Jump to content

RobLe

Members
  • Posts

    56
  • Joined

  • Last visited

Personal Information

  • Location
    Cambridge

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

RobLe's Achievements

Member

Member (3/5)

14

Reputation

  1. How about heated chair covers? 100W at 12V into a strap on seat cover, intended as a retrofit for a car, makes a huge difference. Uses a lot less power than a radiant thing heating the world.
  2. I think with the symmetric scheme where each batt has a CT with 2 wires through it in order to negate the other battery current flow, and a 2kW kettle turns on, then: Batt1 will provide 2kW Batt2 will provide 2kW So yes, the kettle will be powered by the batteries. But then the main tail will export 2kW, surely not wanted?
  3. I think the mutual nulling you suggested will: Make PW supply the same power the house uses, and at the same time Make SE supply the same power the house uses That's too much power, that's why I don't like it! I think it will provide 2x the required power. The cables are unlikely to fit through the CTs, as an aside. I would love any/all schemes to be tried though, out of interest!
  4. Why do you suggest to do this? I'm with OVO, they pay 3% interest on any balance, so why not stay in credit? We've been with them for 10 years, the DD bounces up and down as their "algorithm" tries to figure out what we will use over the year, I just let them "manage" it.
  5. @Radian: Agree syncing charging is part of the issue, but it alone wouldn't be terrible. The dominant thing is an underconstrained system, where it is completely valid for PW to charge SE and the other way around. Hence the horrific up/down spikes in power in the top graph above from the op.
  6. If the PW and SE are wired in series, as (d) above, I think that they both will work. When the sun shines, PW will gobble it up assuming PW=B1 in (d). It's not a voltage thing, it's just where it's CT is positioned that matters. When the kettle is on, the PW will provide. When there is too much sun, the PW will be full and stop taking power.... then the SE will fill up. If the PW were to fully empty, then the SE would get a chance to power something. It might not be quite as above, as the PW can also be controlled by Tesla(?) - I think Tesla might tell it to power the grid sometimes and pay you for it - eg via Octopus "Tesla Tariff". This could screw up if so, as SE will immediately thwart that behaviour, absorbing that power. You might be better off with SE on the inner loop, so PW can discharge into the grid if you want that to actually work(don't bother if you don't want this!). You'll also need to make sure any night battery charging times are synchronised between PW and SE. Eg if PW charges in the middle of the night from the grid, but SE isn't programmed to do that, then SE will charge PW, which is inefficient. By and large though I think d) is good..
  7. It's not my house, so I don't know for sure how it's installed... a piccy of the CTs would help - I think they will be on the same wire, and I would be willing to bet money that it is actually installed like (e) below. That is, each batt connects to the house consumer unit or a henley block next to the consumer unit. Then each batt CT connects to the tail (the order of the CTs here makes no difference). And wiring it like this is the cause of the issue, even though each battery may be individually wired as it's install manual - I'm guessing they haven't considered anybody would be ambitious enough to have 2 batteries. In contrast, I think that (d) will fix the issue completely. I'm not changing my mind, it's the same schematic as (c) I did a while back, and also the blurb I wrote some time back. It is clear to me, although I'm having trouble convincing the internet!
  8. You don't need to worry about the voltages - the PW and SE will do all of that complicated stuff with their power electronics and control algorithms. All you need to know is that the PW is trying to minimise the power flow through its CT; it will put out whatever current at whatever voltage is required to achieve this. Imagine that you sublet a room in your house, where you had a single power feed in, and there was a PW battery, and a kettle which somebody turned on and off, maybe even a bit of PV. The PW battery would effectively power the kettle. You might not even know that they had done this, inside their sublet room. Then imagine you fitted an SE battery to your whole house - this would work too, but nothing it does will affect the PW operation.
  9. I'm going to have another go at convincing people πŸ™‚ The pic below is in 3 parts, showing a progression of the idea in my mind, and without pesky equations: a) is as the manufacturer intends, batt1 having a CT on the tail before it, so that it can identify the power flow and try to make it zero on average. It's doing a lot of maths and clever power electronics to achieve this, but we don't need to know about that. A bit of power sneaks in and out on transients, but otherwise it does a good job. b) is just like (a), but I've put a dashed box around the house+batt, and now we are just going to ignore what's inside - it all just works in isolation - any interactions are inside that box, and ultimately just give or take power from the grid. Importantly, nothing from outside the box affects the inside (neglecting stuff like the grid is broken etc). c) now we apply another battery, but you can see that box from (b) replaces what was just "house" in (a). So batt2 should work, just as well as batt1 in (a). As described before, the 2nd batt won't do much work - the first will preferentially absorb PV and make the tea, 2nd batt only operating when the 1st for some reason cannot. πŸ™‚
