Microwave was dead yesterday. Opened up and the internal fuse had blown. After replacing the fuse I noticed that one of the door latch micro switches was stuck. On inspection, it was sticky so I cleaned it out with isopropyl alcohol, and fiddled with it, which loosened it up and it seemed fine. So I put the microwave back together and it worked fine for a bit, but a few hours later the fuse blew again and the switch wouldn't work at all, stuck fast.
So I took the switch apart to see what was going on.
I couldn't work out how the internals of the switch worked at all. It is a double pole like one of these.
formatting link
The only plausible explanation I have is that a high electrical current has welded two of the switch contacts together, in the open position (unpushed).
I've seen that in the past but the electrical weld has been weak, always broken apart with a little force. This one was rock solid like it was welded properly. Is this a genuine switch failure mode and does the repeated failure point to a likely error elsewhere in the microwave?
Relay RY7, on the microprocessor board on the lower left side of the schematic, applies power.
Notice the magnetron circuit is basically modular and self-contained. The HV transformer, filament winding, oil-filled cap and half-wave rectifier, are all on their own circuit. This minimizes the chances of a HV fault into LV circuits.
RY7 controls the feed to that chunk.
For safety, "Secondary Switch (Top)" is a safety feature. It interrupts the flow of current to the magnetron, but not to the microprocessor board, any time the door is opened.
"Primary Switch (Bottom)", over near the microprocessor board, is a logic level switch, indicating the "same thing" to the microprocessor.
When a microwave door opens (and microprocessor finds out), the mains circuit via RY7 will not re-energize until the door is closed. This means that "Secondary Switch (Top)" never suffers the indignity of its contacts closing with mains potentials ready to go, on either side. It is not "arc welding on closure".
However, the secondary switch does suffer the same current flow as the fuse, when the magnetron is operating. And the secondary switch will be subject to some need for arc extinction, if/when the contacts open with mains potentials on either side.
Microswitches (if this is what we're talking about as a switch), the piston is spring loaded. The switch should return to piston-out, if pressure is removed. The spring constant then, is all the force available to separate the contacts. the switch would be open with the door open, so the N.O. and COM contacts must be in usage right now. When the piston is depressed on the microswitch, the switch contacts close, because the switch is now in the non-default state.
If the switch will not close, too much material could be shaved off the top surface of the microswitch piston. Or, the door could be too loose, to apply sufficient pressure to the piston, to cause circuit completion as you desire.
Summary - the Secondary switch which is under mains current (full load), only opens with current applied. It never closes with RY7 enabled, thus causing the indignity of mains current flow established purely by that switch. The switch is there to interrupt current flow under abnormal conditions (microprocessor goes nuts, ignores Primary switch, doesn't remove RY7 drive). The switch contacts are already closed, when RY7 energizes seconds later, and mains current starts to flow. If the mains current was high enough to weld Secondary closed, only the microswitch spring would be there to open the contacts.
Root cause - abnormally high magnetron load indicating need to replace magnetron. Since fuse did not blow immediately upon replacement with new fuse, oil cap and rectifier are OK. Either Primary or Secondary switch could be defective, as loss of primary means microprocessor will not close RY7. Listen for RY7 closure for evidence microprocessor wants the power on.
I'd say that it does point to some kind of over current somewhere that is perhaps intermittent in nature, and when it does happen the switch is overloaded and the fuse blows. Those sort of faults are a swine to find, and especially on such lethal gear with thousands of volts and microwave energy floating about while you are poking about inside!
Thanks everyone. I binned it. Delay for parts and £20 quid for new switches and correct fuse, make repair impractical, even if I could fix the root cause.
I went up a wrong path assuming the broken switch had caused the problem, rather than the problem had caused the broken switch.
The point that was really surprising to me was the strength of the weld between the switch contacts, couldn't break it apart even with pliers/screwdriver etc.
It's a failsafe arrangement that with the door open (where the switch closes), places a crowbar short across the magnetron supply and if running, violently blows the fuse, welding the switch in the process.
The cause is normally a bad microswitch, can replace that for a quid or 2. Then need to check the short limiting resistor is still ok. Failure to check that is a safety problem.
Just to emphasise your point. The OP mentions an official spare is 20GBP, but such microswitches are generic and widely available. RS or Ebay for instance.
From memory, two switches, quick delivery ~£5.00 each. +Box of two types of non standard fuse, quick delivery about £10. Both fuses were
10amp but for some reason the microwave internal one was sightly longer than standard plug size. I blew the plug one, when I used it to replace the microwave one.
From China I could have bought 10 switches for not much, but it would take a month.
The problem was that I use the microwave all the time.
I'm still not sure what *caused* the problem. Adrian's comment about the fail safe crowbar short made perfect sense, but what caused it? Maybe as simple as the other switch? (the non crowbar one), maybe the crowbar one itself? But as the microwave is at the dump it is hard to tell.
I also said the switch was double pole, which was nonsense, as I didn't understand what double pole meant. (I think I meant two way)
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.