# Practical questions

Things that I sort of know but can't recall offhand.

Or happened to have more interesting details.

## Around the microwave oven

 This article/section is a stub — probably a pile of half-sorted notes, is not well-checked so may have incorrect bits. (Feel free to ignore, or tell me)

The most minimal microwave oven would consist of

• a 2.45GHz source (a magnetron)
• a chamber that is insulated for that frequency, and
• a way to guide waves from the source to the chamber (a waveguide)

Microwaves deliver energy most efficiently into polarized molecules, creating friction and thereby heat.

There are various molecules we can get to, but the most abundant in the list that is present in (most) food is water (though also fat and sugar(verify)).

It turns out that at 2.45GHz (wavelength of approx 12.2cm, 4.8")) we heat primarily water.

So any food with high enough moisture content can be heated in a microwave.

Unaffected materials include most types of glass, paper, ceramic, and many plastics, though there are exceptions.

### Uneven heating

Simpler designs will have some amount of standing waves, implying there are also areas receiving much less energy than others.

This is one reason you may see unevenly warmed food.

The rotating platter reduces this by moving your food through these spots, but isn't perfect.

Microwaves commonly have a stirrer, sort of a metal fan that helps jumbles the waves around, making for less pronounced standing waves. This helps, but isn't perfect either.

Instructions might include stirring halfway through, or letting it sit a while after cooking, depending a little on what kind of food it is.

Denser food sees only 2-3cm (~1 inch) of penetration.

If heating something large, the inner parts may need to be cooked via conduction from already-warm parts. If you want to do that evenly in a microwave, you'd probably turn the power down and wait a lot longer -- but at this point, you might as well use a regular oven, or pan.

It also turns out that ice does not absorb energy as easily as water, (it melts, but mostly because the energy gets delivered to water on its surface) so a frozen product will effectively warm slowly at first, then heat faster, and when something was partially frozen thing, that can make some parts be very hot and cooked while others are still frozen.

Combined with the last, it means cooking a large piece of frozen something isn't great.

#### Some plastics

Whether a plastic is microwave safe is not about it getting too hot because of microwaves, it's about what it might contain and release when hot (mostly because of hot food).

So if it has a microwave label, it's an indicator of what it does not contain: additions known to not less-than-great for you, or even just taste bad. (The main names here seem to be BPA, some phthalates, and some others)

### Other potential problems

#### Sudden boiling liquids

It is possible for some liquids to superheat, which for these purposes means their temperature is above their boiling point but do not look like they are boiling.

This doesn't happen easily. Pans and heaters avoid this because they heat locally, and it takes quite-uniform heating to avoid creating local nucleation sites for the bubbles (and movement) we call boiling to start.

It also requires fairly pure water - this isn't easy to do with most tap water.

The movement involved in picking it may be enough to shift things and start the boiling, as may adding something at a different temperature (spoon, another ingredient).

### Some more notes

Isolation

Microwave ovens are well isolated by the internal metal walls, by the door's grated metal sheet (because of the microwave wavelength), as well as the chassis.

The little microwave EM that gets out is several orders of magnitude weaker than what's happening inside; it's milliwatts rather than the kilowatt-ish rating the microwave probably has, and rather smaller than what is known to be able to hurt us, or even be perceptible as heating.

Microwaves won't get out the air exhaust, largely because of the wavelength involved combined with the designed physical characteristics.

Pacemakers could in theory be affected, with the small wires acting as antennae. It can't hurt to be overly careful, and you might want to avoid a job at a microwave radar site, but microwave ovens are quite unlikely to cause problems.

Microwave frequencies mean this is non-ionizing EM radiation, meaning that they won't destroy cells, just heat them (comparable to cell phones).

You wouldn't want to stick body parts inside of an over, because it will heat you just as well as it heats food, which working organs won't like. But the little energy that escapes is unlikely to be even felt.

