How is Chipageddon affecting you?

[KE5FX]

Apparently the neon thing is a problem because the lasers used in fabs require extreme purity, something like six or seven nines.  The story I read was quite a yarn: the Soviets needed high-purity neon for laser weapons projects back in the Star Wars era, so they developed the necessary process.  Nobody else needed it, so nobody else built the facilities needed to produce it... and now, decades later, nobody has given it much thought.

Needless to say, don't take that story to the bank, but I thought it was an interesting and halfway-plausible explanation.

[tszaboo]

Me searching for high voltage LDOs. I also need negative LDOs.

[tom66]

Pre-chip shortage finding precision negative LDOs was a bit like hen's teeth.  You could find them, but far less selection than you would have liked.

For a few designs I just rolled my own using a spare op-amp on the board.

[Ed.Kloonk]

Pre-chip shortage finding precision negative LDOs was a bit like hen's teeth.  You could find them, but far less selection than you would have liked.

For a few designs I just rolled my own using a spare op-amp on the board.

This is where blokes like you and TszaNand will separate the sheep from the goats.

When the rubber hits the road, a lot of spoilt kids are going to left wanting.

[peter-h]

The problem with "renewables" is that they are not 24/7, so you need gas power stations to make up the fluctuating generation. The more wind you install, the more gas you need to install. Nuclear is a good solution - the very best we have - but not if you also have a lot of wind generation. And nuclear is not (yet) politically achievable in Germany... well, give it a few more weeks, the landscape on the mainland is changing fast

For sure all those gases are easy to make anywhere. Cryogenic distillation. Most stuff RT puts out is BS, at the best of times. Been watching it here for years; it has a certain entertainment value

On the topic, I am sure a lot of lead times quoted are pure BS. I am looking at 10k of a certain HP optocoupler; dely time quoted Aug 2022, which is OK. But nobody can tell if it will turn up. Lead times quoted on ST CPUs are BS; I have a few hundred 32F4s on order from a well known UK disti and the lead time turned out to be pure BS.

[JPortici]

Pre-chip shortage finding precision negative LDOs was a bit like hen's teeth.  You could find them, but far less selection than you would have liked.

For a few designs I just rolled my own using a spare op-amp on the board.

This is where blokes like you and TszaNand will separate the sheep from the goats.

When the rubber hits the road, a lot of spoilt kids are going to left wanting.

yesterday afternoon i had to redesign a board on the fly because the main boost controller disappeared. i admit was tempted to change the main role for the MCU to be digital power

[coppice]

On the topic, I am sure a lot of lead times quoted are pure BS.

Lead times are always BS. Like a salesman telling you the thing they are pushing is "off the shelf", you order and you get 10 weeks delivery. I guess the salesman just forgot to specify which planet that shelf was on.

[peter-h]

Right now that may be but generally (I've been in design and mfg business since 1978 so have "seen a fair bit") you could place a PO for say 10k, 50% in 3 wks' time and 50% on some date next year, specify dates with "not before", and the stuff would turn up. 99% of the time.

Now there is a lot of pisstaking.

In the meantime the fly by night crooks, who bought stocks early, are minting it. Just had a quote for a MAX3089CSD at £13 I was tempted to reply to it accordingly but I am not going to waste the electrons. I hope these bastards get hit hard when this collapses.

[NiHaoMike]

The problem with "renewables" is that they are not 24/7, so you need gas power stations to make up the fluctuating generation.

Not if you adapt demand to supply. Then the 24/7 generation only needs to cover the loads that cannot be rescheduled.

[IDEngineer]

If there's demand, supply will eventually rise to satisfy it. Even if people simply run private generators on gasoline. This is the same basic argument as "We must lower our standard of living" but consumers will not put up with it, and business people will not leave such an opportunity unharvested.

[coppice]

If there's demand, supply will eventually rise to satisfy it. Even if people simply run private generators on gasoline. This is the same basic argument as "We must lower our standard of living" but consumers will not put up with it, and business people will not leave such an opportunity unharvested.

You forgot the magic ingredient - oppression.

[Kasper]

Some people run their gas generators inside their home and then die and then demand from them at least is reduced.  I looked into this once and found 1/4 of the deaths from a hurricane were attributed to loss of power, mainly from running generators indoors and house fires from candles etc.

I think wars and shortages are good examples of why centralized power generation is not ideal.  Lots of small generators that don't rely too heavily on supply chain seems safer. 

When people trash alt energy's fluctuating supply but omit the existence of the largest batteries on earth, hydro dams, it seems like they aren't giving alt energy a fair chance.  All-though times of war highlight one of the downsides to dams (and nuke plants): their vulnerability.

Most consumers aren't going to change their demand habits just to feel good about themselves but once time of day billing comes in, in response to the massive increase in electric cars and solar panels then some consumers will change.

Some demand is already becoming flexible.  HVAC, water tower pumps and other things that operate periodically to keep their system within its limits but they have enough flexibility to run when the sun is shinning or the wind is blowing and turn off quickly when a cloud blows by and causes a sudden drop is solar supply.  I had a prof who came out of retirement to start a company and get a PhD on this subject of flexible demand.  He made it sound like the sky is falling (re the duck curve) and flexible demand is one of the solutions.  Supposedly some hotels in Hawaii were encouraged to allow control of their HVAC so it can be turned off when there is a sudden dip in supply.  Their other option was to lose all their power instead of just HVAC power.

[Bud]

No need to go to Hawaii. Same thing was promoted here in Ontario some time back, they were giving you a free thermostat in exchange of you allowing the mother ship to control the air conditioner in your house

[IDEngineer]

Oppression just means the supply will be via a black market, and thus crime will rise and tax revenues will fall. History is full of examples proving that "bans" don't work. They sound emotionally satisfying but only yield the negative results noted above. See "The War on Drugs", "The War on Poverty", US Prohibition, etc.

[Kasper]

No need to go to Hawaii. Same thing was promoted here in Ontario some time back, they were giving you a free thermostat in exchange of you allowing the mother ship to control the air conditioner in your house

A free thermostat, that doesn't work sometimes   I think Quebec did a similar thing with water heaters.

[IDEngineer]

...they were giving you a free thermostat in exchange of you allowing the mother ship to control the air conditioner in your house... A free thermostat, that doesn't work sometimes

I would think the crowd here would be uniquely empowered to, ahem, "fix" that problem. In a manner not visible upstream. For research purposes only, of course.

[Kasper]

...they were giving you a free thermostat in exchange of you allowing the mother ship to control the air conditioner in your house... A free thermostat, that doesn't work sometimes

I would think the crowd here would be uniquely empowered to, ahem, "fix" that problem. In a manner not visible upstream. For research purposes only, of course.

Upstream visibility might be a problem thanks to smart meters.

[IDEngineer]

I thought of that. Remember, they don't know what loads are active. Only that the HVAC isn't active because they remotely turned it off, right?    Could be anything... water heater, pool pump....

[coppice]

I thought of that. Remember, they don't know what loads are active. Only that the HVAC isn't active because they remotely turned it off, right?    Could be anything... water heater, pool pump....

There are load disambiguation systems that try to figure out which loads are turning on and off, and how much they consume. They don't work nearly as well as the developers would like to claim, but you might be surprised how well they do work. For many of the things they are being developed for they need to be almost 100% reliable, which they aren't. However, to look for usage patterns, rather than fine details, they can do a pretty good job.

[Kasper]

I thought of that. Remember, they don't know what loads are active. Only that the HVAC isn't active because they remotely turned it off, right?    Could be anything... water heater, pool pump....

There are load disambiguation systems that try to figure out which loads are turning on and off, and how much they consume. They don't work nearly as well as the developers would like to claim, but you might be surprised how well they do work. For many of the things they are being developed for they need to be almost 100% reliable, which they aren't. However, to look for usage patterns, rather than fine details, they can do a pretty good job.

Since they know when the a/c should be turning on or off, I think it'd make this easier than the usual load disambiguation systems. They might already include this as error checking for their system: look for incr load when they try to turn a/c on and a decr when they try to turn it off.

[coppice]

I thought of that. Remember, they don't know what loads are active. Only that the HVAC isn't active because they remotely turned it off, right?    Could be anything... water heater, pool pump....

There are load disambiguation systems that try to figure out which loads are turning on and off, and how much they consume. They don't work nearly as well as the developers would like to claim, but you might be surprised how well they do work. For many of the things they are being developed for they need to be almost 100% reliable, which they aren't. However, to look for usage patterns, rather than fine details, they can do a pretty good job.

Since they know when the a/c should be turning on or off, I think it'd make this easier than the usual load disambiguation systems. They might already include this as error checking for their system: look for incr load when they try to turn a/c on and a decr when they try to turn it off.

You should be able to game that. Turn off the load when commanded, but turn it back on some random time later. Never turn it on immediately, when commanded to. It should be possible to fuzz things up enough to confuse something following the timing.

[Kasper]

I thought of that. Remember, they don't know what loads are active. Only that the HVAC isn't active because they remotely turned it off, right?    Could be anything... water heater, pool pump....

There are load disambiguation systems that try to figure out which loads are turning on and off, and how much they consume. They don't work nearly as well as the developers would like to claim, but you might be surprised how well they do work. For many of the things they are being developed for they need to be almost 100% reliable, which they aren't. However, to look for usage patterns, rather than fine details, they can do a pretty good job.

Since they know when the a/c should be turning on or off, I think it'd make this easier than the usual load disambiguation systems. They might already include this as error checking for their system: look for incr load when they try to turn a/c on and a decr when they try to turn it off.

You should be able to game that. Turn off the load when commanded, but turn it back on some random time later. Never turn it on immediately, when commanded to. It should be possible to fuzz things up enough to confuse something following the timing.

Ok, now we're onto something!  If only we could get the components, we could make a business out of this!   

[IDEngineer]

EDIT: The key would be to show changes in load when the spy device requested them. That way it appears to the outside world that things are behaving as expected. Essentially, after each "external command" there would be a semi-random period of time during which that external command would take precedence. (Again, avoiding patterns which could be discerned from the outside.)

But after that period elapses, priority would revert back to internal commands. In this mode the load could be switched on, or off, based on internal preferences. This too would add a degree of randomness since it wouldn't always be true that your actual wishes were opposite the external commands.

There are other large loads in most households. Ovens, if electric, are several kilowatts. So are electric water heaters. You might have supplemental heaters or A/C units. Large pumps for pools, hot tubs, etc. have big motors just like A/C compressors. So nothing definitive could be learned even by analyzing the power factor, because large motors are used in lots of things besides A/C compressors.

Superior (command of) technology for the win!

[tom66]

The problem with "renewables" is that they are not 24/7, so you need gas power stations to make up the fluctuating generation. The more wind you install, the more gas you need to install. Nuclear is a good solution - the very best we have - but not if you also have a lot of wind generation. And nuclear is not (yet) politically achievable in Germany... well, give it a few more weeks, the landscape on the mainland is changing fast

For an engineering forum this is a place full of misconceptions and pseudoscience.  Maybe 10 years ago, you would have been correct.

However now we can build models on a certain number of wind turbines, combined with a certain amount of storage, and in the southern parts of the country combining this with solar PV, and test such a system in various scenarios that shows that we will never have a power outage in 300 years based on known and changing weather patterns.  No (natural) gas required, no nuclear required.

One of the best technologies (in my opinion) is to generate hydrogen from electricity when there is an excess of renewable supply.  This hydrogen can be used for domestic heating, it could be used to power trains and trucks and industrial processes, or it could be converted back into electricity when there is a shortfall.   Short-term (1-8 hours) storage could be battery-based, combined with demand shifting things like EVs into charging overnight.

You don't need nuclear;  that's not to say it's a bad technology.  Nuclear is expensive, so except for a few locations (inland areas without access to much wind, for instance) it isn't really that sensible of a choice any more.  This is why ~2-3GW of new wind is being installed every year, not nuclear.  I think Hinckley Point C will end up being seen as a massive white elephant, especially given the cost of the build out.

(Edited to correct error about wind turbine power.)

[peter-h]

Ah, storage, yes...

And hydrogen

It's gonna be fusion next.