1

So, I have essentially a door lock in a box; a generic 60kg electromagnet (example), power supply (IRM-10-12) and control relay (G5LE-1A4). The electromagnet claims 150 mA, the relay is ~35 mA, and the PSU is good for ~850 mA¹. (All of this is 12V.)

In an open-air, climate-controlled environment, I probably wouldn't worry about it, but... this is "in a box". Specifically, an enclosed, metal box (approx. 80 mm wide, 40 mm deep, 130 - 200 mm tall, including the electromagnet) with minimal (and weather-resistant) ventilation, and which might be exposed to 45°C ambient temperature.

Questions:

  • Do I need active cooling?
  • If "yes", are a pair of NF-A4x10 FLXs (with probably 50Ω resistors) even going to be enough?
  • Is active cooling going to have a sufficiently meaningful effect that it may be worth doing even if it isn't strictly necessary?
  • More generally, how does one design for cooling requirements? In other words, what guidelines can be followed at the design phase for knowing whether active cooling is likely to be required, as opposed to making a prototype and testing to see if it overheats?

(¹ Probably overkill, but IRM-05 is not significantly cheaper, and I'd prefer more margin than an IRM-03 would provide. Also, I penciled in the IRM-10 in anticipation of it also needing to power fans.)

Matthew
  • 572
  • 3
  • 13
  • There is a whole field of thermal analysis of systems. There are also simulations that can be done by some professional software packages. FWIW, in the companies I used to work it was done by mechanical engineers, not electrical ones (well, the electrical ones had to supply data about individual components). – Eugene Sh. Jan 23 '23 at 19:23
  • Can you assemble it on the bench and find out how hot it gets without fans? – Andrew Morton Jan 23 '23 at 19:28
  • Does the manget needs to be energized all the time? Or only for "turing the lock"? If the former: how you make sure that the power to your box is not simply cut? – Christian B. Jan 23 '23 at 20:06
  • @AndrewMorton, hmm, possibly, though that isn't going to tell me how hot it gets in the enclosure. Also not sure how I feel about running mains through a breadboard, though perhaps you meant to assemble a PCB (which I guess would have to have traces for fan(s))? – Matthew Jan 23 '23 at 20:06
  • @ChristianB., only being energized when the power is on is a design requirement. – Matthew Jan 23 '23 at 20:13

1 Answers1

1

First of all, when heat is concerned, it is power that you have to look into, not current.

If I'm not mistaken, you plan on 12V (based on the magnet). So:

$$P_{magnet}=0.15 \times 12 = 1.8W$$

$$P_{relay}= 0.035 \times 12 = 0.42W $$

So a total of 2.22W without the AC/DC converter, which has efficiency of 82% typical, so losses are arround 20%, ie 0.4W

So a total heat of 2.7W to dissipate (lets say 3W to be safe).

The operating temperature for the magnet is 55°C max, so if you have 45°C ambient temperature, it gives you just 10°C margin.

Without doing any calculation, I'm not very optimistic that with a closed box without nearly any airflow you manage to dissipate the 3W through the air.

You might get better of by having the components directly connected to the box with thermal paste, but even so, I wouldn't bet on it being enough without testing, simulation or computation.

I think using a 55°C magnet with 45°C ambient temperature is probably not a good idea.

Also, a small frame challenge :

  • do you really need to wast 3W in permanence the whole day, during the whole year, just to avoid turning a key?
  • do you think that 60kg is enough holding force? Any strong person will be able to just pull open your door
Polynomial
  • 10,691
  • 5
  • 49
  • 88
Sandro
  • 7,277
  • 11
  • 37
  • I wasn't going to get into application requirements, but... yes, the intended application is not readily conducive to a purely physical system. As to strength, I don't anticipate the lower-power magnet being inadequate, but the electronics should work equally with a stronger one, if needed. – Matthew Jan 23 '23 at 20:06
  • 1
    @Matthew consider using the magnet to pull a deadbolt into position (against gravity). If it's hidden inside the box, then the user can't just oppose the magnet, he has to deal with the strength of the deadbolt. And you can use a much weaker magnet with lower waste power. – Ben Voigt Jan 23 '23 at 20:24
  • @BenVoigt, leaving aside other reasons that may not be practical... do you have a specific suggestion for such a magnet? I've seen very few magnets intended for continuous power besides the sort I'm looking at. (Also, don't forget such a magnet has to be strong enough to overcome whatever you're using to overcome residual magnetism. Even in general, I suspect there's a lower limit to how powerful a magnet can be while being reliable in such an application.) – Matthew Jan 23 '23 at 20:48
  • @Matthew overcome residual magnetism, you mean, your deadbolt getting magnetized because it's made of iron? you might consider using a different material, or, reversing the magnetic field direction so it helps you instead of hindering you – user253751 Jan 23 '23 at 20:52
  • If this is for a door, also considere what happens in case of a power outage : if magnet has to be on to close the door, your door is unlocked. If the magnet is needed to move a lock, then if there is no manual mechanism as well, in case of an power outage, you are locked in/out of the house. This can be a serious hasard in case of electrical fire (likely to also trigger the circuit brakers) if you don't have ground floor windows you can use to escape – Sandro Jan 23 '23 at 21:28