1
\$\begingroup\$

I am working on a critical circuit that drives a DC motor in an aircraft landing gear system. The electric motor is a simple DC motor where one polarity extends the landing gear and reversing the polarity retracts it. My question is are there any reasons why using MOSFETs instead of mechanical relays would present a reliability issue? Obviously it is important that the gear comes down when it is commanded to so making sure this is reliable is high priority.

\$\endgroup\$
  • \$\begingroup\$ MOSFETs should be more reliable than a mechanical relay if you use suitable derating criteria. \$\endgroup\$ – John D Feb 13 '17 at 0:42
  • 1
    \$\begingroup\$ As mentioned somewhere in "The City and the Stars", from Arthur C. Clarke": "No machine may contain any moving parts." \$\endgroup\$ – Dirceu Rodrigues Jr Feb 13 '17 at 0:52
  • \$\begingroup\$ This is not really my field. I think the relay will be more reliable and more serviceable, but will have a shorter service life. I would use a relay. What stops the gear, a limit switch? I think the relay has a better chance of surviving a lightning strike (which will eventually happen if it is a production airplane). Don't you have other designers or mentors you can ask? \$\endgroup\$ – mkeith Feb 13 '17 at 2:36
  • \$\begingroup\$ Which airplane (model, company) it will designed in? (to avoid the model in the future) :-( \$\endgroup\$ – Ale..chenski Feb 13 '17 at 2:39
  • \$\begingroup\$ Additionally, MOSFETs due to their wide availability and relative low cost, would allow you to implement series high side switches structures as well as low side structures and redundancy to allow you to retain control of the motor even if devices do fail \$\endgroup\$ – Kvegaoro Feb 13 '17 at 17:52
2
\$\begingroup\$

I have some background in high reliability electronics for space applications. The way you define "reliability" here is the key.

If you're thinking just in terms of the the random failure rate, a properly derated and protected flight-qualified MOSFET-based assembly can easily beat any equivalent relay.

But that's not the only thing to consider when chosing between those two technologies, of course. That would be too easy. :)

  1. Are there any specific derating or reliable use guidelines for MOSFETs in aeronautical applications that make them unconvenient for your company? Has this technology been used before for this very purpose?
  2. Are there any intrinsic advantages of one technology over the other? Such as higher immunity to single events (like lightnings, atmosferic radiation, etc.). Is there the possibility of single-event latch-ups (i.e. destructive response to an environmentally possible event) in any of those technologies?
  3. Are there any environmental parameters that impacts more to one technology than the other (vibration, shocks, temperature, thermally induced mechanical stress, etc.) and that can precipitate accelerated aging and an early wear-out failure?
  4. Which are the failure modes and their criticalities? Does any of the two alternatives have a significant advantage over it?

Sorry for raising more questions that answers, but your question can't be easily answered without having a broader view of the design problem.

\$\endgroup\$
1
\$\begingroup\$

Here's an excerpt from the link below showing that the life expectancy of solid state relays (using MOSFETs) is much greater than electromechanical relays:

"Using the daily number of operations from our previous example, this means using an SSR instead of an EMR could extend the life of the switching component in the oven [a hypothetical example in the app note] from 2 months (with the EMR) to 833 years (or 83 years at the lower end of the calculation, just to be prudent)"

Crydom application note

Of course the design has to be appropriate for the environment, including ESD, weather and temperature conditions, transient voltages, EMI, etc.

It's also possible to use additional FETs (or relays) such that any single switching component failure does not cause loss of functionality, though that's much more complex and costly.

\$\endgroup\$
0
\$\begingroup\$

Well, relays can be easily replaced and can generally take much more abuse than mosfets can. I also imagine that, at least depending on the mosfet, some condensation due to temperature changes during flight could cause the mosfet to act in weird ways.

For example if lightning or whatever hits the plane, it is much more likely for the mosfet to break compared to the relay.

