pins used for programming - to make unsoldering easier.

Is there something I must take into the consideration when creating such a footprint with “missing” pads?

Can I just take the existing QFP64 footprint and simply delete the pads that are not used or do I have to renumber the pins or to somehow mark the pins which don’t have pads? I’d like the pins in the schematics to have the original numbers as before removing the unused pads.

Here is one example of such an adapter containing a partial QFP64 footprint:


Can I just take the existing QFP64 footprint and simply delete the pads that are not used


If you use the standard symbol (for schematics design) with the standard 64 symbol-pins and than assign a footprint with fewer pads (as your custom 1/4-QFP64) you get error-warnings during the “update pcb from schematic” command. But the created netlist should work ok - so you could just ignore these errors.

1 Like

Consider In-Circuit-Programming and you will not have to have adapter and you will not have to unsolder your microcontroller. For AtXmega microcontrollers we use 3 pin adapter to connect to GND, Reset and PDI pin.

Thank you very much for your reply!

Now that you mentioned the standard symbol, what if I try to make the special symbol for the footprint with fewer pads - is it then possible to have the correct pin numbering which in that case would not be continous?

Yes. ----------------

1 Like

Yes but this is not the adapter for programming the device designed by me and the microcontroller can be read only by desoldering it because of a few reasons:

  1. Microcontroller is secured by security bytes and programmer can find them only if it can monitor the current drawn while sending the bytes so the processor must be powered by the programmer with no capacitors or some other circuits interfering.

  2. For the same reason as in #1 clock signal must be generated from the programmer.

  3. To enter monitor mode one of the pins has to be powered by 12 V