# EM spectrum

## Quick(ish) reference

 This article/section is a stub — probably a pile of half-sorted notes, is not well-checked so may have incorrect bits. (Feel free to ignore, or tell me)

Note: There are a lot of infographics you can find that are much nicer than the below. (This is here for an experiment which I have not nearly finished yet)

Perhaps most broadly, you can see the spectrum as

• light (IR, visible, UV)
• harmful radiation (UV, X-ray, Gamma)

Or, a little more detailed:

 Name Wavelength Frequency (Hz) Photon energy (eV) grouping ionizing or not? blocked by atmosphere? Notes Radio 1 meter – 100,000 km 300 MHz – 3 Hz 1.24 μeV – 12.4 feV thermal [1] no lower wave radio is blocked, higher end not under ~30kHz (1km) there are only a few niche uses that can barely be called radio Microwave 1 mm – 1 meter 300 GHz – 300 MHz 1.24 meV – 1.24 μeV many applications are much like classical radio Infrared 700 nm – 1 mm 430 THz – 300 GHz 1.7 eV – 1.24 meV thermal/optical mostly blocked (the thermal/optical split is sort of arbitrary) Visible 400 nm–700 nm 790 THz – 430 THz 3.3 eV – 1.7 eV optical mostly passed Ultraviolet 10 nm – 400 nm 30 PHz – 790 THz 124 eV – 3.3 eV bond breaking yup mostly blocked bond breaking starts in the higher range of UV X-ray 0.01 nm – 10 nm 30 EHz – 30 PHz 124 keV – 124 eV damages life Gamma ray less than 0.01 nm more than 30 EHz more than 124 keV damages life

[1] - 'thermal' is the classical name, named not for feeling warm, but the mechanism: thermal radiation is EM that originates in thermal motion of matter. This is much wider than heat/infrared. But yes, you can warm things with much of this range if you're really trying. 'Thermal' seems contrasted with 'optical' but that's a fuzzy distinction too.

In a little more detail:

• Below 3kHz or so
static field
power-line
easily influenced by atmospheric changes, and by noise from devices, and the antennas would have to be impractically large
...so doesn't have a lot of radio-like uses
most broadly 'Radio waves' is 3Hz .. 300 MHz (100000km .. 1m)
and if you consider microwave part of it, up to 300GHz (1mm).
as just mentioned, there are few practical radio-like uses below a dozen or so kHz
a lot of the Radio/TV frequency band names come from incremental amounts of use during the era that radio and TV were in active development
microwave itself is arguably just the next "well, turns waves this small are also usable for similar uses" (they were initially harder to generate)
Examples:
TV, AM and FM radio typically within 50MHz..1500MHz (60cm .. 20cm)
mobile phones are largely at a few GHz (order of 10 cm)
frequency choice is also influenced by practicality of antenna size
there are radar-like uses between 3Hz and 100GHz (see e.g. SUMMARY OF RADAR OPERATIONS in [1], though their practical use is fairly specific to range)
though a lot of it is on the order of hundreds of MHz (longer range) to dozens of GHz (shorter range)(verify)

• the terahertz gap
roughly 0.3cm .. 30um, 0.1 THz to maybe 10 THz
so roughly "the bit between microwave and lower infrared"
called this because generation and detection here is inefficient for practical reasons, and not in mass production
sub-mm wave, is an often narrower part of this, used in astronomy

• Infrared, 1mm .. 0.75um, 0.3THz..400 THz. There is more than one sub-categorisation of IR, but we often go roughly by:
FIR (far infrared) (1mm..14um, ~0.3THz..20THz)
IR-C (14um..3um, ~20..100 THz)
IR-B (3um..1.4um, ~100..200THz),
IR-A (1.4um..0.75um, ~200..400THz)
Infrared is called thermal, but
that's partly because it's part of the (much larger) range of thermal radiation.
If you meant actual warmness
it's only perhaps half of e.g. the sun's thermal delivery
only part of IR is particularly warm, or practical to warm people - preferably far, but mid also works

• Optical
~700nm, 450THz - red
~600nm, 500THz - orange
~580nm, 520THz - yellow
~530nm, 560THz - green
~480nm, 650THz - blue
~450nm, 700THz - purple

