Intercept pin headers, three way or the header?

Question

Is there some kind of T-splitter 3-way-splitter etc for pin header blocks? Something that passes through the pins but also contacts them and exposes a set of right angle pins too? So you can use a regular female strip on top (just sitting a mm or so higher than usual) but access the pins from the side too?

For a project I'm working on I have a 40 pin header (typical GPIO, SPI etc block on an rPI) and a board that stacks on top. The top board has a female block pre-soldered on and it's not feasible to de-solder it. I need to access some of the pins in the rPi's GPIO block for another purpose but have very little vertical space to work with. I can't add another full height stacking layer and a breakout board or some kind.

I expected to find sets of pin header taps or similar that formed a collar around the headers and projected at 90 degrees. But can't find anything of the sort.

I'd rather not solder a PCB straight onto the base of the headers and break it out that way. Fiddly access, sensitive components and too easy to melt the base of the pin headers.

There is presumably some "everybody knows" answer I don't know the words for . Help?

Answer 1

You can make your own if you just want a few.

Take some Right Angle headers and solder them to the straight header of the other gender.

Tape them onto the workbench to align them as best you can and solder the two ends first. Then solder the rest. You will have pins sticking out the side.

If you want to have longer pins look for wire wrap sockets and headers.

If you need a DIL header with break out you will have to use two SIL RA headers soldered to the DIL header so that the break out pins stick out to either side.

If you just need a couple of signals I would use a header with longer pins and just solder your wires to the pin root. If you have easy access to wire wrap wire and tool you could just connect wire to the long post/pin and use it to make your connections. However remember that wire wrap wire is solid conductor so has a finite number of bends before it breaks. I find that too often the number is N-1 available bends where N bends are required to complete the assembly task.