-rick-

This is a known problem with existing solar/battery/etc installs. The electrical regs were relatively conservative, preferring the new generation to cut out early and delay reconnection. Since new generation got bigger these limits started becoming problematic. There's more noise on the line than 0.2v, also you might lose several volts on the connection from the grid to the inverter in local cabling. Having a window of 2-3v might help. (Each inverter choosing a value at random within that range every 24 hrs or similar). Yeh, that's the long term solution, making new energy inverters be grid forming not grid following. I believe this is already happening to a degree, though maybe this will never happen in residential solar, only commercial scale for the reasons of making sure grid workers aren't exposed to unexpectedly live cables. (Having said that they work on live cables an awful lot as it is so maybe worth evaluating exactly where the line should be drawn). It's a key signalling mechanism that is used by modern inverters too, so frequency modification is not going to go away. Victron inverters can vary the frequency of their output to control 3rd party inverters connected downstream. (I think this only works when they are working in a UPS mode, ie, not connected to mains).

Not watched it, but just skimmed the wiki page. Backing away now ............ 🫥

Cost of that Mac will pay for 5 years of Codex pro which I'm hearing is much much more efficient than Claude Opus 4.6 so could get an awful lot done with it plus can always move subscription if something better comes along. As someone who has traditionally loved playing with hardware seeing what Alex Zsikind* and others are doing is making me envious. But then I think 'What would I actually do with any of this stuff?' and the answer is just play around and not much more.

Mac Studio M5 Ultra 512G when it's announced? That's a mighty big hole in your pocket!

Sure that's been the problem for ages and what is trying to be addressed AFAIK.

The excuse before we left the EU was that it was EU rules preventing us from doing anything. Now we've left we could have done whatever, but the government/officials were all too busy dealing with brexit. We'd be further forward with addressing this issue if we'd stayed in the EU.

The big brands offering things like this cost a more than that. If the regs say that to legally sell this product it 'MUST' ensure that the pins are not live unless a mains signal has been seen within the last 20ms then either the product complies and is safe (ie, wont kill anyone even if they touch live pins) or it doesn't comply and is therefore defective. That should be the end of it. Really not sure what you are getting at here. The number of things that have to happen to produce mains output from a solar panel these days is huge. The computers involved in generating the waveforms are far far more powerful than desktop computers from the early/mid 2000s. It's not a case that the waveform will be produced unless blocked by a safety system, it's more that the system can't produce a waveform unless a long list of conditions are met. This is technical but heres a reference design from TI for a microinverter. Not suggesting you take anything from it other than it's not a simple system that will just continue working in a fault condition (the document doesn't touch on the safety side unfortunately). https://www.ti.com/lit/ug/tiduf63a/tiduf63a.pdf?ts=1774517354380 The MCU (brain) they use to do the control is this : https://www.ti.com/lit/ds/symlink/tms320f280039c.pdf?ts=1774508361987 Again not really an accessible document, but you'll note that it supports various functional safety standards. Those are some pretty rigorous standards and if you design a product to meet them then you can make guarantees about the behavior. ie, make it fail safe. Edit: I'm not saying this is currently part of the regs for this, seems a little OTT for the situation, but it's certainly something that could be added to regs if a need is identified.

Sure, but that is a separate issue. We as a country need to start actually fixing the issues rather than trying to work around them by banning things outright. (Which still doesn't solve the problem because people will still get and use them anyway). Other countries manage this much better than we do. I haven't read the regs but I very much doubt that the regs allow for a plug to become live without an active current for anything more than a fraction of a second (and only then if there was a signal immediately prior that disappeared - eg power cut or socket pulled from wall). Either case I don't see how you could be in that situation while at the same time holding the plug. Others have raised the possibility of two systems plugged in sensing the other as 'mains' and hence continuing to work. I don't know how possible this is for systems designed with the current regs but it's certainly something that I think can be designed out and even in this situation the live pins will be actively buried in the socket, not in human contact. Electronic inverters can assess the mains waveforms thousands of times a second and with each sample detect a wide range of issues so any of these safety issues should be addressable if they havent already been (I suspect they already have been). Edit to add: A standard RCD cuts the current within 30ms of a leak to earth being detected. This is deemed sufficient to save someones life who touches live conductors. An inverter should be able to switch off much quicker than this if it detects any unusual conditions. It would be relatively unimpactful to efficiency to set the cut off conditions much tighter than RCDs as a cut off for a couple of cycles would not be noticed by anyone using the system (given the system should only ever operate when the mains is live).

