Covert REX trigger

Request-to-exit trigger disguised as an emergency door release

Similar projects worth following
An idea to disguise a mechanical "Useless Machine" in a ubiquitous wall box to release a Request-to-Exit button on demand, for physical penetration testing.

I've built a working "v1" of the project, proving the concept, and I'm currently refining a "v2" which will be 3D printable. The BoM below is for "v2".

Here's a bit of "proving the concept" video:

The project consists of a servo, battery, charge circuit, microcontroller, and RF relay. When the RF relay closes, it signals to the microcontroller to start a 3 second timer. If the 3 seconds elapses without the relay opening again (to allow for power tests) the servo does a smooth 180 degree sweep, lifting the lid up and out, with the intention of pressing a green mushroom REX button.

This the current motion and hardware for Version 1:

Here's as "how it would work" video with V1 hardware:

And here's a little bit of how V2 is progressing:


(NOTE: This is WIP and will change regularly) This is the STL file for the latest version of the V2 mechanism, with an early version shown in

Standard Tesselated Geometry - 3.21 MB - 06/16/2024 at 01:00


View all 10 components

  • 9 months

    Robert Wallhead7 hours ago 0 comments

    I just realised I've been working on this project for 9 months, and figured that was worth a short montage of the development!

  • Another iteration

    Robert Wallhead06/16/2024 at 00:40 0 comments

    I've managed to print the base of the latest iteration and it seems to fit the everything quite well!

    I've - of course - got some more tweaks to make, but I think we're really getting close to the final 1 or 2 iterations of "v2" of this project.

  • Integration hell :)

    Robert Wallhead06/09/2024 at 06:17 0 comments

    Here's some updated images of how the 3d-printed mechanism fitted in the box - after a lot of sanding, filing, trimming and pushing... This design is going to work but I've had to make a lot of small tweaks and I'm looking forward to a re-print soon.

    The parts that I tried to integrate today:

    What I'll be printing next time:

  • Base updates

    Robert Wallhead05/25/2024 at 07:20 0 comments

    Here's a couple of pictures of the updated mechanism base, with places for all the components, routing for the wiring and some protections for the battery:

  • Progress on v2, demo of v1

    Robert Wallhead05/19/2024 at 03:41 0 comments

    I've just put together the following demo of v1 in action:

    And v2's mechanism works pretty well on the first print!

  • Finishing V1​

    Robert Wallhead04/27/2024 at 23:21 0 comments

    Version 1 is complete! Even though I identified a number of shortfalls with this first iteration, I wanted to complete the build of this one as a standalone unit. Everything fits, which is great, though I'd like version 2 to have places for each component and something to assist with wire routing. Here's the finished interior:

    And here's a video of the motion:

  • WIP on v2 3D-printable linkage

    Robert Wallhead04/14/2024 at 05:48 0 comments

    I've uploaded a pic to the gallery showing the WIP design for the v2 3D-printable linkage

  • Work to date

    Robert Wallhead04/14/2024 at 01:09 0 comments

    As this has progressed enough to at least be functional, I've decided to publish it. The videos and pictures here are of my "V1" of the project. I only need to install the battery and charging circuit to consider this version complete.

    Testing has shown this can generate 150-170g of pressure at full reach, which is reasonable as the cheap 9g servo I'm using is supposed to generate 1.8kg/cm of torque, and the distance between the servo's pivot and where it was pressing on the scales is about 10cm. The problem is that REX buttons are specified as taking anywhere between 200g and 1.5kg (expressed as approximately 2-15N) of pressure to trigger, though the one's I've observed have all been at the lower end.

    V2 then really has to target 300g/3N to be considered worthwhile, which will require a servo that's more than twice as powerful. I'm considering this one as a candidate: with an estimated 3.2kg/cm at the 5V I'll be working at.

    Additionally, V2 will be built using a 3D-printed 4-bar linkage so that this project is repeatable, and all the files and code will be published at that point.

View all 8 project logs

Enjoy this project?



Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates