This is a DIY project. The title question is the first problem I'm considering, but the entire project is the goal I'm not an expert in hardware, but neither am I uninitiated. Put me at around an intermediate. I'd like to not break the electronics I'm working with so I'm trying to square away everything I need to do well before I get started, which is why I'm here. I'd like to not give a lot of unnecessary details so I might be withholding relevant details that would help give an answer. Please comment to let me know.

These are the criteria I'll know i need for this portion of the project:

  • I'll need to apply a connection to about 20 different small PCB test points of varying sizes located at different parts of the board

  • The connections need to be stable

  • The connections need to be detachable (no solder)

  • The connections need to be insulated from the rest of the PCB

  • I need to be able to reattach the casing

    The casing has two "layers" when disassembling it. The first layer in has access to the rechargeable battery and a "door" to the "innards".

  • I'd like to place my array of test point extensions so they stay inside the first casing so I can clip them to my microcontroller only when I want to make use of them and so I can use my device normally the rest of the time.

  • 3
    \$\begingroup\$ Generally when I need to do something like that I solder 30 gauge wire wrap wire to the test point and unsolder it when I'm done. If it needs to come on and off a lot I will solder a small 1 or 2 pin header type connector to the test point and use a socket attached to some wire wrap wire to get the signal out. Most of the things I work on are very dense PCBs with very little room for any kind of connector on the test points so I have no choice but to use solder. I'm interested to hear if anyone has a better solution to this. \$\endgroup\$
    – John D
    Commented Aug 25, 2014 at 0:48
  • \$\begingroup\$ Product recommendation requests are off-topic on EE.SE . Please edit your question and do something about the "What could I buy..." in the title. \$\endgroup\$ Commented Aug 25, 2014 at 0:48
  • \$\begingroup\$ What about a secondary PCB with miniature pogo pins bolted onto the primary one? \$\endgroup\$
    – venny
    Commented Aug 25, 2014 at 0:57
  • \$\begingroup\$ @NickAlexeev is "What kind of equipment would I need" better? I'm editing that in. \$\endgroup\$
    – 0xFFF1
    Commented Aug 25, 2014 at 2:44
  • \$\begingroup\$ @venny how little perpendicular-to-board-plane space could pogo pins take up? I'm thinking space will be cramped, bu I haven't opened it up yet, having only seen high-resolution pictures of the insides online. \$\endgroup\$
    – 0xFFF1
    Commented Aug 25, 2014 at 2:48

3 Answers 3


This is one of those things where what you think you want and what is possible to do are about 10,000 meters apart.

It is possible to make a jig that has spring loaded pogo pins. Pins with sharp points can be made to line up with small test points on one side of the PC board. Generally the board is guided in place on the jig via tooling posts that line up in holes in the circuit board. For longer term use a clamping mechanism is needed to hold the board down in place on the spring loaded pins.

The pogo pin fixture is going to take up a good amount of height so give up the idea of it being portable and fittable inside the existing housing. It is also nearly impossible to build a successful pogo pin fixture unless you have the original design data for the circuit board. Gerber files would be a minimum. You also need a precision drill fixture that can drill perfectly straight holes down into the material that holds the pogo pins in place. That material needs to be a good thickness to hold the pins in a sturdy manner. Expect a pogo pin fixture to take up about 2 -> 3 cm of height depending upon which pogo pins you choose.

For trying to fit the connections inside the existing case you are going to have no choice but to solder the wires. Give up trying to think otherwise. Wire the 15-20 wires over to a small connector that you might choose to poke through a hole that you Dremel out in the existing housing.

Keep in mind that once you start down the road of hacking into a pre-existing product that it becomes dedicated to that effort. Erase thoughts of trying to keep your hack platform pristine. If you need the product for normal use then go purchase another one for that purpose.

Just for reference here are a couple of pictures of a small pogo pin fixture that was made to allow programming of an AVR microcontroller via test points to its SPI programming port directly on its product circuit board. In this case the programming operation took only a few seconds so the product circuit board was simply held in place on the tooling posts. Note in the side view that the polycarbonate that holds the pogo pins is ~9.3mm thick.

enter image description here

enter image description here

  • \$\begingroup\$ @oxFFF1 I realize you don't want to use solder, but I agree with this answer and the others advocating it -- if you have existing pads, then soldering small gauge (28-30) wires is the way to go. Once soldered, you can use Kapton tape to hold the wires in place. Later, you should be able to remove the tape and desolder the wires without leaving hardly any evidence they were there. Also, using such fine wires should allow you to reattach the casing. \$\endgroup\$
    – tcrosley
    Commented Aug 25, 2014 at 5:42
  • \$\begingroup\$ This is similar to the test fixtures we make for production testing our own circuit board assemblies; we fork our PCB design files, strip out everything except the mounting holes and the contact points, then add retention clips to hold the board at the correct spacing against the spring-loaded 'pogo pin' contacts during test. But this type of fixture won't fit inside a product case; neither will any mini-grabber test clip I've ever seen. I see no way to avoid soldering wires to the test points. \$\endgroup\$
    – MarkU
    Commented Aug 25, 2014 at 8:43
  • \$\begingroup\$ If you have the skill to make a pogo programmer (or whatever fixture) at all, I don't think not having Gerbers is really a problem (though it helps). You could just measure out pads you want to stick, I'd probably do it on the same mill I'd use to drill the holes later. \$\endgroup\$
    – Nick T
    Commented Aug 25, 2014 at 10:06

If there is 20mm x 6mm clear area near the edge of the PCB with 2.1mm height clearance, you could permanently glue or epoxy a Flat Flex Cable connector to the board. This provides a solid mechanical strain relief for those AWG30 wires you will need to solder to the test points. Secure the wires with Kapton polyimide tape, and now it looks like it was always part of the board... Thread the flex cable through the inner housing so that it comes out the battery door, and close up the case. You might need to file a slot along the lip of the inner case to accommodate the flex cable; won't know until you fit it together. Finally, connect the breakout board to the free end of the flat flex cable and secure to the outside of the case with Kapton tape.

This approach minimizes the risk of soldering damage, because you are only adding wires, never desoldering them. The wires, connector, and tape are consistent with the kind of cable assembly commonly used inside laptops, cellphones, and similar small, multi-PCB handheld devices. So there's a good chance you can fit all this inside the exisiting case. The flat flex cable itself is removable, leaving behind the flat flex cable connector and the 'patch wires'. Note that this type of connector is a bit fragile, it's not meant to withstand heavy use or repeated insertion. Mechanical strain relief is essential.

Flat Flex Cable is different from Ribbon Cable, it's much thinner and stands a better chance of fitting through the case.

One example set of 20-position connector + cable from mouser.com; other distributors also carry similar product and in other sizes.

  • Flat Flex Cable connector: mouser.com 710-68710614022
  • Flat Flex Cable assembly: mouser.com 763-20POSFFC
  • Breakout board: mouser.com 763-NHD-FFC20-1

I'd distinguish between test points and connections for in-circuit programming.

Route the signals you need later on to a separate header, and verify during PCB manufacture that the connection to this header is working.

For your in-house units, populate the header and connect a ribbon cable that you can easily pull out of the case, and for mass manufacture, use a board with pogo pins for the initial programming.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.