# Toy "pull tab"-style feature on PCB

I have recently designed a passive-based modular PCB.

When I click the secondary board into the base board, I would like it to break the current circuit at a fixed point, and insert itself in series instead:

I could easily connect the secondary modification board in parallel with the wire connecting the two passives halves, but then I must break the wire connection for it to be in series, so I have a plan.

I remember back to my childhood days when McDonald's toys had little white pull-tabs that you needed to pull out before they became operational.

One way or other, the pull-tab would keep the circuit open and close it when removed. The easy way to do it would be to place the tab between two small coin batteries, with the spring pressure holding it in, but I have seen toys that have the tab on the side away from the batteries. I suspect the circuits inside are using two plates that push against each other on the PCB, and would close the circuit without the tab being inserted between them.

I would like to do the same thing in my design, only with a permanent tab on the underside of the secondary PCB and the contact plates on the base board. When I clip the secondary board into the base board, the insulator tab pushes the contacts apart, allowing the connected modification to be in series with the rest of the circuit.

Are there parts out there already that are designed for this? Maybe something similar? The boards are passive, so non-mechanical options are not viable.

• The problem with contacts is they oxidize unless there is a wetting current or they are gold plated or there is a friction to wipe oxide off the surface. I imagine you also did not specify but you want these to be low cost (how cheap?) and off the shelf Jun 30 '19 at 3:30
• Electrically it is seen as a DPDT switch with 2 inputs and 2 outputs with outputs shorted when not installed and then connected in to out when installed. Or as a SPST-NC switch not installed then Open and yet connected to 2 external contacts so there are 3 mating surfaces that need to be the same metal. Either Nickel plated or Au plated over Ni. Jun 30 '19 at 3:35
• @sunnyskyguy EE75 yes I was going for relatively low cost (1-3 cents). I suppose an inverse push-button could work... I think they are meant to be oxidation-resistanant. Jun 30 '19 at 3:38
• OK Now you have a mechanical problem to make an inverse NO- pushbutton into a NC switch on disconnect. Not an electrical problem media.digikey.com/Photos/CK%20Comp%20Photos/MFG_PTS815_sml.jpg They don't sell NC pushbuttons in this price range $0.14 (6k pc) or media.digikey.com/Photos/CK%20Comp%20Photos/… ($0.07 (5kpc) Jun 30 '19 at 4:28
• NC reed switch on main board ... magnet mounted on the mod board Jun 30 '19 at 5:41

In "phone" connectors (also variously known as "headphone", "aux", or by their size (3.5 mm or 1/4") or contact configuration (TS or TRS)), many jacks have switch contacts that break a circuit when a plug is inserted. The simplest such configuration is one where the tip contact touches a switch contact unless a plug is inserted.

If you wire the switch contact to the sleeve contact of a TS (tip-sleeve, two conductor), then you have a circuit which is broken by inserting a plug, and the two contacts of the plug are the ends of the break point, exactly what you want.

You might choose to, instead of using TS plugs and jacks, use TRS, and carry the signal on the tip and ring. This has two advantages:

• It provides a third conductor for common or shield purposes.
• If there is a definite direction of signal flow, in which the tip is "to mod" and the ring is "from mod", then your usage is now identical to audio insert jacks, right down to the switching.

Picture for context: a 3.5 mm TRS jack with two switch contacts (top left for tip and middle just below the plug for ring).

(Photo taken by me. Licensed CC-BY-SA as per Stack Exchange terms. Please feel free to reuse elsewhere.)

• When a plug is inserted or removed, temporary mis-connections/shorts occur.

If you use a TS style then the worst this can do is short the two contacts on the mod board (this is not a problem if it's unpowered). If you use the TRS "insert" style then this can also connect the mod board between the main board's output side and ground, which also shouldn't be too much of a problem.

But, consider the electrical characteristics of your particular application. It may be easy enough to ensure that the outputs are current-limited and inputs are protected so that arbitrary misconnections don't damage anything — and this is a good idea for any exposed connectors, anyway!

• The male connectors are usually on cable ends only. Board-mount male connectors exist (example I just searched for) but are not common parts with many varieties.

• Depending on the design of the jack, sideways force may cause the sleeve connection to become intermittent. This is more commonly an issue on the 3.5 mm size than the 1/4" size, in my experience.

This could be reduced by choosing the right orientation of the jack versus any other mechanical interface features on the mod board. Or, it might not matter if the mod board has nothing else connected to it and isn't expected to be bumped in use.

If your mod can be a "dongle" on a cable, then this is not a problem and neither is the availability of connectors.

• Also consider the temporary connections as the plug rakes past the jack's spring-loaded tabs. For ground-centered analog audio with that ground already established on both sides, that's not really a problem - I've jammed an outboard compressor into a live channel many times without issue. But if your signals are not ground-centered or at least at ground potential at the moment of movement, or if the common ground is not yet or no longer established at that time, then you've got problems. Jun 30 '19 at 7:00
• @AaronD Very good point, which I should have thought of! I've added a section on that. Jun 30 '19 at 14:39
• @Jasen I already included that link. Jun 30 '19 at 14:39
• I didn't know headphones jacks worked like that! (Although if they didn't, I suppose there would be excess noise on inputs with nothing plugged in and it would be much harder to detect when a device was plugged in). This is still not ideal for my design though. I am looking for something that is only 1-3 mm at it's largest dimension. Jul 1 '19 at 1:42