I agree. Though again, I don't think we should be making decisions about what is allowed based on people doing things that are explicitly not allowed. UK based sellers are required to ensure the products they sell meet UK standards*. If you buy an electrical product from a UK seller that doesn't meet regs they are liable. We shouldn't set regs based on the possibility of gray/black market imports. As I've said I think the requirements we have already discussed are sufficient to avoid serious harm. Compliant products are unlikely to cause deaths. These products have been available and legal in Germany for a while so we should have good data to see if there are any gaps that need adjusting in regs for here (though we will tend to follow EU regs anyway) and so long as decision makers are taking this evidence into account and setting standards based on them then I doubt they would have liability either. * If you import yourself then you are liable to check compliance.

Those outlet covers are cute. What are they?

If that's true then anyone who installs an oven/hob and then someone else manages to cause a deadly fire using it could also be held liable. Thats absurd. If you buy a dodgy uncertified product and install that then I can see there being a liability. But if you install a UKCA/CE marked product and do so in accordance to the manufacturers instructions then you should be ok. I'm not saying don't regulate this stuff. I'm saying these devices are safe enough with the restrictions as discussed (800W limit, automatic shutoff if no mains detected, no more than one unit installed). Trying to say 'well someone might install 2 therefore we can't allow anyone to install any' is ridiculous. Edit: I suspect the 800W limit is set as such as it is so far below the limit on most domestic circuits that even if the instructions are violated and 4 units are installed you are still going to be within the safe limits of any circuit.

We have to draw a line somewhere. At some point the attempts to make things safer just make things impossible, unaffordable or impractical. Germany now has 1 million on these plug in devices, I expect thats enough to generate some fairly robust safety data to see how dangerous plug in solar actually is. I suspect not very much at all. No, and we have to draw a line somewhere, but in this case my strong suspicion is that the products are safe enough. Sure, the question is how likely is it to happen? Can you address the bulk of risks through safety/product standards and public information. And that increases the cost 10x. Are the safety risks of plug in solar (when limited to 800w) sufficient to justify that? I argue not. Edit to add: I say 10x because as soon as you start getting installers involved whats the point of a tiny system? You end up with a bigger one and before you know it you are at 10x cost. The whole point of these systems is that you can buy them as a renter. Put them in for a short time and then take them with you when you move. This doesn't work if you need to get the landlord + electricitians involved.

This isn't a fault and can happen even with apple original chargers. Happens with mine. It's caused by the noise suppression capacitor in the power supply. The supply is fully isolated from mains voltage except for that capacitor. Because the power supply runs at high frequency a limited amount of current can pass through that capacitor and that can lead to the tingling feeling. It's a very small amount of current and considered safe. (And the capacitors are special safety rated ones which means they are tested to fail in a safe way).

Where do you draw the line though? Just because someone CAN harm themselves (or others) with a product doesn't mean we shouldn't offer it for sale. Mandate clear instructions, mandate regular reminders on bills and questions when people change energy suppliers. Make sure information is regularly shared publicly but don't say 'oh someone might do something stupid with this so we won't offer it for sale'. The products will still be available, idiots will still buy them and the safety of those products will be less than if you had sensible rules around them rather than saying no.

It's a bit more complicated than that. They should warm up at the same rate. Even in a somewhat unbalanced system the rads might eventually get to the same temp but if they take a long time to do it then over that time they have put out a lot less heat. You say that with confidence but I'm not so confident. I'm not sure your most recent installer would have done balancing. It may have taken several additional hours once the install of the heatpump was complete and involved him going round your rooms repeatedly. Did that happen? If he balanced the system then I would have expected him to bleed the radiators also. (Bleeding the radiators is step one of balancing the system).

Also, the overbuild of supply from China is likely mostly resolved now and the Chinese government have stepped in to regulate the market there to limit overproduction. The £50 a panel price we saw a year or two ago was because a significant number of factories all came online at the same time and the market couldn't absorb the supply.

