Undervolting magnetically held switches

Question

I have acquired a Honeywell ET series switch (best datasheet I could find) whose holding coil is rated to run on 28 VDC. I know it is often deployed in aeronautical environments where voltage can drop to near 24 V when operating on battery.

From my testing, the switch seems to hold reliably with voltages as low as 7 V. With such low voltages, the force required to bring the switch to the OFF position without cutting power is much lower.

Up until now, I've considered its coil analogous to the one of a relay. I know most of these really shouldn't be undervolted, but is that only coming from the fact their construction is usually of lower quality and thus wouldn't operate reliably on lower voltages?

Would I need to beware of undervolting this switch to something like 9 V to reduce the force required to turn it off mechanically? What precautions should I take if I go forward with such a design?

This is only for a personal project of mine and wouldn't be deployed at scale so any variance in the manufacturing of different switches is not a factor to me. I can adapt my system to the tolerances of mine if need be.

I find it very hard to find such information on the limits of MIL spec components since they are almost never used by most hobbyists.

Answer 1

Looking at the datasheet, it appears as though the solenoid is only used to hold the internal plunger in the activated position once it has been placed there manually with the switch baton, not to electrically remotely activate the switch, hence the absence of separate "pull-in" and "holding" current ratings. However, the switch is qualified for flight deck use on military aircraft, where the plunger must not move under extreme shock or vibration conditions so the holding force is probably far in excess of what you'd need to hold it in under "normal" use.

For casual, non-critical use, I'd do a few tests (let the coil warm up) to see where it drops out and then give yourself a healthy margin (25-50%) between the highest measured voltage and 28V.

Or you could try emailing Honeywell. I've had long, involved conversations with TDK-Lambda engineers about products that cost less than one of those switches.