Microwave oven shielding

[JimB]

Well if you look at it that way, a microwave oven is say 1000 watts, to make it safe outside the microwave you need to drop that to 1watt, that's 1/1000th the power. Only 10db attenuation required.

Wrong, a power ratio of 1000:1 is 30db.

About screened enclosures,
where there are holes in the enclosure, for a fixed hole size the effectiveness of the screening is lower at higher frequencies.
The effect is known as "Waveguide Below Cutoff" (Google it), essentially the attenuation of an electromagnetic wave travelling along a waveguide or through an aperture increases greatly below some frequency.

As mentioned earlier, the effect is well demonstrated by the effects of radio reception in a vehicle passing under a bridge.
Reception of VHF signals (about 100Mhz) is relatively unaffected as the aperture under the bridge is large compared with the wavelength of the radio signal.
Reception of MF (0.5 to 1.5Mhz) on the other hand, is severely affected because aperture under the bridge is small compared to the wavelength of the radio signal.

In this context the bridges I am thinking of are small, less that 10 metres high, bridges like the Forth Bridge, Humber Bridge or Golden Gate Bridge are an order of magnitude larger.

JimB

On edit:
I have just had a look at my microwave oven, the flange on the door is about 3.5cm wide.
The oven runs at 2400Mhz or so, which gives a wavelength of 12.5cm.
The door flange is a quarter wave choke at 2400Mhz, at 1900Mhz it will be much less effective, hence some of the effects observed with the mobile phone.

 

[rmn_tech]

Anybody ever heard of a Faraday cage ? The holes in the cage are a lot larger than a microwave mesh but I would bet that neither RF signal would penetrate the cage.

Just a thought........

 

[Sceadwian]

rmn tech you'd be wrong though =) Like I said this has been tested with my phone measuring the signal strength, I only lost 2 decibels with it in there. It's got me curious.

The screen being a faraday cage makes sense because the holes are much smaller than the wavelength but that doesn't make sense with the cell phone because the frequency is lower so it should actually block better. Sorry about the miscalculation on power, I forget it's calculated differently than voltage.

Little confused about how the door flange forms and RF choke though.

 

[ccurtis]

I have just had a look at my microwave oven, the flange on the door is about 3.5cm wide. The oven runs at 2400Mhz or so, which gives a wavelength of 12.5cm. The door flange is a quarter wave choke at 2400Mhz, at 1900Mhz it will be much less effective, hence some of the effects observed with the mobile phone.

That makes sense. Good observation. I'm inclined to think it is the door gap, as mentioned earlier also.

Update: As an experiment, I secured aluminum foil between the mating surfaces of the door and oven and closed the door with the phone inside. I then secured aluminum foil over the exterior of the gap that exists between the door and oven, all around. The screen mesh is left totally uncovered. The phone doesn't ring. I add that removing even a small section of the foil that is over the exterior of gap allows the phone to work. Granted, the exterior finish is non-conductive, but it tends to validate the choke flange explaination. Additionally, I have a small oven and a large one. The thickness of the door flange appears to be the same for both ovens (within 1/8 in, or so, as the eyeball can easily detect).

 

[kchriste]

Now that is just petty:

 

[Sceadwian]

I'd have to agree kchristie, I think I demonstrated a greater lack of understanding than you did, I didn't think my posts all the way through before I made em, and I don't think you posted anything that was wrong at all. If it makes you feel better you have a higher rep than me, mines only 670

I would like to understand the filter characteristics of the door seal dimensions better though, I find it interesting that you can effectively make a relativly narrow bandpass filter using only simple physical construction.

 

[kchriste]

Here is a detailed article about the choke formed by the door seal as described by JimB:
https://sweet.ua.pt/~f1011/SoELoMO.pdf
Not written in the best English, but it gives you the general idea.

 

[rmn_tech]

Sorry the point I was making is that I worked for a company that manufactured and serviced highly sensitive equipment.

The Equipment was sensitive to almost all RF radiation and the only way to calibrate it was in a Faraday cage. (Walk In) The mesh on the cage was at least 2 - 3cm and I can assure you nothing got in. (No SW, MW, LW, VHF, UHF, SHF, no RF at all.)

The holes in a microwave door are a lot smaller but with a 3cm mesh Faraday cage 2.4GHz can't get in.

odder and odder.

 

[Sceadwian]

rmn tech, the gaps on the door of the microwave are NOT part of a Faraday cage. Early microwave ovens relied on an electrical seal from the door frame but dirt and warping prevent this from being a good RF seal. It doesn't get in through the mesh you're right about that, it gets in through that gap in the edges of the door. I read that really nice PDF kchristie posted and it's pretty informative. Suffice to say it has nothing to do with the screen itself, it's the RF blocking properties of the door barrier. At 2.4ghz it's quiet good, enough to take 1000watts down far enough to only cause minor interference on my 2.4ghz phone less than 2 feet away But at 1.9ghz and 900mhz the two frequencies used by my cell phone carrier it lets a large portion of the power through.

 

[Mr RB]

The cell phone probably has TONS of gain and a gain control circuit, so when the signal drops heaps of db it just cranks itself right up.

And just checking my microwave door, the door is supported entirely on plastic hinges so the screen in the door is NOT earthed. It's probably slightly transmissive to the cell phone signal.

 

[Sceadwian]

Not really Mr RB, it should act as a reflector, though it might re-transmit a tiny bit, like ccutis said, he tried a foil seal on the door and it shut the phone right up. So the mesh is definitely a fine RF blocker above whatever the wavelength of the hole spacing is.