Bounceless switching: Aluminum foil to the rescue


A newly designed power supply required load testing with no contact bounce allowed. The load would be first unconnected from the power supply and then it would be connected using a mercury switch to ensure bounceless load application. The problem was that we didn’t have any mercury switches on hand nor could we find any to purchase. All such switches seemed to have been discontinued by their suppliers because mercury is a hazmat item.

This was once briefly discussed at “Bounceless Switching Without Mercury,” but we will look a little deeper into the issue as follows.

First consider how contact bounce arises as per the example   in Figure 1.

Figure 1 An illustration showing how contact bounce arises.

Starting with a relay or a switch in its open circuit position, we energize a relay coil or push a switch lever to make an armature’s moving contact come into touching a nearby stationary contact. However, when the armature contact first hits, it can rebound. There can be a momentary closure, then an opening and then a final closure. This is contact bounce.

In the above sketch, one cycle of contact bounce is shown; but in the real world, there can sometimes be several such cycles. This was forbidden to happen during our power supply testing.

Mercury switches have the virtue of no contact bounce. A blob of mercury within a glass envelope either engulfs two contacts or is separated from those two contacts depending on how the glass envelope is positioned as shown in Figure 2.

Figure 2 Operation of an SPST mercury switch.

The mercury switch simply does not bounce; but we couldn’t obtain any such switches so we needed an alternative means of bounceless switching.

Aluminum foil came to our rescue. Using a loosely packed ball of aluminum foil and a copperclad board, we devised the bounceless switch shown in Figure 3.

Figure 3 A bounceless switch using aluminum foil and a copperclad board.

It worked as follows: As I would SLAM the ball of foil down onto the copperclad board, the foil would collapse and not bounce. After some number of uses, the ball of foil would need to be “fluffed” out again; and when eventually that could not be done anymore, a new foil ball would be used.

When we tested the power supplies in this way, we had to prove our point with scope photographs to show that contact bounce had not occurred. We did exactly that and our equipment was accepted.

John Dunn is an electronics consultant, and a graduate of The Polytechnic Institute of Brooklyn (BSEE) and of New York University (MSEE).

Related Content







Source link