First a question for the OP please? How far in metres or yards is your breaker board from the supply transformer? If your supply is overhead you ought to be able to see a round drum like object to which your supply cables connect to.
====================================================================
Now to my learned colleagues. I have taken literally what the OP has said - the mcb has burned out - so it may well be the microwave oven is faulty in some way - but I suspect not since it is new. Anyway, it seems to me this MWO is a powerful model - 1700W. The Panasonic we had before disposing of it was only circa 800W.
The way the high voltage is generated for the magnetron is something like a step-up isolating transformer, capacitor C1, charging/swing diode D1 see:
https://www.researchgate.net/figure...in-a-microwave-oven-The-switch_fig1_234930316
When energised with HV the magnetron is pulsed on/off at to produce bursts of microwave radiation with a repetition rate equal to the mains frequency 50/60Hz.
Such a circuit will produce a primary current waveform which is not sinusoidal since the diode conducts on one half of the cycle and the magnetron on the other -
once there is sufficient applied emf for forward biasing of the diode and circular thermionic conduction in the magnetron. You get the gist.
Such a power supply has a low power factor - I'd guess 0.65. Thus for a 1700W real power output the apparent power is 1700/0.65 = 2600VA If the supply voltage is 120V then the rms current is 2600/120 = 22A when the MWO is operated continuously at 1700W. I doubt this happens since it would quickly turn the contents of the oven to burning carbon. The magnetron's HV will be further pulsed on and off to achieve a lower average cooking power. But for the on periods the current will be circa 20 -22A - the
continuous average rating of the breaker.
The cycling on and off of the magnetron is normally done on the primary side of the transformer by crudely switching the supply. No switching to minimise transients then - unless the MWO uses an appropriate solid state switch. The transient current for such a transformer rectifier combination can be up to 10, maybe 20 times the steady state current with a decay dependent on the L/R time constant.
The current waveform will have a high crest factor - the ratio of peak current to rms current - greater than the usual purely resistive (PF=1) circuits sinewave crest factor of 1.414. And because of what I said in red above will be of short duration something like this:
https://www.researchgate.net/figure...orms-in-a-Typical-Peak-Rectifier_fig1_3627656
I wonder whether then the Square D QO 20 A breaker has unsuitable overcurrent/short circuit characteristics and
contact type to pass the peaky high current short duration waveform causing the magnetic trip to operate though I think the OP suggests this has not happened since the lever does not move.
Instead the contacts burn out because they are subject to a current density far higher than they are able to cope with - resulting in gradually and eventually catastrophic failure - they burn out. The Joule/Ohmic heating of these contacts depends on a square law which rises very rapidly with increasing current I/current density J eg: I/Isq = 20/400, 21/441, 25/625, 30/900, 100/10000, 200/40000. These very high peak currents albeit of short duration can cause very high local heating of the contacts. And small contacts will suffer most.....
Interestingly, thanks to Westward10's identification of the breaker as type QO, there is a specialist range of QO breakers viz QO - HM and QO -HID to handle one off switch on surge currents and cyclical surge currents respectively. See page 8 of:
https://download.schneider-electric.com/files?p_Doc_Ref=0730CT9801
QO® and QOB Miniature Circuit Breakers Special Application Circuit Breakers 8 07/2008
QO-HM and QOB-HM High Magnetic Circuit Breakers QO-HM and QOB-HM high-magnetic circuit breakers are recommended for area lighting (such as athletic fields, parking lots, and outdoor signs), when using lamps of inherent high inrush current, individual dimmer applications or other applications where high inrush currents exceed standard tripping conditions. These circuit breakers are available in one-pole 15 and 20 A ratings only. QO-HM and QOB-HM circuit breakers are physically interchangeable with standard QO and QOB circuit breakers and accommodate the complete range of QO accessories. QO-HM and QOB-HM circuit breakers are manufactured with the magnetic trip point calibrated at a much higher level than standard QO and QOB circuit beakers, as shown in Table 3.
QO-HID and QOB-HID High Intensity Discharge Circuit Breakers QO-HID and QOB-HID circuit breakers are for use in high intensity discharge (HID) lighting systems, such as systems using mercury vapor, metal halide or high-pressure sodium lighting units. These circuit breakers are designed to handle the high inductive loads, harmonic currents and cycling which are inherent in HID lighting systems. QO-HID and QOB-HID circuit breakers are physically interchangeable with standard QO circuit breakers and accommodate the complete range of QO accessories. QO-HID and QOB-HID circuit breakers are manufactured with larger contacts than standard QO and QOB circuit breakers to allow switching of high inductive loads. They also have magnetic characteristics similar to QO-HM and QOB-HM high-magnetic circuit breakers to allow the circuit breaker to hold in against the high starting inrush currents which are typical in HID lighting systems.
Or something along these lines... ?
====================================================================
For the OP - I am providing no advice just yet on what to do.