@zoothorn Will reply to your PM in a bit, been a bit busy last few days. Firstly, the thermometer is useful for other things than just checking the radiators. Once you have it I would like to you measure the temp of various things in your house, such as the different walls. However, the reason for suggesting it for the heating is two fold. 1. It's a way of working out the approximate flow temperature without having to know how the Valliant menus work. That's something you've had some issues with and I don't have any experience with Valliant menus specifically so can't easily help with what steps to take. 2. It is not correct that the radiators will always be the same temperature if the incoming water temperature is the same. Water will take the easiest route, so if you have two radiators connected in parallel, but one has 10m of pipe between it at the heatpump and the other has 1m of pipe then a large portion of the water will flow through the shorter pipe starving the more distant radiator of heat. The way this problem is addressed is called balancing. This is done by adjusting the locksheild valve that is hidden on the caps at one end of the radiators. This restricts the flow of water to radiators so you can reduce the amount of water the nearest radiators get. This forces more water through to the more distant radiators. Do not attempt to adjust these valves right now. I would hope that your system is already balanced, getting the thermometer will help confirm that. If the temps vary significantly from one radiator to another then it's something that needs looking at.

When I looked, the window/door companies seemed to offer this as an option on their products. I'm sure vastly overpriced for what is a magnet and a reed switch (cost <50p). With planning I expect you could plan to retrofit something to windows doors just after they are installed but you would be putting holes in the frame and not sure how that affects warranties, etc. (though a common window install method holes in the frame as well so can't see it being a huge deal). Not got far enough to think about this in more detail.

I should add, on the fire/smoke issue, we do have one in the building I'm a director of the RTM (separate from the BMS I mentioned above). It's entirely seperate from everything else. Designed to be manually controlled by breakglass (though our recent FRA suggested it might need to be wired into the fire alarm now). Quick look at our budget at it looks like it's costing us about 2k a year for firealarm/smoke ventilation system maintenance, for a 6 story/27 flat block (below 18m where extra regs kick in).

Doesn't sound crazy. (I'm not expert, I've been involved in precisely 1 BMS system and was mostly hands off and not impressed with the final result). The above can all be done with relays (and a timer) so don't necessarily need a computerised control. Worst case a PLC with the ladder logic printed out and stored in the cabinet so it can be replaced if it breaks. Having said that if opening windows if the fire alarm goes off is a requirement for fire safety you may well get into a much more regulated area with the need for certified systems, battery backups, regular maintenance, etc. The other points seem trivial to do. I would stay away from anything more heavily computerised if you can. You'll just be paying an endless series of extra bills. Every bit of equipment you buy will likely cost more up front and then you'll be paying specialists to configure and maintain them, etc. I'm about at my limit of useful contributions here but even if it doesn't help me say more it may help others if you could say more about what the requirements/specifications are for this building.

Only looking at the proposed panel, I see nothing on there that necessarily requires a BMS system with complicated setup. Just hard wires into the various systems. Does he want this wired up to a BMS controller that then has interfaces with all these systems or is he just proposing a panel where you can see the status of everything at a glance. Doesn't look like there is anything complex going on, so keeping it simple with some hard wired indicators and switches and no computers involved seems like the obvious answer. (If you need a simple PLC for sequencing relays then sure but can't see why you need a full BMS which will a likely 5 digit number to your costs).

As much as dealing with the developer is on this, I feel its probably worth persuing at least a bit more. Maybe worth considering getting an independent report of what needs to be done before approaching them though, that way you have a defined list of issues and wont be fobbed off with inadequate proposed fixes. I know this doesn't affect you personally but the more people complain about these issues (and cost developers in fixing the issues), the more likely developers are to take more care in future.

I've seen someone do it by making the jig stationary and pushing the xps through it. Like a tablesaw/bandsaw. Vaguely remember that the standard wire was too flimsy to do at any speed but with a bit of juice and a thicker wire was able to get through it quite quickly. Thinking more, I have very vague recollections of someone using a welder as the power source (not sure it's the same one). In which case this is maybe not the easiest route to go down.

Sounds like you should be able to get this fixed? Some ISPs are terrible for this but others will badger BT until the get an engineer out to move you to a different copper pair. (The reasons I've had in the past for performance like this is corrosion/moisture on the line).

@flanagaj If you are going ahead without experienced hands on-site, remember that concrete can burn, so don't get it on your skin and if you do wash it off PDQ. Would be sensible to have washing area ready and spare clothes just in case. I've seen others talk about how they let themselves get too focused on getting the concrete in and ended up getting badly hurt. Even the worst pour can be fixed later, burns can do permanent damage.