I have a latch design problem and need your help - many years in software engineering aren't helping me much here!

Two pieces of 60x30mm RHS become parallel/adjacent (on the 60mm face), with a 5mm gap, via a pivoting motion between two sub-structures. A rotary latch is installed into the bottom piece of RHS and a pin/striker in the top piece engages into the latch. All works well but this is just the initial latching mechanism to allow the user to engage a stronger mechanism - which is what I need help with: designing/choosing a suitable latch.

Firstly, it will need to hold a large force - 2275kg if I take the 7:1 leverage ratio and x2.5 safety factor into account. If this is wildly optimistic, I can use one on either side to halve that requirement. Secondly, it must be able to pull the two pieces of RHS into near perfect alignment as the rotary latch has 2-3mm of free play. There can be no free play at all once the secondary latch mechanism is engaged.

Any ideas? I attach a drawing of my current best idea - a lever laser-cut from 10mm steel plate, with a small bend, rotating on a shoulder bolt which goes through both 30mm faces of the top piece of RHS. Currently shown in the engaged position. Turning it 180 degrees anti-clockwise will disengage it. The bend is to allow it to pull the top piece into alignment before securing it. In case it isn't clear, the top piece of RHS will pivot down from top left to bottom right in the diagram (the direction the pin is pointing) and the rotary latch/striker ensure it cannot move further in this direction.

Latch idea screenshot

Apparatus slightly 'open'

  • $\begingroup$ It is still a little unclear to me exactly the motion of these pieces, can you provide an "open" and "closed" macro view? I'm also confused why you need a second, stronger latch if the first one is working well. Also why the 2.5 safety factor? Is there inherent risk of injury if something fails? Otherwise 1.5 or even less would suffice. $\endgroup$ – jko Jul 30 '20 at 18:39
  • $\begingroup$ thanks @jko, I'll try to get that open & closed view together this evening to make it clearer. The structure is 2m tall/long and sometimes is just 'out of the way', allowing the user to use the floor space to exercise. So the basic latch makes this possible - to lift it up (helped by gas struts) and it automatically locks into place. It is now safe and the user hasn't had to perform any further action other than simply lifting the apparatus. When the user wants to use the apparatus for exercise, much larger forces are present. x2.5 was the figure given to me by an exercise safety expert. $\endgroup$ – Monty Jul 30 '20 at 19:07
  • $\begingroup$ @jko added another view/diagram showing it slightly 'open'. This 3D diagram I'm playing around with doesn't include the rest of the structure/apparatus, so it won't show the path exactly as I've just moved and rotated the upper piece slightly. The lower piece now shows some cut outs for the existing rotary latch. Does this make it clearer? $\endgroup$ – Monty Jul 30 '20 at 20:04
  • 1
    $\begingroup$ This is unanswerable if we don't know the load path of your 2-tonne force. For example If it all goes through the latch, the wimpy thing you have drawn won't last long. (And even if we do know the load path, IMO you don't design something to support a 2 tonne load unless you either know exactly what you are doing, or you want to apply for a Darwin award!) $\endgroup$ – alephzero Jul 30 '20 at 21:36
  • $\begingroup$ thanks @alephzero, I appreciate your concerns. My prototype has been working fine for 6 months, but is considerably different in design. Entrepreneurs don't always have experience or expertise in the area where their idea lies, and that's the case for me. I have a factory building a pre-prod model right now and we'll be adding a suitable latch mechanism. Before this gets anywhere near the public, it will go through strenuous official testing. Load is 130kg max user weight x 7:1 lever x 2.5 safety factor. Load path will be perpendicularly away from the bottom RHS, so bottom right to top left. $\endgroup$ – Monty Jul 31 '20 at 11:44