Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> there's usually another anchor point

The original design supposedly had two screws, but was manufactured with one. When you have two fasteners (typically nuts, as opposed to screws...) they can be mutually secured with lock wire. That is very common.

It is decidedly not common to lock wire electrical terminal fasteners in aircraft, or anywhere else for that matter. Typically lock washers and self-locking nuts are used. I'm not a certified aircraft electrical system designer, so I can't say definitively, but there are some pretty obvious reasons: First, these are typically small fasteners: lock wiring is hard enough on large fasteners. Lock wiring tiny little nuts with small gauge wire approaches the unreasonable. Second, lock wire is not insulated, so you would end up with a rats nest of exposed conductors leading to your terminals. Clearly unworkable. I suppose someone, somewhere has done something that involved insulated lock wire, but I've never seen it, wouldn't know where to buy it, and can't imagine how you'd employ it without abrading the insulation.

The problem here is the shared ground terminal. Stacking wires on ground terminals is common, stupid and a plague on electrical systems. The clamping force on a wire's ring terminal (and, thus, it's contact resistance, mechanical friction, ability to inhibit corrosion, etc.) is distributed among all of the ring terminals stacked on a stud, screw, whatever. Stack just one new terminal with another and you've cut the clamping force on both in half.

Half.

You are now the "engineer" and when someone dies as a result of your field engineered electrical system it's your fault.




I don't follow how stacking two terminals cuts the clamping force in half.

If a screw provides 10lbs of clamp force, the equal and opposite reaction is that the thing it is screwed into must resist with 10lbs.

If you put one terminal in between it must transmit all 10lbs through itself or else the forces don't balance out and something must be accelerating.

If you then stack another one in there all the force must be transmitted through it also. So the screw clamps with 10lbs of force and both terminals feel 10lbs of clamp force.

I just can't figure out where you got the idea of "it cuts the clamp force in half" but I'm interested to hear.


The clamping force is distributed over an area: \sigma = F/A. Adding ring terminals increases the total clamped area, reducing the pressure seen at any point on the surface. Since increasing from 1 to 2 doubles A (assuming each has the same contact area) these surfaces see half the distributed force at every point.

It's easy to visualize if you replace the two rings with one enormous ring (and fastener, etc.) while F remains the same: obviously the distributed force at any point will be low.

The distributed force is crucial. Friction in real mechanical systems is non-linear. Conductors made of real materials vary in yield strength. A correctly engineered terminal must account for force, yield strength, area, vibration, dissimilar metals and other factors to prevent back off, gas ingress (thus corrosion,) high resistance etc. Real engineers don't do all the materials science involved here and no one would trust it if they tried: they rely on published standards, authored in blood.

Stacking ring terminals torpedoes all that: what was (relatively) simple with one ring becomes unanalyzed and prone to failure when stacked.


Ah I see. That analysis works if you neglect the free body diagram of the system. I agree with your analysis if one larger terminal were analogous to two smaller ones. But it is not.

When a bolt applies clamp force it does so to every thing in between the bolt head and the anchoring threads. All of the force is transmitted along the bolt shaft from the mating threads to the head of the bolt and then back down whatever is in between the bolt head and the mating threads.

If this were not the case no fasteners would work. The only exception is when you have multiple mating threaded regions, rust, etc.

If you put one washer in between the bolt head and the threads obviously it must feel the full force of the clamping, or else some force went missing.

If you put two washers in between they're both still trapped in this identical clamp force situation.

This can be extended by induction about as far as you'd like. Certainly 2,3,5,10 washers. Even to 100. Eventually gravity and other things start to creep in at much, much larger scales. If you wanted to clamp 1 million washers this simple analysis would fall apart of course.


>The original design supposedly had two screws, but was manufactured with one.

I'm sure many here heard the saying(especially when doing FMEA): One screw/bolt is no screw/bolt. ;-)




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: