Help me to understand relay datasheet

[kanonka]

In particular, I don't understand "shock" numbers. I'm looking for a miniature signal relay for my circuit (up to 100 mA). Some of them list "shock resistance" of 50g, some go up to 300g. But, in the same document they also say "please avoid mishandling. Simple drop from a workbench to the floor can cause device damage". And this is where I got stuck. Simple drop from the table is 1g. They claim shock resistance "up to 50g". The how in the world device with rating of 50g can be damaged from 1g? What am I missing?

Basically, I just want to make sure that my final product can survive normal handling. And yes, electronic devices occasionally fall off the tables ("life happens"), and this is perfectly normal situation.

P.S. Please, don't suggest SSR - they just don't suit my circuit.

 

[KlausST]

Hi,

While a relay or any other thin is still or with constant linear velocity, then it feels gravity = 1g.
This is when you stand still on a personal scale. But as soon as you try to jup ther is added acceleration ad the scale shows decreased values. As soon as you are in the air the gravity is still there, but inside you ( or the relay on it's way to the floor) there is no acceleration. The same is inside an airoplan on an parabolic zero g flight.

With a free fall, seen from the earth, the airplane speeds up with 1g, while inside the airplane there is zero g.
( free fall physics is a bit different than parabolic...)


But as soon as the relay or the airplane hits the ground there is a huge (negative) acceleration. Some hundreds of g.Distroying.

Physically/mathematically:
If the relay falls from 2m down to a concrete floor, then it needs about 0.65 s. It speeds up to about 6.5m/s.
But when it hits the floor the speed is reduced within milliseconds.
If it is 2ms, the it is about 6.5m/s / 0.002s = 3250m/s^2 = about 330g.

Additionally the hit on the floor causes a lot of frequencies to acellerate inside the relay, and eventullly existing resonances gain the destroying effect.


Klaus

 

[mrinalmani]

No, that's not how you should interpret it!
A normal reaction force of 1G acts when an object is lying stationary on a surface.
When the object is falling freely, it does not experience any force. (Just like in space)
However, at the instant when a falling object hits the ground it experiences a shock (negative acceleration in this case) which may be more than thousands of G! The falling object comes to rest and reverses direction within milliseconds. This is a tremendous rate of change in velocity.
If you compressed a spring that would cause an initial acceleration of 50G and fastened your relay to it and then let it loose, your relay will survive the oscillations. But falling and stopping with a shock is beyond the Gs!
The G rating can be a good parameter for determining the RELATIVE ruggedness of relays, but it has nothing to do with falling.

 

[kanonka]

Physically/mathematically:
If the relay falls from 2m down to a concrete floor, then it needs about 0.65 s. It speeds up to about 6.5m/s.
But when it hits the floor the speed is reduced within milliseconds.
If it is 2ms, the it is about 6.5m/s / 0.002s = 3250m/s^2 = about 330g.
Klaus

Thanks! This helps to understand the issue, but now the practical question - how we determine that "stop time" is ~2ms? Obviously, it cannot be zero in real life - this would mean indefinite acceleration, and destruction into the dust . So, what would be a realistic number? Where to get it from?

Also, it seems like the best solution would be to amortize PCB inside the case - for example, add rubber washers in between bolt and nut on both sides. But then again, how to get "drop time" number from that? Simplest way would be just do an experiment , but it would prove nothing - I just might get a super-good variation of a relay that doesn't break , but "normal" ones would break. Any advice?

 

[KlausST]

Hi,

It depends on many parameters.

Imagin you jump.
From 1m with feet first it is OK. With head first it hurts.
Head first from 1m into water is still OK. But from 10m head first in water may hurt.
A crash with a car equipped with airbag and 50km/h on a concrete wall may be OK, but with a bycicle ....

Means: it depends on speed, on the materials, on the direction, on the crumple zone and other protection systems.

Klaus

 

[Borber]

Shock resistance expressed in g is measured on vibrating test machine. Object under test will be shaken with different amplitudes and frequencies and thus at different accelerations in order to find what it can survive.
As previously described free fall on concrete floor where change of speed in certain time what acceleration or deacceleration is may be higher than acceleration it can survive. Free fall conditions are not controllable and that's why it must be avoided.

 

[crutschow]

The shock and vibration G rating of a relay is the point at which the relay contacts may chatter or momentarily open. It's generally not related to the damage G rating, which likely is much higher.

If a relay is mounted in any type of enclosure it's unlikely to be damaged by a drop to the floor.