And also, if you use a relay and it breaks, you can borrow another from another device on the plane temporarily to operate the landing gear in emergencies, whereas a mosfet isn't commonly found in other places.

\$\endgroup\$
  • \$\begingroup\$ The MOSFET could very well be found in other subsystems on the plane if the airframe manufacturer stipulates the common use of parts. However the MOSFET is far more likely to be soldered in place making swapping harder whilst the relays could be socketed. \$\endgroup\$ – Michael Karas Feb 13 '17 at 1:06
  • \$\begingroup\$ Yeah but seeing that he's designing the motor control and is concerned about using a mosfet, one could assume that it doesn't originally come with one. Either way in a situation where one needs to react quickly, there isn't much time to unscrew the mosfet, place the new one in, and attach it to its heatsink... \$\endgroup\$ – Bwinzey Feb 13 '17 at 1:08
  • \$\begingroup\$ I do have to beg the legitimacy of the lightning argument. The relay itself may be more rugged than the MOSFET but BOTH are highly likely to be controlled by electronics circuitry that is susceptible to upset and damage by severe lightning jolts. So in that case neither subsystem can be considered overall reliable. This is one reason that some critical systems on a plane are still equipped with entirely mechanical control systems and linkage override systems. \$\endgroup\$ – Michael Karas Feb 13 '17 at 1:11
  • \$\begingroup\$ Safety critical systems cannot be blindered to just one component or part. It is essential to consider the entire ecosystem and evaluate that to as much detail as possible. It is also RIDICULOUS to think that aircraft crew, in the case of a critical emergency, are going to be trouble shooting and trying to swap parts in the seconds or few minutes that remain before the final calamity. Any redundancy and override rollover mechanisms need to be designed in from the get-go. \$\endgroup\$ – Michael Karas Feb 13 '17 at 1:19
  • \$\begingroup\$ If it is a small airplane, the gear descent may be controlled by a switch and simple relay logic. You flip the switch, the motor puts the gear down until it hits a limit switch. You put the switch in the up position, the motor retracts until it hits another limit switch. \$\endgroup\$ – mkeith Feb 13 '17 at 2:45
0
\$\begingroup\$

Mechanical relays dont care of your smartphone's or other medium intensity rf fields. Requires a microwave owen to make some effect. Semiconductor parts are much more sensitive. Maybe there's no radio transmitters built to be in a near contact with your relay, but some non-thinking idiot can carry in one just when an aeroplane is landing.

Mechanical relay tears and wears in the use. But so does your motor. This kind of stuff has a regular pre-emptive maintenance program. The wear-out prone parts will be changed and refurbished before anything happens. The relay and the motor could be in one swappable assembly.

\$\endgroup\$
0
\$\begingroup\$

As far as average lifetime goes, the MOSFETs will probably win if properly derated and protected with surge absorbing devices.

Reliability- I would say it would go direct switch, relay, MOSFET, in order of decreasing reliability. It takes very little energy to cause a MOSFET to fail 'on'. There are plenty of sources of spikes in most aircraft. A relay will seldom fail in the middle of its useful lifespan.

Whatever you do I suspect you'll want a reasonable and well-documented plan 'B' available to the pilot. What happens if the main bus power goes down?

\$\endgroup\$
0
\$\begingroup\$

MOSFETs would be more reliable than a relay due to total cycles, but you probably have a larger issue. I would suggest using a BJT for the following reasons:

  • The MOSFET is sensitive to ESD. When you are flying around, you are actually creating charge. Planes create lightening. I would not be surprised to find that you blow out the MOSFET gate due to a static discharge.

  • Relays have a fixed amount of cycle times, cold metal shrinks, vibrations can destroy the aperture.

The advantage of the BJT is that is can self-heat down to about -80C, and it is largely immune to ESD. The scariest part of any implementation is always the mechanical components. You could also use the BJT current as an indication of up/down instead of limit switches by having a resistive shunt that would make it completely solid state.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.