Combi boiler and unvented hot water cylinder

Depends on the combi I expect -easy enough to add one if required.

I would just shut off the output of it instead rather than leave the plate heat exchanger empty / at low pressure.

It's an amusing exercise. It's like planning an installation for the "yard" at Dow Chemical.

Powerwall 2 Powerwall 3

Energy Capacity 13.5 kWh* 13.5 kWh

On-Grid Power 5 kW continuous 11.5 kW continuous

Backup Power 7 kW peak 11.5 kW continuous 106A LRA motor start 185 LRA motor start Seamless backup transition Seamless backup transition

Up to 10 units Up to 4 units

It would depend on how long you expect to draw this 20kW. When assembling this thing, I've tried to arrange the pack discharge time to 3 hours for the packs, to limit stress on the batteries. Since the load is non-inductive, the inverter might not blow up. Even though an inverter has a Locked Rotor Amps rating, inverters do not really "like" running inductive loads, or handling the back-EMF from such loads. You would not try to actually *use* the 185 LRA, maybe a single furnace air handler (2HP motor) would be as far as I would go. The Tesla inverter might be orders better than a similarly rated Chineseum inverter. When there is no way of knowing what you're getting, it adds to the "excitement" at the 20kW level.

(+)----(+)----(+)----+ | | | | PW3 PW3 PW3 Inverter ---> 11.5kW continuous | | | | 40kWh (-)----(-)----(-)----+ 3 hour run time (charge batt 80% not 100%)

(+)----(+)----(+)----+ | | | | PW3 PW3 PW3 Inverter ---> 11.5kW continuous | | | | 40kWh (-)----(-)----(-)----+ 3 hour run time (charge batt 80% not 100%)

And you're splitting the electrical load into two pieces. As far as I know, the PW inverter applies to the whole bank (of 3 in example), rather than each pack "load sharing" and increasing output radically.

Your solar array in this case, could store 80kWh * 0.8 = 64kWh (get 5000 charge cycles max)

And now, you're capturing a lot of your solar output. You could even export (some of it) :-)

You will likely get a much better quote locally. You could tighten up the project as you saw fit.

Paul

If you do that you're going to be limited by the output of your inverter. A typical domestic inverter is say 3-6kW, so you'd need several. A domestic battery may not be designed to produce enough current (20kW = 416A @ 48V) if it's tailored for a 6kW load, so you'd then need several packs. As Paul suggests, a trio of 7kW peak Powerwalls would do it, if you don't mind them being £5k+installation each.

More efficient would be to heat the instant hot water from the DC battery voltage directly. Most EVs are much more than 20kW so an EV pack plus a big water-cooled resistor would work, but that's all custom engineering. The safety system on what's effectively a DIY DC electric shower would be interesting.

Theo

If you had a switch, you'd have to work out where in the cylinder to put the sensor - maybe putting it at the top of the cylinder would be enough so that you'd use any stored hot water, and then switch to combi when it got cold? Or maybe it would oscillate when water was drawn and heated at the same time?

I think there's a limit to how much a combi can modulate down, ie if takes

I'd guess the safety interlock would refuse to heat if the input temperature was out of range. You might end up with the input water being 'too hot' while the output is 'too cold' - eg if the input was 35C that could be too high for the boiler to heat but not enough for bathing.

I don't see why you'd try to only run it during the middle of the night - that's going to be a terrible way to heat a house. At normal rate electricity after the COP it's roughly in line with the same price of gas, so it would be more or less neutral in terms of running costs. Of course you can optimise that with fancier tariffs (eg Octopus Tracker which gives you a daily market price ~15-20p rather than the standard 25p, or time of use tariffs) but even the standard price cap tariffs are about break even. It depends a bit how much you drive, whether the car usage or the heating usage is greater.

I'd do the calculations for your specific needs - how many baths / showers / sinks / etc you have. John's wiki page is a good one:

Last time I looked the value was 280g for the South West. I get your point. However is that the true average "total CO2" / "total generation" or one over time? Does it take into account the network losses.

Thanks for the link.

Yes.

In fact if the desired final output is. in fact, heat, a massive insulated tank or even an insulated concrete block with pipes cast in is going to be a way cheaper and higher capacity energy store than any battery.

It varies over time based on the weather.

For the regional figures:

"What’s included in the forecast The Regional Carbon Intensity forecasts include CO2 emissions related to electricity generation only. The forecasts include CO2 emissions from all large metered power stations, interconnector imports, transmission and distribution losses, and accounts for regional electricity demand, and both regional embedded wind and solar generation."

Theo

You could possibly connect the output of the combi to the input of a unvented direct cylinder. In the winter the combi would deliver hot water to the cylinder, and the immersion would just boost and maintain the temp. In the summer, turn the boiler off, and then the cylinder will still replenish from the (now unheated) DHW output of the combi.

No valves or extra controls required, just an immersion with a stat (and timeswitch if required)

(it would limit your flow rate through the cylinder to whatever you can get out of the combi)

Isn't that going to depend on the solar conditions, which will vary from day to day? If it's tipping it down with rain there will be less solar output and the water might be cold. Ideally on those days the combi would come in as a backup.

(when we had solar thermal there was reliable heat maybe 4-5 months of the year, and maybe 3-4 months it depended a lot on the weather. Of course solar thermal is better at collecting low grade heat than PV, although you can make up with it with lots of PV)

Theo

Exactly.

Even with economy 7, gas is far cheaper per kwhr than electricity.

I have a box that diverts excess PV generated electricity into the immersion element of our (primarily heated by gas boiler) water tank. It actually pulse width modulates 240(ish) Volts DC, so for instance if only 1kW is available, only 1kW goes in.

I've never properly done the sums, as I flog back to Octopus surplus electrons at 15 p/kWh and gas is about 7p/kWh, but how efficiently is that gas generated heat dissipated into the water, compared with the immersion element (which must be close to 100 % ?)

Is it better financially to heat by gas, and flog my ergs to Octopus, but from a CO2 perspective using the PV is thing to do ?!

I'd assume all the heat of the gas ends in the water, modulo the flue temp.

This is defined in one of the (several) heating values: lower, higher, and ISTR a handful more, and should get you in the right ballpark.

Thomas Prufer

This depends on your electricity tariff, if on intelligent octopus go and your DHW tank is low loss, J Rumm mentioned 60W which over the day is 1.2kWh, then heating offpeak overnight is mostly by wind and nuclear, could be as cheap as gas at 7.5p/kWh and selling at 15p/kWh during the day might mean less fossil fuel is used for generation.

I wouldn't assume that.

suggests 70-90%. But it doesn't mention the water return temperature which I happen to know is critical.

Vaillant mention the temperature

but don't give figures. I can't see any anywhere.

Andy