0
$\begingroup$

I'm trying to open a door. More precisely, I'm trying to design a mechanism to open and close a door mechanically which also allows me to open the door manually.

Now, I thought this would be fairly straightforward - just a simple servo, a rod, and an arduino then I'd be good to go. Easy Peasy.

But man, the more I try to think this I find I have to keep throwing out designs.

For instance, the problem with the basic servo idea is that I'd also like to be able to open and close the door on my own, without use of the mechanism. And, from my (admittedly very limited) experience with servos, I think this poses a problem. All the places I've read say that you shouldn't rotate a servo's arm manually - it can damage the gearing inside. If I'm trying to have a long-term installed mechanism, then the odds are pretty darn good I'll end up breaking things. (Plus, if it's just a direct attaching of the servo arm to the door then there's also be resistance from the arm against me pushing the door open).

So now I'm trying to find ways to design this thing. But my mechanical knowledge is every bit as limited as my servo knowledge... so I'm asking for help.

I have a few questions that hopefully the SE folks can answer:

  • Does anyone know of a mechanism design that would allow for the door to be openable both mechanically and manually? I have found designs like this, but am not sure I'd have the ability to build something with that many moving parts. Is there any way to build this more simply?
  • Are there any resources y'all could point me to to learn about designs for things like this? I've gone looking for a database or library of mechanisms, but don't know where to start.
  • Dumb question: does anyone know of a servo model where the arm can freely be rotated manually? That'd.. actually be the easiest option.
  • Anything obvious I'm missing in a plan like this?

I've looked at linear activators, too, but from everything I've seen they have the same problem as the servos. A simple solenoid could push the door open, but I'm not sure there's a way to make sure it can close it.

Note that this is a door that opens inward on a hinge (so your standard bedroom door).

Thanks for your input, and happy to answer any questions

Improve this question
$\endgroup$
5
  • $\begingroup$ Locks have been designed for 2000 years. Must be one that suits… $\endgroup$
    – Solar Mike
    Commented Aug 8, 2023 at 15:43
  • 1
    $\begingroup$ have you considered going and looking at doors that have the handicap button to open them? These are ubiquitous, I can't imagine designing one myself. $\endgroup$
    – Tiger Guy
    Commented Aug 8, 2023 at 16:33
  • $\begingroup$ Your requirement specifications are incomplete. Can the door spring closed? 1. That way the automated pusher mechanisism can open and hold it open and let it close when retracted. 2. Meanwhile a manual push would open against the spring and the door would close when released. 3. It might not be possible to close the door manually if the automatic pusher is extended. Hit the Edit link below your question ... $\endgroup$
    – Transistor
    Commented Aug 8, 2023 at 18:51
  • $\begingroup$ Have you considered a drive gear or clutch that disengages the servo motor for "manual push mode"? This can be triggered by a capacitive sensor in the doorknob. I've seen security doors operate similarly, where they allow exit but not entry. $\endgroup$
    – RC_23
    Commented Aug 8, 2023 at 19:53
  • 1
    $\begingroup$ Another option is to couple the servo drive rod to the door in a way that is one sided (can push but not pull). E.g. the motor pushes the door as with a piston head, but when the user pulls the door retreats from the piston head $\endgroup$
    – RC_23
    Commented Aug 8, 2023 at 19:58

3 Answers 3

Reset to default
1
$\begingroup$

The electric mechanism is engaged by a push of a button. Once it has completed its activity have it disconnect physically. Every time the button is pushed, the first thing it does is mechanically engage, operate and then mechanically disengage.

Any manual operation in between would still operate normally

Improve this answer
$\endgroup$
4
  • $\begingroup$ This is a ok solution. But mind you were i work we have doors like this and the motors are broken alot because people still interact with the door when its moving. $\endgroup$
    – joojaa
    Commented Jul 20 at 20:30
  • $\begingroup$ @joojaa You could add an additional slipper clutch, or if you're clever perhaps get one of the existing components to operate as such. $\endgroup$
    – Drew ♦
    Commented 2 days ago
  • $\begingroup$ @Drew I dont think that works because both of the actions move in the same direction part of the time. So the clutch does shield from opposite movements. So it does eliminate 50% of the problem cases. $\endgroup$
    – joojaa
    Commented 2 days ago
  • $\begingroup$ @joojaa Sorry wrong term. I meant a clutch that slips when the torque is excessive. Torque limiting clutch? $\endgroup$
    – Drew ♦
    Commented 2 days ago
0
$\begingroup$

Consider a Yale lock night latch.

It has the cylinder outside the door - that's the part the key fits in.

It has the nightlatch inside the door - that's the part with the knob you can turn with your fingers.

Either turning the key or turning the knob opens the door.

I think you're asking for a mechanism like that. A servo turns "the key" or the person turns the knob. Either opens the door.

I'd suggest you take nightlatch apart to see how it works. I can't find a simple web-page that explains it but around 4:03 of this video you can see the mechanism

https://www.youtube.com/watch?v=-IUwEtRak_s

There are arms on the rotating disk that pulls the tongue back. There can be two disks with arms: one for the knob and one for the key. Turning either retracts the tongue.

Improve this answer
$\endgroup$
1
  • 1
    $\begingroup$ the OP is asking about opening a door, not about unlocking a door $\endgroup$
    – jsotola
    Commented Aug 8, 2023 at 23:28
0
$\begingroup$

Install a lever handle. (Consider an ADA handle, specifically one that has a long piece sticking back toward the door at the end.) Drill a hole in the end of the handle. Stick a cable through the hole. Attach the cable to some mechanism that can yank on it.

The mechanism will pull down the handle and open the door. When turning the handle manually, the cable will go slack. This has the disadvantage of making the handle only usable in one direction.

If you can figure out a way to install a servo arm that rotates around the door handle, have a cam arm that sticks straight up and rotates 180° and engages with something after 90° of rotation. It'll act as a stop if you turn the knob 'up' but won't interfere with using the manual handle.

If you're able to install something inside the door itself, some mortise latches have a bit on the end that let you pull the latch to open it aside from using the door handle.

The general name of the game is to use something that only engages the mechanism in one direction but doesn't interfere with the mechanism being moved independently.

Improve this answer
$\endgroup$

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.