• Bond-breaking / ionizing
'UV' is a very wide range (0.3um..10nm, 790 THz .. 30 PHz) that it is usually subdivided into UVA, UVB, UVC based on the varied effects.
UVA - 315–400nm, ~800THz, (tanning beds. The little UVA from the sun that makes it to the surface causes at most minor skin damage)
UVB - 280–315nm, ~1000THz, (sunburns, skin cancer)
UVC - 100–280nm, ~1200THz (used for sterilization, ozone generation. Will burn flesh if strong enough, and damage your eyes. The ozone it generates from oxygen isn't ideal either)
Notes:
there are other classification, like far/middle/near UV
and there is extreme ultraviolet, 10-120nm
Because of the scale, the numbers are not very precise. UV could be said to start maybe around ~750THz, while e.g. 700 THz is still just visible purple
EM starts to be ionizing somewhere above UVA, so UVB and UVC are considered ionizing.
Luckily, our atmosphere blocks almost all UVC, and a lot of UVB (and relatively little UVA)
Blacklights often straddle visible and a bit of UVA, so are mostly harmless
you can get UVC germicidal sterilizing lamps, but you don't want to [2]. You can usually tell by them having clear glass, because you need fairly special glass to even pass UVC. (No, blacklight/purple style UV does nothing in terms of germs)
window glass take out most UVB, though only some of UVA. This is part of why you won't easily get sunburned indoor - and your body produces less vitamin D if you never go outside.
• X-ray is roughly (10nm..10pm) 30 PHz to 30 EHz
Soft X-ray are those with lower energy (30PHz to 3EHz)
Hard X-ray (above ~5eV, 3EHz) have better penetration but do more damage
• gamma waves is above 30000000THz+ (30EHz), shorter than 10pm

## On EM and harm

 This article/section is a stub — probably a pile of half-sorted notes, is not well-checked so may have incorrect bits. (Feel free to ignore, or tell me)

For context

We are used to 'radiation' meaning 'nuclear' and 'danger'.

But any EM radiates. From the harmless visible light, to heat, to radios, to the harmful UV, gamma, and other nasty things in space.

We are used to 'photon' meaning 'visible light'.

But in physics the term is valid for all EM, from radio to IR to visible to gamma.

Harm how?

EM radiation can damage in one of two ways, and the distinction turns out to be a specific place, a specific threshold in the EM spectrum, which happens to lie in the UV range.

Below that threshold (what we call radio, thermal, optical), the only way to harm people is for there to be so much of it it heats them up too much. This is easiest with a specific range of IR, but possible with any.

Doing that enough to cook them is hard to do accidentally, and hard without them noticing (pain receptors being what they are).

Most natural sources aren't strong enough to bother us much - the sun's heat being the strongest example down here on earth's surface.

Most man-made sources of these are often not strong enough due to regulations, or if they are, they typically have fences around them.

But perhaps more practically, if the only mechanism is heating, it can't be damaging without first being noticeable (there are stories radio tower engineers having snacks melt in their pocket but not really noticing themselves), then painful, and only then being harmful. So this is not generally not an issue.

The only everyday example I can think of being potentially risky is a microwave - which is why those are shielded, and have interlocks based on its door.

Whereas e.g. your phone tops out at about 2 Watt (and it lowers it as much as the distance to the tower allows, to save battery), and most variants have a duty cycle of order of 1:8 or so[3], meaning the average power (which is what matters here) is an eight of that peak power.

Smush that between your face and hand hard enough that they recover all that power, half in your face and half in your hand, and that's still not an amount that matters. Your hand puts out more than that (you put out ~100W overall when relatively idle, hands are ~1% of your body surface).

Above that threshold, it turns out that it's no longer about collective effect - every individual arrival (and they are tiny and many) individually has enough energy to react with molecules.

(The table above lists photon energy because it's around ~10eV that EM quanta are strong enough to be ionizing, that is, break molecular bonds - in the UV range)

This means that any amount of it can and will do damage to our cells.

We typically call this ionizing radiation, because that's how it does that damage to our cells (and to other things).

If you really wanted to heat people, you would generally choose the highest frequency you can easily generate, because it more easily carries more energy.

Below ionizing radiation, the easiest happens to be the infrared range.

There is no accumulative cell damage until you use temperature that start cooking people, and people will notice that.

Our bodies can deal with a tiny amount of ionizing damage - and continuously do, in the form of background radiation and the small amount of UV(-B) that isn't blocked by our atmosphere.

and we still like to keep this low.

