The Heat Death of 5G (digitstodollars.com) 92
An anonymous reader shares a report: Yes, 5G is coming and data rates will improve, but we, the mobile industry, still have a lot of work to do. We could regale you with litanies of woe about roaming and hand-offs, or belabor the small cell backhaul density logjam. But perhaps the best example of roadblocks to 5G is much easier to grasp -- Heat. 5G phones get hot. Really hot. Probably not hot enough to ignite your battery (probably), but enough to generate a definite burning sensation in your pants pockets. At Mobile World Congress in February, we spoke with an engineer from Sony who was demo'ing a phone (behind glass) that was clocking 1 Gbps speeds. Wow, fast. We asked the engineer why it was not going faster and he said "It overheats." A good solid answer, from a nuts-and-bolts-and-antenna person. We will wager any amount that at next year's show, no one on the floor will be as open about this problem.
The big improvement in data rates for 5G will only come with mmWave radios. This is a whole new spectrum band that allows for really high data rates (again, let's set aside the whole densification issue for now). The trouble is that mmWave radios generate a lot of heat. To greatly oversimplify, mmWave frequencies are pretty close to microwave frequencies, as in the thing we use to reheat our lunches. From some of our very recent industry conversations we know that the handset industry is using a tried-and-tested method for dealing with this problem -- ignoring it and hoping it goes away. The whole issue strikes us as one of those issues where middle management really does not want to raise the subject with senior management who have wrapped themselves so tightly around the 5G flagpole. "Uh boss, your pants are literally on fire."
The big improvement in data rates for 5G will only come with mmWave radios. This is a whole new spectrum band that allows for really high data rates (again, let's set aside the whole densification issue for now). The trouble is that mmWave radios generate a lot of heat. To greatly oversimplify, mmWave frequencies are pretty close to microwave frequencies, as in the thing we use to reheat our lunches. From some of our very recent industry conversations we know that the handset industry is using a tried-and-tested method for dealing with this problem -- ignoring it and hoping it goes away. The whole issue strikes us as one of those issues where middle management really does not want to raise the subject with senior management who have wrapped themselves so tightly around the 5G flagpole. "Uh boss, your pants are literally on fire."
Seriously? (Score:5, Interesting)
With the RF transmit power of your average phone well under 1000mW, are the authors of the story seriously suggesting that RF heating is an issue?
Hey, let's not allow science to enter the argument eh?
Really, if that much of the RF energy was being turned into heat then the range of these things would be next to nothing.
Perhaps it's just that the silicon isn't very good at turning electricity into RF at these frequencies and thus a higher than expected amount of the battery's energy is going directly into creating heat (I2R losses mainly). It's *NOT* because the mm-frequency radiation is causing heating via RF absorbsion.
Sigh!
Re: (Score:2, Funny)
While this is a very good post, I don't think it's proper to have an exclamation point after "sigh."
Re:Seriously? (Score:5, Funny)
Re: (Score:1)
Re: (Score:2)
This is a dumb article for exactly the reasons you point out.
The problem with heating is exactly the same as it was for 4G when it first came out: Processing requirements for the specification and the PAs required to do the transmission. Not the transmission itself.
"Sigh!" indeed.
Re: (Score:2)
The problem with heating is exactly the same as it was for 4G when it first came out: Processing requirements for the specification and the PAs required to do the transmission. Not the transmission itself.
"Sigh!" indeed.
Yeah. I bought an HTC Thunderbolt when it came out (one of the first 4G phones). Had horrendous battery life (maybe 6 hours) and got very warm when using data. I had to turn off 4G most of the time to get any kind of battery life out of that thing.
Re: Seriously? (Score:3)
It's a dumb article for other reasons too. This part in particular made me bang my head off the desk:
To greatly oversimplify, mmWave frequencies are pretty close to microwave frequencies, as in the thing we use to reheat our lunches.
Not only is the idiot conflating the phrase "microwave frequencies" with "microwave oven frequencies", but he obviously has no fucking clue that good old 2.4ghz WiFi operates in the same rough frequency range as a typical microwave oven. So by his logic your phone should already be overheating every time you use WiFi.
He apparently "greatly oversimplified" his own ignorant interpretation of something someon
Re: (Score:1)
How? If you don't know you don't know. (Score:2)
Or don't know what specifically you should do searches for.
Re: How? If you don't know you don't know. (Score:1)
Re: (Score:3)
Thanks for stating the obvious. Sad to see such an article here. It checks all the boxes. ... as if there were no study on this)
- Fear from future technologies
- Technical nonsense (confusion about high freq waves and chip power/heat)
- Conspirationist theories (the usual : companies don't want to know, "tried-and-tested",
- The out-of-context quote from an "expert" from a well-known company (Sony) so that every person can assess it's serious.
- A clickbaity title.
Sigh.
Re: (Score:3)
Okay, then how about this. We don't need 5G technology if it's
- gonna require littering our sidewalks with transmitters with much less range than the existing cell transmitters - which already litter all of our rooftops, but at least provide a useful service.
- not going to be everywhere - ever, for the same reason. Who's gonna put expensive transmitters in any but the most densely populated areas?
- gonna provide services nobody really wants to buy. Yes, there's such a thing as 'good enough'. Today's wir
Re: (Score:1)
- gonna require littering our sidewalks with transmitters with much less range than the existing cell transmitters - which already litter all of our rooftops, but at least provide a useful service.
- not going to be everywhere - ever, for the same reason. Who's gonna put expensive transmitters in any but the most densely populated areas?
So which is it? Littered everywhere or not?
OK. Just the useful places then.
- gonna provide services nobody really wants to buy.
Except these people
Out on the street, only an idiot 'really wants' that. It's a phone/text/email/google query device. Anything more, and you're one of those idiots walking down the supermarket aisles, face-timing your spouse about whether to pick up bananas.
OK. Just the people who want to use it will use it.
Yes, there's such a thing as 'good enough'. Today's wireless is essentially that.
Nobody will ever need more then 640k
Maybe you won't need it. So don't get a 5G phone then. I hear Apple still make 4G phones, try one of theirs.
Re: (Score:2)
The range *is* next to nothing though. ;) (Score:2)
Agreeing with your comment nonetheless.
Re: (Score:2)
Re: (Score:3)
Re: (Score:2)
With the RF transmit power of your average phone well under 1000mW,.
Under normal operational conditions it is way under 25mW. Once upon a time I used to design cellular equipment (one of the predecessors of what we know as the femtocell) and our cells maxed at 25mW.
In fact, 5G should use LESS power, not more because it very heavily relies on MIMO and beam steering. That is the real reason why 5G deployments are such a bugbear for people like Mike Pompeo, nothing to do with the network side of the security equation: https://www.fagain.co.uk/node/... [fagain.co.uk]
In any case, back to
Re: (Score:2)
How you put a directional anthena in a cell phone without a need to orient your phone properly when communicating?
Phone antennas, and practically all antennas, are directional anyway. What they are referring to is an antenna with multiple elements with controllable gain and phase for each element acting as a phased array which is what MIMO systems rely on.
Re: (Score:2)
In any case, back to the RF bit - as your RF is now directional you should be using less power than before. Not more.
Path loss is roughly the same (it is worse but can be ignored for this) but the basic antenna size is smaller yielding a smaller aperture. So the phased array is used to both make up for a smaller aperture and reduce multipath. Unfortunately, this does nothing to decrease power requirements and the the whole thing is less efficient. The only gain comes from using a smaller cell size.
Why would you think wavelengths don't matter? (Score:3)
The peak power for cell phones are typically:
GSM: 28-30dBm
UMTS: 24dBm
LTE: 24dBm.
This is of course only the transmit power of the radio. Each band has it's own power limit in the US, so the phone maker will use an antenna with the appropriate gain in order to be under the EIRP limit. FDD4 and FDD7 have a limit of 1 watt, where as other bands like FDD5 have a limit of 11 watts.
Though, you won't see that on a cellphone since there is no possible way you could ever pass the SAR requirements, which has a frequen
Re: (Score:2)
You are partially correct. Any time you increase frequency significantly, amplifier efficiency drops. Innovations in semiconductor technology will help though. The other issue is the change in modulation techniques and the digital processing required to handle the new modulation techniques. This also consumes a lot of power. This will reduce significantly as the technology progresses.
Re: (Score:2)
With the RF transmit power of your average phone well under 1000mW, are the authors of the story seriously suggesting that RF heating is an issue?
The efficiency of the RF power amplifiers at higher frequencies is poorer and phone design was already power dissipation limited making the situation worse. And that is after taking into account the smaller cell size.
Heat, eh? (Score:3)
If it doesn't make your battery ignite, it will almost certainly shorten its life.
So far, not impressed (Score:4)
I'm been hearing that 5G is the next big thing since merkins and witch burning. I've seen several review videos about how you only have line of site access and it only delivers the promised speed if you are standing right next to the pole.
I'm not buying into the hype just yet.
Re: (Score:1)
Would you buy it at half the price?
This 5G stuff is going to cause a lot of headaches
How does one so completely miss the point? (Score:2)
He clearly knew that. Because he said so.
And it was in fact one of his points.
As opposed to your existence, which seems to have none.
Re: (Score:3)
I do not think so as he also said:
"mmWave frequencies are pretty close to microwave frequencies, as in the thing we use to reheat our lunches"
microwave frequencies? as in "microwave oven" really?
Unless you hang yourself on the meaning of "pretty close". The mm waves lengths are somewhere above ~24GHz while the ovens operate at 2.4GHz (ISM band).
The mm frequencies that the 5G is after are around 28G, 38G, and 62GHz.
Re: How does one so completely miss the point? (Score:1)
Re: (Score:3)
Frankly as a mobile phone, the higher frequencies will cause serious reception issues. No question in theory that it will improve data rates, but for mobile devices is that really a big problem? It hasn't really been an issue for me.
The big thing I see with 5G is fixed wireless to residential and small business customers. Instead of costly cable and phon
Re: (Score:3)
Frankly as a mobile phone, the higher frequencies will cause serious reception issues.
5G is not about high frequencies, and unlikely to be widely used at those frequencies. The mm band is just one of the things the new standard enables.
Is 5G an over-hyped solution in search of a problem or am I missing something?
You're missing something. You're missing a lot of things. Hell improving subscriber count per base station, and less airwave contention, alone would be worth the technology change even if it didn't bring about a single bps of speed improvement.
Re: So far, not impressed (Score:2)
"Costly wire"
RG-59 installed 30 years ago is still viable for gigabit DOCSIS 3.1 over distances similar to those that 5G can hope to serve it (curb your home). 30 year old data radios have been in the scrap heap for 20-25 years.
Investing in fixed line service is the best long term use of money... but that isn't what carriers care about.
Re: So far, not impressed (Score:1)
Here's how you can really tell it's not ready (Score:2)
I'm not buying into the hype just yet.
You can tell it's not even close to being really ready, because there is so little complaint Apple doesn't support it in the new iPhones. Not even the Apple Haters are willing to claim 5G is a very useful feature at the moment to call out for.
Re: So far, not impressed (Score:3)
Good choice. I just read the requirements for the number of antennas they have to place in Amsterdam to get meaningful coverage. They can completely forget about that ever happening.
So I don't know where 5G will be rolled out, but it seems unlikely to be rolled out anytime soon. Loads of problems still to be solved.
Re: (Score:2)
I've seen several review videos about how you only have line of site access and it only delivers the promised speed if you are standing right next to the pole.
Then you and the reviews you are reading have no idea what you're talking about. 5G and the line of sight issues related to the millimeter wave have almost nothing to do with each other beyond the fact that the former standard enables the latter to be used.
Personally I cant wait for 5G and all the benefits it will bring to normal signals in the existing LTE spectrum.
Wifi alternative (Score:1)
With Laptops that can better manage their heat, it would be interesting to see 5G embedded into them as a replacement to 802.11[x]. Maybe thereâ(TM)s a future there too.
Close to microwave frequencies? (Score:5, Informative)
Your lunch is cooked with the same frequency as old school 2.4ghz wifi.
Re: (Score:2)
That might explain the burning sensation in my ear when my wife calls me.
Re: (Score:2)
As the name implies mmwave frequencies have wavelengths under 1cm (hence millimeter wave). By contrast your "microwave" and your WiFi use frequencies with around 12cm. The frequencies recently acquired by Verizon for their 5G deployment, 38GHz have a wavelength of around 8mm.
Re: (Score:2)
Irrelevant. The 2.4 WiFi band covers a wide spectra and only the few frequencies close to the resonance frequency of water have any effect on heating food.
Learn how microwaves work instead of spouting nonsense.
Re: Close to microwave frequencies? (Score:2)
Ehm? Wifi is 2.41 to 2.48 GHz (not particularly wide-band) and microwave ovens operate at 2.45 GHz. That frequency is not a resonance frequency. See https://en.m.wikipedia.org/wik... [wikipedia.org]
You can heat your lunch with a microwave oven and not a wifi router because one is 700 W and the other is 20 mW.
No (Score:4, Insightful)
It doesn't have anything to do with the wavelength. The heat is 100% determined by the energy used from the battery. If it heats up, it's because the transceiver and the other chips use more energy from the battery. Battery life is probably pretty bad on that phone.
You can look forward to buying a better phone next year, and probably the year after that. If you bought the first 5G phone, you got the one that's missing all the optimizations in next year's and future years' phones.
Too Hot (Score:2)
Too hot to handle
Too cold to fold
You call the ghost busters and their on patrol..
Just not on your 5g phone, those things are dangerous.
Re: (Score:2)
Hard to handle [youtube.com]
Microwave frequency bullshit. (Score:2)
A microwave oven doesn't work like that.
Its frequency is like that of wifi. But that alone does not heat the food.
It is a *rotating* field. It rotates polar molecules (like water).
You would need 500 phones with active UMTS calls at 1m distance, rotating around your food, for it to have any effect.
What we now get a bit (!) closer to, is terahertz radiation. Like in body scanners at airports. So penetration is not as deep, but stronger.
You need to go even (a 1000 times) further and worse, you end up with ioni
Re: (Score:3)
No, nothing special and "rotating" about the field that doesn't happen in any other microwave field.
The polar molecules rotate, but the source doesn't need to, lolz.
It's purely a question of power. No rotation of sources required.
Re: (Score:1)
Giggity!
Re: (Score:1)
"It is a *rotating* field. It rotates polar molecules (like water)."
You might want to go re-ask your high school physics teacher about that.
Comment removed (Score:3)
Reading this with the right inner voice. (Score:2)
Reading this in AvE's voice, with his delightful Canadian accent and adding a bit of his prose, makes it all sound just right.
Diamond Age (Score:2)
Everyone who has not read Neil Stephenson's Diamond Age, please go read it now, and shame on you for not having read it already. Those who have read Diamond Age would do well to read it again, it's a rollicking good ride.
Headline writing (Score:2)
Oversimplification? (Score:2)
"mmWave frequencies are pretty close to microwave frequencies"
Sounds like more of an error to me.
mmwave 30 GHz - 300 GHz.
microwave ovens 2.45 Ghz.
Re: (Score:2)
Microwaves are from 300 MHz to 300GHz, it includes mm waves.
Anyway, you could make a microwave over at 900MHz or at many GHz.
Re: Oversimplification? (Score:2)
It's close enough. The word "resonance" is misused, but 2.4 GHz is actually kind of an ideal frequency for coking food because it's very efficient at inducing rotation in water molecules inside a microwave oven. You could certainly use higher or lower frequencies to do the same thing, but it wouldn't be as effective. And the further you get away from 2.4 in either direction, the worse the results would be.
Re: Oversimplification? (Score:2)
Re: Oversimplification? (Score:2)
You have that completely backwards; the reason that the 2.4 GHz band is unregulated is because, when the ISM bands were being established, the Americans asked for it to be included in order to accommodate their newfangled microwave oven technology.
Re: (Score:2)
Usually, for normal values of 10, the micro is 1000x smaller than milli.....marketing at work.
The customary, unscientific term "microwaves" specifies wavelengths in 1m...1um range, so it's particularly imprecise (source: wikipedia). In contrast the mmWaves is used in reference to wavelengths around ~1mm.
Re: (Score:2)
You fail in your use of wikipedia. The definition of microwaves are very precise and what I said, even in wikipedia:
https://en.wikipedia.org/wiki/... [wikipedia.org]
You are ignorant of electrical engineering, that's fine, most people are.
Except for 2G evrey single G is hotter and (Score:5, Interesting)
Consumes more battery than the G before it.
Is the law of the land, and the laws of physics too.
If you are signaling faster, you are switching transistors on and off faster, which consumes electricity and gets hot.
If you are connecting to more than one frequency band (or more than one cell site in the same band) at the same time, you are turning on multiple radios, which consumes more electricity and gets hot...
If you are transmitting more data to do the same things as before (say, HD voice, SMS over packet data, or more signanling) , you are using more electricity and generating more heat.
Trust me, I was working in telecom in my country in the shift from 1G to 2G and from 2G to 3G, and was closely involved from 3G to 4G. 5G is no different at all.
Over time, the kinks will be ironed out, and hot hadsets will be adistant memory, just in time to complain that 6G handsets get hot, and the battery lasts less, just as many so called reporters, punduits and bloggers just forgot that this happens every time you go from one G to the next
Re: (Score:1)
why? (Score:1)
Who the fuck is pushing this crap?
It feels like wankery.
I hear the wavelength is so small that transmit power needs to be increased anyway to get the same range, sounds wasteful and stupid.
And now they're saying the chipsets are wasteful and make loads of heat?
Send this back to the lab for another 10 years and then it might be necessary.
This stinks of progress for the sake of progress. With this and t
Re: (Score:2)
No, the GP is right. LTE's maximum throughput is around 700Mbits/sec. The base tier of my two different residential broadband options is 100Mbits/sec. One offers 940Mbit/sec in certain areas, but the other maxes out at 400Mbit/sec download speeds. Essentially, at least where I live, LTE can already outstrip residential, wired internet. On top of that, how many of the 5G towers are getting 100Gbit/sec on the backhaul? If they are getting merely 40Gbit/sec, then more than 50 people on a tower and you're back
Re: (Score:2)
Why would anyone need those horseless carriages?
GET OFF MY LAWN
The key phrase was "good enough". Plenty of better technologies have failed because the currently dominant one is "good enough". 5G has major costs connected with it and very little to offer 99% of users. The other 1% may, or may not, prove to be enough, but for now I personally couldn't care less about 5G.
Re: (Score:2)
Users don't need 5G, especially in the USA. If they had 5G's maximum theoretical data rates they'd only burn through their data allotment in moments. But if we're going to add a jillion IoT devices all phoning home via cellular modems, we'll need 5G to accommodate them. Plus, autonomous vehicles are all going to need always-on internet connections to share data back and forth.
How do we prevent 5G mast from being installed? (Score:1)
Re: (Score:2)
Ruh roh! (Score:2)
Me thinks the chips are not being cooled properly. They need to add a heat sink or some other cooling system and that means
*NO MORE PAPER THIN PHONES*
The companies are in a real pickle here: Make phones a reasonable thickness or have them super thin and fragile because they've been conditioning the public into thinking they want thin phones.
All signs are pointing away from heavy adoption. (Score:1)
Summary correction (Score:2)
No, microwave ovens operate in the 2.4GHz ISM band, same as Wi-Fi and Bluetooth, not the 24GHz band used for NR FR2.
mmWave != microWaves (Score:2)
Your oven works at 2.45Ghz (nominally), these are microWaves (but actually UHF). Some 4G can be found in the 3-6GHz area - these are still microwaves (but we have now strayed into SHF). 5G is currently slated to use 24Ghz and then higher frequencies, but 24 Ghz is really ~1.25cm, we don't really get into millimeters or mmWaves until beyond 30Ghz (the start of EHF).
But don't let any facts get in the way of a marketing slogan or trademark!
This article is nonsense (from a technical viewpoi (Score:2)
It's new so it gets hot, this is the same pattern we find again and again. Why it gets hot is likely not the reason the author suggests.
In any case, IMO, this is a technology update that probably benefits carriers more than consumers for now.
The average consumer already hits reasonable speeds on 4G/LTE (at typical densities, but maybe not crowded-stadium densities) -- anecdotally, I often find my LTE speeds beat any public space WiFi options I may have.
Speed gains help to future proof, but with a 2-4 year l
So it's like the '77 Monte Carlo I used to drive (Score:3)
It's overhyped, it's ugly and it overheats when traffic gets heavy.
Don't need it (Score:2)
mmWave is not close to microwave ovens (Score:2)
Most microwave ovens operate in the 2.4 GHz band, the same one that is used for WiFi and other unlicensed devices. (That's why running your microwave oven can interfere with WiFi.) That is right in the neighborhood of some of the bands used by cell phones, which have spectrum at 2.1 and 2.6 GHz. The reason the cell phone (probably) doesn't cook your head is the much lower power level used.
Millimeter wave is technically frequencies from 30 GHz to 300 GHz. All of the ultra high speed 5G service that is curren