  10. That's a very interesting arrangement, but I'm not sure that will do what you want. As an aside, you may find it difficult to fit 2 thick cables through the clamps. Anyway - considering the power into B1, B2, and House - I think the schematic above, with knowledge that B1 is trying to null CT1, will create 2 equations: B1=-House B2=-House Thus the overall power from the Live Tail will be the sum of B1, B2, and House: B1+B2+House = -House. The units will each independently compensate the house, but that's unlikely to be what you want, as you've gone too far and are still interacting with the grid. In contrast, the version I intended a while back gives equations: B1=-House B2=-House - B1 Thus the overall power from the Live Tail will be: B1+B2+House = -House-House-B1+House = -House-House+House+House = 0 No current at the tail, just what we want I think πŸ™‚ Honest.
  11. So both PW and SE are home batts, and there's no useful software settings on them to prevent interaction? I think the issue is that the combined system is underconstrained. PW presumably has a current clamp on the main Live tail, and is attempting to make the current there zero, using a software feedback loop with likely a PI(proportional + Integral) controller. SE is presumably the same. While you might think they would both sit there doing a similar thing - slowly feeding on PV say, an equally valid solution is for one to discharge into the other, while still overall feeding on PV. I think if the PW clamp were after PW+House+PV, then SE clamp after PW+House+PV+SE that this stops the PW responding to the SE as it can't "see" that current flow. The SE can still respond to the PW - but the overall feedback loop is broken, so I think this might fix it πŸ™‚ One of the home batts will be exercised much more than the other doing this - the way around I described it, the PW will do all of the work until it cannot, being full or empty or having too high power flow, then the SE will mop up the rest. Disclaimer: I'm just a diy-er with ideas above my station πŸ™‚
  12. A friend of mine had I think a similar, but simpler problem. He had V2H and also a solic 200 diverter. With no sun, the solic should do nothing, and V2H supply the house. What actually happened is that the solic sometimes turned on and β€˜stole’ all the V2H energy, everything ramping up to full power. The solic and the V2H had slightly different ideas about what the power flow into the house was, causing the issue. We fixed it by bodging 10 thin turns of wire around the solic sensor in addition to the mains tail, with 100mA ac flowing in it from an (mains)ac-(selv)ac adapter, convincing the solic that there was 1A*240V less available power than reality! The above fix only works because the solic takes power only, and (obviously!) can’t deliver electrical power. Perhaps you could see if either the PW or SE(another home batt i assume) have adjustable dead band regions? That is, if one of the units allowed power flow in either direction of less than 100W or so, then the issue might resolve. The bodge we did will not work for you, if both units can have bidirectional powerflow; you need a software fix, or a more complex bodge.
  13. Ooooo maths! If it's a cube of insulation, each side of size X (metres), with a U-value called Y(W/m^2/degC), full of water DT(degC) above surroundings(it makes the maths easy, and you'll not do better than water): Area = 6.X^2 Losses = 6.X^2.Y.DT/1000 (kW) Volume = X^3 Energy storage = DT.X^3 (kWh) - just so happens that 1T of water stores 1kWh/degC πŸ™‚ So the time constant = energy / losses = DT.X^3 * 1000 / (6.X^2 . Y . DT) = 180 X / Y (in hours) Lets say Y = 0.1W/m^2/degC, so the time constant = 1800X. So if X is 2metres, using a U of 0.1, we can get to 3600hours, a time constant of around 5 months - just enough. If we have water in it. Sand won't be quite as good I think. Agree with last post .... it needs to be big!
  14. Great work! Could do with 2 of them to the same cpu - we have a main meter and a PV meter. I think we might need another meter again really, in series with the main one but backwards, to measure power flow both ways. Then we have V2G....Why is it so complicated?! Some elec meters have isolated outputs, as well as an led, that pulse. This one that I've used does, pulsing at 1 pulse per Wh - prob worth getting that sort if you're buying one. I used a 10k pullup resistor to 3.3V, connected -ve to cpu 0V, and fed the pulse signal into a cpu input: https://www.amazon.co.uk/Electric-Electricity-Backlight-Certified-Calibrated/dp/B096YBRRYK/ref=sr_1_7?crid=3VCCVWXIONQEG&keywords=electricity+meter+din+pulse&qid=1657462762&sprefix=electricity+meter+din+puls%2Caps%2C106&sr=8-7
  15. We have a vented DHW system, pic of similar attached. The tank is nicely insulated, it's even boxed in with poly beads down the back! It occurs to me though that the vent itself is a horrible heat leak mechanism, as it leaves the hottest part of the tank, then travels upwards into the cooler loft. Sure that pipe is insulated, but not as well as the tank, and that whole 2m length of that 22mm copper pipe thermosyphons away so is permanently at the maximum DHW temperature. Is there any reason why there can't be a thermosyphon break (london loop, short down then up segment)? I've looked in the regs, and they require that the vent be 19mm ID minimum, but there's nothing I can see about it having an upwards only slope. Every picture I have even seen on the internet shows the vent heading straight up, or horiz then up (hardly any better). It just seems a huge waste of heat to install that way. Thoughts ?
×
×
  • Create New...