-rick-

Feels like this is one of those new technology startup issues. Turbine size is still growing but I'd guess we are near the stabilisation point where its not worth going bigger. Once that happens then replacing like with like at end of life is likely something that can planned for. Not sure it's necessarily been something on the radar up to this point.

Congratulations on your first shower in the new bathroom!

They are, though I don't think this is really an argument that can be used against wind turbines or similar. The amount of concrete used in them (even if massively expanded) pales into insignificance compared to other big projects. Concrete + steel are pertty good in the end. Steel can be endlessly recycled pretty much. Using ground concrete as filler for other things is relatively good compared to many of the other things we do in terms of materials recycling. I believe ground concrete can be used in place of virgin sand/aggregate when making new concrete at least in a decent percentage of use cases. Concrete over its life absorbs a lot of the CO2 put out during manufacture and if we can switch from gas firing to renewable heat we lower the carbon footprint a decent chunk. A lot of more renewable building methods have overinflated/misreported environmental credentials once the full lifecycle is considered and once those things are considered concrete really doesn't seem that bad, especially given it's about the cheapest construction method. At a societal level spending more for a more environmental construction method may mean less money for more beneficial environmental spending.

My point is that they haven't replaced the cores on the existing fleet that is going to last to 50 years. They have had to do some very expensive repairs, some of which were particularly expensive because the design didn't allow for them. New designs shouldn't make the same mistake and either design out the need for those types of maintenance or plan for them. Well quite. That gets to the bottom of my issue. I don't particularly have an issue with nuclear power if it is competitive price wise. Given the ginormous construction costs the only way it can make sense is if the plants last a long time.

Yeh if it's a case that the financals work out for 30 years but the plant has been designed to last longer then fine. Someones planning for future profits, that's a different issue. If we get to 30 years in and they then turn around and say we need further massive investment if you want to continue for another 20 years then given the previous massive investment it's very poor planning.

Sure. I would expect that for a plant opening today that very few of the components would be original in 30 years time. But that to me is maintenance. If our existing fleet is running for 50 years, it just seems like we should design new plant to have the same life (with the expected maintenance). It's the parts that can't easily be swapped out that need to be designed to live the longest. And if we have managed to get parts that we built 40-50 years ago to live until today then we should be able to design the new parts to do the same. Not maintenance free for 50 years, but to last 50 years with planned and scheduled maintenance. Most importantly with a design that allows for all the maintenance tasks we have found necessary on the old plant to be possible and planned for on the new plant.

Prices at Bimble Solar don't seem much higher than last time I checked. Guess they have older stock but worth looking at if looking to buy.

Well quite, it's why I'd expect a new plant to be designed to last at least as long without the expensive life extension. Obviously maintenance, even fairly expensive maintenance, will be required to last that long. But if planned for upfront it's a lot cheaper than just building for 30 years and dealing with extension at a later date.

Yeh you clearly have the design right for the 'rip it out and replace with dumb wiring' option. But personally I want a solution to be that plus 'if the central hub fails, the light switches still work'. This is easy with sonoff style smart solutions (the switch modules can be programed to work like normal switches if the central hub is down). Little more difficult to find solutions with centralised dimmers. If installing a relay based solution then the european way of doing it works well (assuming you can get three way retractive switches and use latched relays). Switches on the wall are directly wired into the on/off coils of the relevant relays and in parallel you wire your smart system. Thus a switch on the wall directly triggers the relay but so can your smart system.

Honestly surprised by this. Given the cost of it (financial and environmental) I was thinking it was more like 50years. AFAIK, the last gen of nuclear is generally delivering 50 years ish of life.

*tongue in cheek* That's not the answer I expect from someone who decided there weren't any good heating installers so became one himself. Time to become a 'pv and heatpump' specialist

To add, having a fancy custom automated home is a nice idea but it is well worth taking a moment to step back and think: If I get injured in some way that temporarily or permenantly stops me being able to maintain the system is what I'm doing going to continue it's basic function? Can someone else understand enough about it to fix an issue? If you live with others, can they maintain/operate the system if you are temporarily unavailable? If you can't say yes to these then well worth taking a moment and thinking what changes you'd need to make to allow that to happen. In my mind, automation should be an add on for extra functionality and if it breaks the basics like light switches should still work.

Stop looking at aliexpress/ebay. Start looking at CPC/RS/Rapid/TLC-direct/CEF. More expensive but no worries about certification and going to be at least reasonable quality Exactly. Presume you are installing something you want to last and not require constant fiddling. In which case get something from known sources that your electrican is comfortable with. Even if you do want to constantly fiddle, I'd guess fiddling with the software/end result side of things rather than replacing iffy parts is where you want to spend your time. But you said you can design PCBs? Even if you buy one off the shelf. Options exist: https://paradar.co.uk/products/owl-driver-board-raspberry-pi-automotive-relay-hat?variant=55334464618879 (not keen on form factor) Things like this is another option: https://revolutionpi.com/en/ Look at latching relays. Requires more IO/drivers but no consumption during use. Still I will go back to my original point. If you are going to the effort of a custom system why not look at using off the shelf dimmer modules that you can control via the Pi? Plenty of options. Yes more costly than buying iffy stuff from AliExpress but a long term solution that offers way more flexibility than simple relays.

Have to say that DNS being the cause of this surprised me a bit, despite 'it's always DNS' being a thing that people put on t-shirts. Was expecting it to be more NAT/firewall stuff. In any case, i expect cloudflare will be faster than almost anything else. Because a huge chunk of webtraffic flow through them anyway, using their DNS can often speed things up vs slow it down. 9.9.9.9 is another service thats worth looking at (Google Quad9) Both cloudflare and Quad9 offer filtered 'safe' versions (using different addresses) that might be preferable to use if you have kids.

If youve gone to the effort of star wiring I'm surprised you aren't looking at dimmers rather than pure relays. Having said that, if you want to use relays have you looked at how the europeans do it? They just have extra rows in the CU and have DIN mounted relays. Given we are in the UK and our regs I wouldn't want to try and put this in the same CU. But using DIN rail mounted relays would be the safest choice for electrician sign off I would have thought. Plus points that each relay is in a socket so can be individually changed if they suffer early death. Pricer than just a PCB but still a professional type job. You can get raspberry Pis mounted in DIN rail cases too for the extra professional look. Edit: Raspberry Pi would need relay drivers if you do it this way, so either you get the DIN packaged versions that have this already or design your own HAT for it.

Just to eliminate the obvious when you've encountered this issue have you rebooted both the router and your phone/device?

Easier to clean around too.

Not an electrician but looks like the insulation is damaged and theres signs of heat damage so at the very least you need to cut back and remake the ends. I don't believe there is an issue continuing to use this older cable so long as the pvc is sound and flexible. If it's cracking when you move it then no.

Yeah, that for a lot of things relied on being one of a small number in the niche across the EU single market. We left that and those manufacturers suddenly have a much harder time competing with the other EU companies, while at the same time Chinese companies are also coming in to compete in the same areas. No, though I think there is opportunity with modern, low labour, manufacturing methods that we will likely mostly miss because other western countries are just more attractive. Theres still a plan to get HVDC setup to Morocco to get access to cheap solar exports though. https://xlinks.co/morocco-uk-power-project/ Africa is a long long way away from being able to compete there. Too many different countries, too much inter country strife. India is the next China. Not sure about Africa but with the more recent AMOC data seems quite likely this country will become much less plesant to live in within that timescale meaning that even if future governments do everything right we are still likely to be poorer and less populated than now. Hope that mainly bites after I'm dead though.

Not sure about that but I do know that PoE cabling has been added to the 4th amendment of the 18th edition (current) electrical regs. Which strikes me as an overreach, are data installers going to need to become qualified electricians now?

Careful where you might end up...

Everything I'm seeing suggests that the world has changed somewhat over the last 4 months or so, so if you haven't tried with codex 5.3/opus 4.6 then you might have the wrong mental model of where things are.

Anthropic just announced a big increase in number of tokens included in their subscriptions. OpenAIs codex is already generous (apparently). So maybe worth another look. Seen plenty of people talking about techniques for minimizing token usage. This is the thing that confuses me the most about your plan (not that you have told us much detail). Home Assistant can already do this can't it? (some features need a live cloud connection but some don't IIRC). Even if the operational side of your plan needs a local LLM, the coding/development aspect is one of the hardest aspects of LLM use that the frontier providers are better at. So your best value for money might be using them to do the development and then using a smaller, less memory hungry local llm to do the basic daily interactions. (Should be possible to design it so it can call out to a bigger subscription model for 'hard' problems). Having said all that, if you are enjoying the process, have the spare money for the hardware then do what makes you happy.

Congrats. Work smarter not harder! Always surprised when I see people manual handling relatively large amounts of stuff when there is a machine that can be easily hired/bought and can do the job in a fraction of the time. Sure, money is a factor but in so many cases using the machine works out cheaper even if you don't value your time/health.

Given the craziness around AI it wouldn't surprise me if someone bought this, especially if the new M5 models are delayed/limited availability.