2

I'm trying to understand the lightbulb monitor section of the following circuit, the 2200 uF cap, zener, choke, and bulb. Everything else is a standard microwave oven circuit.

The magnetron will not pass current until the filament warms up, a few seconds after power is applied, at which point it will pass 300 mA. The bulb monitors the magnetron current, and starts out dark and then lights when the magnetron is passing current. This is used as a "properly functioning" indicator light for each magnetron in a system with 12 magnetrons.

enter image description here

I understand that the 200 uF cap has 1.2 ohm reactance voltage drop, which is capped by the zener and filtered by the choke to provide current to the bulb.

What I don't get is how putting that in series with the HV rectifier monitors the current in the magnetron. Hypothetically, if the magnetron were missing then rectified AC would still be present across the capacitor and the bulb would still light.

Can someone explain how this works?

(From an online Article)

Based on explanation below:

Looking at it in terms of charge leads to understanding.

When the magnetron is cold, it's open circuit and the EMF of the winding draws a large amount of charge into the HV capacitor. This causes a large voltage drop across the capacitor, but the charge has nowhere to go - it's blocked by the HV diode and the magnetron is open circuit at that point.

When the magnetron heats up, some of the charge bleeds off into the magnetron, which is replaced by the EMF of the coil on the next half cycle.

The HV cap and HV transformer winding is essentially a reciprocating charge pump.

Thanks for taking the time to explain it.

PWalsh
  • 23
  • 4

1 Answers1

1

All the current through the magnetron goes through that HV capacitor. The average current through the capacitor is zero (or it would build up voltage until something broke). So the same current that goes through the magnetron in one direction must go through the diode in the other. The lamp + choke || zener || capacitor smooth that mains-frequency current, and limit the voltage to what the lamp can handle by dumping excess current through the zener.


Here is a simulation that illustrates what is going on in the average:

enter image description here

For Rmag = 10k, 100k, 1000k we get I(D1) of 29.2mA, 4.64mA, .495mA respectively and I(Rmag) of -29.2mA, -4.64mA, -.495mA respectively.

The simulated average capacitor current is < 1uA (it's not exactly zero because of numerical errors).

Here is what it looks like close-up cycle-by-cycle in steady-state:

enter image description here

The red trace shows the capacitor current. The area under the positive half and area under the negative half are the same, so the average magnetron current flows through diode (and thus the lamp and parallel parts).

Spehro Pefhany
  • 376,485
  • 21
  • 320
  • 842
  • I don't understand your answer. Why is the average current zero? I see no reason why the current through the HV cap on one half-cycle can't be different from the current in the opposite half-cycle. If the different cycles present different voltages to the cap, then the total *charge* on the cap would be zero, but the current could be different in the two cycles. And the voltage drop of the cold magnetron is very high while the voltage drop of the HV diode and small cap would be fairly low. Can you elaborate on your explanation? – PWalsh Dec 24 '22 at 20:58
  • The voltage across the capacitor is proportional to the integral of current wrt time. Since the voltage across the capacitor is the same at the beginning of each cycle in steady-state the average current must be zero. The average current over the half cycle charging the cap is thus equal to the average current over the half cycle discharging the cap. When the filament is cold the capacitor will accumulate a large DC charge and the average current will be low in both half cycles. – Spehro Pefhany Dec 24 '22 at 22:03
  • When the filament is hot the capacitor discharges during the half cycle and more current through the rectifier is required. – Spehro Pefhany Dec 24 '22 at 22:03