Note that there are other reasons for things to have more energy. For example

Aurora are charged particles (which?(verify)) from the sun, which need to go moderate speed to even make it into our atmosphere(verify).
But don't worry, it mostly happens 100-200 kilometers up, and only a small amount of it makes it to ground level, rarely enough to affect our infrastructure much (verify)
cosmic rays, which are actually particles, have lots of energy because the ones that we detect are typically going near the speed of light, making their effective energy high enough to be damaging and, when these are also charged particles are then ionizing.
But don't worry - they're rare enough individual events, so they bother computers more than us.

## How much gets to us?

Broadly, see this image from NASA, via wikipedia:

## Infrared notes

 This article/section is a stub — probably a pile of half-sorted notes, is not well-checked so may have incorrect bits. (Feel free to ignore, or tell me)

While infrared often refers to near-infrared, it's actually a fairly huge range, split into ranges in a few different ways[4].

One common seems to be the near/mid/far split (near/far from the visible spectrum, 380nm..740nm), like:

• Near-Infrared
• Wavelength approx 740nm to approx 2500nm
• IR LEDs are usually near-IR, typically somewhere within 700...1000nm
• most relevant to optical astronomy(verify)
• Mid-Infrared
• 2500–25000nm
• Far-Infrared
• 25000–1000000nm(verify), getting close to microwave region

### Infrared and cameras

Infrared around optical cameras mostly deals with the fact that camera sensors can see some amount into near infrared.

For reference, our eyes see ~700nm (~red) to ~400nm (~violet), while CCD and CMOS might see perhaps 1000nm to ~350nm (this varies - sensitivity shapes differ).

In other words, they look slightly into near-IR, and slightly into UV-A - but far too little to be useful as a thermal imaging camera, because it doesn't even cover a lot of near-IR. Thermographic cameras often sensitive to a larger range, like 14000nm to 1000 nm

So while you can modify a bunch of regular digital cameras / webcams to see infrared (mostly depends on how easily you can remove the IR-cut filter), you won't see a lot of the IR range.

Broadly speaking, there are two main things we might call "infrared filters"

• IR-cut filters
often a filter in front of a camera image sensor
look transparent but bluish from most angles (because they also remove a little visible red)
cuts a good range above some point (well, transition), often somewhere around 740...800nm
Since it's a transition, bright IR might still be visible. For example, remote controls (typically in the 840..940nm range) may still be (barely) visible, in part because they're actually quite bright and concentrated

• IR-pass, visible-cut filters
often look near-black
cut everything below a transition, somewhere around 720..850nm range
using these on a camera that has an IR-cut will give you almost zero signal (it's much like an audio highpass and lowpass set to about the same frequency)

You now have some options, like

cutting IR and some red - similar to a regular blue filter
passing only IR
passing IR plus blue - happens to be useful for crop analysis(verify)
passing IR plus all visible
passing everything (including the little UV)

So for DSLRs, you probably want those in lens filter form.

Actually, there's another filter in color cameras, namely the Bayer filter that basically are per-pixel color filters, making different pixels sensitive to different colors.

Bayer filters care mostly about the visible range and tend not to filter out much outside it. The seem to tend to pass above 800nm partially and roughly equally, so IR comes out looking white anyway(verify).

### Communication

Two-directional communication is typically half-duplex because a device can easily be blinded or confused by its own signal.

Consumer IR (TV remotes and such)

• Often uses a continuous pulse, not continuous sending
• Pulsing with low duty cycle also means you can pulse the LEDs with more current without destroying them.
• sending and a specific speed of pulsing is something that won't occur naturally, so helps confusion from environment IR.
• Carrier usually 38kHz. More generally it's somewhere in 33..40kHz or 50..60kHz, often 38kHz, 40kHz, or 36kHz
• In the case of remotes there are hundreds of variant protocols (that is, bit patterns that are specific to brands and devices)
• Universal remotes usually have a lookup table from brand-and-model to one of hundreds specific code sets that the remote supports
• and occasionally the ability to learn codes from an example

IrDA

• Speed: 2.4 kbit/s to 1 Gbit/s (faster speeds primarily at close range)
• Modulation: baseband, no carrier
• Has a few different layers