Open source deadbolt lock

I really think designing the thumb turn is the least complicated part. Take a cylinder and cut a notch out of the center at the smallest known width of a thumb turn(similar to switchbots design). Cut it in half, reattach halves with fasteners. If you have a wider thumb turn add spacers. The simpler and less mechanical parts, the better IMO.

With the electronics we’re talking Bluetooth communication, proximity sensors, door open sensors, servos, encryption, power consumption etc.

I just really don’t think the thumb turn and the lock design itself is what’s gonna hold us back.

The more conversation continues the more it feel like a series of locks, not a simple one size fits all.

  1. apartment friendly version
  2. home owner version that fully replaces and is implant only
  3. one that also has a keypad
  4. one that still allows a key
  5. wifi enabled or not enabled on any combination of the above.

hah to me it’s the opposite… the electronics and firmware seem easy compared to an elegant mechanical design… one that is easily manufacturable and takes into account things like flexibility, durability, simple analog motor control, gears, etc. … perhaps it’s good that various people find various aspects of the project easy vs hard… find enough people who think enough parts are easy and it’ll all be easy :slight_smile:

Have you considered using ready to go chunks?

From the last 2 weeks of hyperfixating on trying to find a ready to go lock…

It seems like 70% of smart locks are using the same battery / motor pack on the back of the door…

Perhaps find the actual manufacturer and simply order those, and just make the rfid part?

No pcb wizard, but surely there’s some way to hack s board like this?

For the apartment style locks that you can’t change, use some like the gimdow or switch not unit,
Why reinvent the wheel when you can just order from supplier

I think it was discussed earlier that it was best to stay away from vendor supplied parts because vendors come and go and specific parts may not be available in the future, rendering the whole project useless.

Edit: then again, reading earlier, this exact topic was considered as well. So :man_shrugging:t3:

Just a thought is all…

That was kinda my thought process when I got started, at least for the mechanical design. I already have an August lock I know will work with my lock, it’s just a matter of adding RFID to it… but I agree with everything above; any long term solution will be best derived from scratch using easy to source and resource parts

It’s one of the things that allows humans to do incredible feats. Our ability to share knowledge and skill is remarkable. Now if we could all just get along with each other…

1 Like

I don’t think lock 2 will be necessary if we create a good lock 1. Total replacement will require significantly more engineering to make a solid deadbolt. My Dad runs his own locksmith business, and has my entire life. I’ve broken into a lot different types of locks and doors, and most of the doors could be easily picked or the lock circumvented all together. Whether we use an already made and designed lock cylinder or not it is crazy difficult to design a deadbolt that can’t be easily defeated. Just check out lockpicking lawyers channel for further evidence of ridiculous ways locks have been defeated. If we were to make full deadbolt design and it’s easily defeated with a paper clip, then we have a big issue. I think we should leave the deadbolt design and manufacturing to the pros and just try to make a universal way to operate their thumb turns like @amal suggests.

1 Like

Why not break this into a couple modularized components? Maybe start with an open source circuit that outputs two signal lines, one for lock, one for unlock. Another component would take those signals and feed into a relay to power a motor to rotate the physical interior. And the last component would be an exterior user interfacement, with light indicators and perhaps a solar panel charger. If you break it out on separate boards, other development ideas can take different avenues while still maintaining a sort of set standard on the other parts. We can then make compatibility community “backpacks” for various store bought lock sets, but not have to rely on the store bought components.

1 Like

If we are interfacing over RF then signal lines won’t be needed.

Reviving this thread. Too many posts about incompatible locks.

I’m going to get started with a gripper design(the part that grabs onto the thumb turn) If I make something out of cardboard(I have a few ideas) can someone better than me at 3D modeling recreate it into an .stl so I can print it? If you already have an idea in mind and can 3D model yourself go ahead and do so. So far I can think of these requirements the grabber needs to fulfill:

-ability to adjust size in both x and y axis to fully support as many locks as possible
-quick-enough adjustment available to those axis
-method of attachment to a servo
-has to be able to stay grabbed onto a lock with pretty great force(some doors are fitted terribly and locks need great effort to lock themselves)

Feel free to add any more items to that list, I’d love to hear any thoughts or suggestions. Let’s get this ball rolling again!

Looks like im late to another party. Heres a diy lock for uk upvc doors. We made it over a year ago.




Wiring diagram and Stl is available

8 Likes

9 Likes

Very nice! Can you still use the physical key to open it from the outside?

This is probably beyond my DIY capability, but then again the only commercial solution I found is evva emzy which costs around $1000 :melting_face:

I like the mechanical design of the Eufy retrofit. So I bought one to tinker with.
No clue on whether I’ll be able to integrate my rfid reader to it.
I’m not planning on mounting the keypad on the outside. So if I can get it to work, will do rf through the door.

2 Likes

As it doesn’t seem to be in this thread, I’ll tack this link & summary here.

Key 2.0 (or Key20 for short) is a Bluetooth IoT Door Lock controller.

Summary:
Key 2.0 Video

The main features of Key20 are:

  • Using state-of-the-art security mechanisms (Elliptic Curve Diffie-Hellman Key Exchange (ECDH), HMAC) to protect against attacks.
  • Open-source software and hardware, including an open implementation of the security mechanisms. No security by obscurity! Source code for the app and door lock controller as well as Eagle files (schematic and board layout) are provided.
  • Maker-friendly: using easily available cheap standard components (nRF51822 BLE chip, standard electronic parts), easy to manufacture circuit board, and open-source software and hardware design.
  • Works with BLE-enabled Android 4.3 mobile devices (and of course newer versions). Porting to other mobile operating systems like iOS should be straightforward.[

Initial Project Proposal

](https://engineering.purdue.edu/477grp12/Files/docs/A0%20-%20Initial%20Project%20Proposal.docx)

1 Like

I don’t really like this design because it is very limiting to only US style deadbolts with nothing close by. Integrated locks and non-us mortise locks would not be able to use this.

I really like a retrofit that skillfully manipulates the thumb turn only. That design is compatible with a really large number of locks because almost every lock has some sort of thumb turn that conforms to the typical design of a paddle that you grip and twist.

1 Like

I really like a retrofit that skillfully manipulates the thumb turn only.

amal, I like this one, as the inside piece is not mounted to the door, but to the bolts via the plate. So it requires no screws or sticky sheet stuck to the door.

Thanks for your comment. I wasn’t certain it would fit, but it did quite well.

I’m uncertain if it has much variability or sophistication in the calibration routine, other that right or left side. I replaced the bolt, as it was sticking. I tried just replacing the lock body, but the smart lock wasn’t able to fully open it (the bolt was sticking). I was replacing the keys(lock) anyways, so had all the pieces I needed.
The bolt was sticking bad enough, that the old lock would turn in it’s barrel. So I was not surprised. Cheers.

I’ll add, this is the video that sold me on the idea.

I beleive eufy is a rather dodgy company with regards to privacy. If it was simple hardware then maybe, but using their software in any capacity is pretty risky. If Im remembering right theyre the company that had home security cameras sending live feeds to unencrypted urls that anybody on the internet could access.

3 Likes