So now I have a solar PV system how do I make the most of it???

EricMears

sly_dog_jonah wrote: »

Ah yes the old chestnut of taking multiple current-sensed values, and doing some arithmetic in real-time to work out what your consumption and export is. I'm yet to see a solution for this that doesn't require a great deal of h/w and/or a PC to do the control (although it should be possible to program such a simple operation into a programmable IC). A difficulty is that many home's electrical layout don't allow consumption to be measured accurately - there simply isn't a single cable available to clamp around.

Somewhat frustratingly, the Wattson has one input for generation and up to three for consumption and does all the arithmetic but doesn't normally do anything more useful than display the answer.

The generation one will always be simple to arrange - but my installer found room inside the isolator back-box for the sensor. With a split load CU, it was relatively easy to fit a sensor around the internal lead that feeds the RCD-protected circuits. The solar installation is of course attached to the non-RCD side and it was relatively simple to arrange that all the other non-RCD circuits (just a couple of lights and a feed to my UPS) were fed from the same CU circuit breaker with yet another sensor on that line.

I don't see it as a 'problem' to need a computer in the setup. I've usually got at least one running anyway and that's already talking to the SunnyBoy. If I do decide to go down the Linux route I'll probably install that on an old computer which I've never got round to throwing away.

zeupater

Martyn1981 wrote: »

Thanks Z. I think it's fair to say that I've run out of at talent at this point.

The 4 cables into the isolation switch, then 2 into the inverter, seems a little odd. Still no better than an absolute beginner on Sunny Design, but it wasn't happy with 5+5 or 6+6. In fact 7+7 was still too low a voltage, thought this did at least get a suitability score sadly it was 45%. 8+8 got 93%.

Like you, I also suspect that the isolation switch is being used to connect the two sets of panels as one string of 10. In order to get a high enough voltage to operate the inverter. The less suitably aligned panels then acting as a brake on the system.

Ed, at this stage, depending on your relationship with the installer, it might be worth just waiting and getting some more time under your belt. Though not fun, the system is working, and it may take a month to get enough comparative data. Unless of course you can find a nearby system with a similar orientation. Size doesn't matter (stop sniggering!), as you can just proportionalise output.

Anyone with recommendations?
BDPV, PVOutput etc?

Hopefully, there is absolutely nothing wrong, but I'm struggling to understand how a single string can operate across two differing roof angles (let alone potentially differing shading). Fingers crossed the fault lies entirely with my limited knowledge, not with the system.

So give it a bit more time, I'll keep digging, as I'm enjoying learning more about strings, inverters etc. If things still look wrong then it's time to have a chat with the installer.

You could also in the meantime, contact SMA, here's the site, send them an e-mail and ask for their opinion on the suitability of a single inverter for 2 roof pitches.

http://www.sma.de/en

Mart.

Hi

The OPs panels weren't available in the current database on my version of SunnyDesign, so I modelled with 10 Schueco polycrystalline panels and the results with a single 10 panel string didn't look too bad, but the accuracy of what I looked at would depend on the performance specifications of the panels which are actually installed.

Regarding the angles of the panels, it looks like we're only talking about a ~5degree deviation from a midpoint for the two panel sets, so the overall performance difference would likely be minimal anyway.

HTH
Z

Martyn1981

zeupater wrote: »

Hi

The OPs panels weren't available in the current database on my version of SunnyDesign, so I modelled with 10 Schueco polycrystalline panels and the results with a single 10 panel string didn't look too bad, but the accuracy of what I looked at would depend on the performance specifications of the panels which are actually installed.

Regarding the angles of the panels, it looks like we're only talking about a ~5degree deviation from a midpoint for the two panel sets, so the overall performance difference would likely be minimal anyway.

HTH
Z

Yep, same here, I used UE Solar ZHM240. In fact Sunny design seemed to like the combination.

Hopefully it is working fine, just poor weather so far, and me just scaremongering. I'm sure Ed will have a better idea in a few weeks how it's all going, and in comparison to PVGIS. Always tricky to judge (compare and contrast) in the early days with so many variables to take into account.

Thanks for having a play.

Mart.

sly_dog_jonah

EricMears wrote: »

Somewhat frustratingly, the Wattson has one input for generation and up to three for consumption and does all the arithmetic but doesn't normally do anything more useful than display the answer.

The generation one will always be simple to arrange - but my installer found room inside the isolator back-box for the sensor. With a split load CU, it was relatively easy to fit a sensor around the internal lead that feeds the RCD-protected circuits. The solar installation is of course attached to the non-RCD side and it was relatively simple to arrange that all the other non-RCD circuits (just a couple of lights and a feed to my UPS) were fed from the same CU circuit breaker with yet another sensor on that line.

I don't see it as a 'problem' to need a computer in the setup. I've usually got at least one running anyway and that's already talking to the SunnyBoy. If I do decide to go down the Linux route I'll probably install that on an old computer which I've never got round to throwing away.

Personally I'd rather not have to have a PC running all the time to control the immersion, but if you have one doing so anyway it's not greater load to use it for this purpose.

Sounds like you were quite lucky you were able to clamp all the leads you needed to. Having the foresight to do this at installation time was good, I'm sure the majority don't realise the practical difficulties in measuring generation and consumption in real-time until long after the installer waves goodbye.

pwllbwdr

sly_dog_jonah wrote: »

Wherever you connect the controller, you'll need to do 'fiddling' - or rather an electrician would. The fusebox is ideal for current sensing as the Solar PV cable and Immersion circuit (if on a dedicated circuit) are co-located.

Out of interest how much is oil per kWh compared to gas?

On a grey day in summer you'll be heating water with your immersion, which could cost around 7kWh depending on the size of the tank, insulation, start temperature and thermostat setting. That's about 70p/day by my reckoning. How would oil heating cost compare?

Oil is currently about 60p per litre, which works out at about 6p per kwh in the oil itself. My boiler is about 70% efficient on the burn, and there will be further losses in the heat exchanger coils to the water tank. Say 8-9 p per kWh. Less than imported electricity, but much closer to it than for a similar comparison with gas.

My PV install is in a separate building linked to the house by a cable. So it isn't possible for me to install a current switch in the fuse box. That's why the light sensitive switch is more appealing to me as it could be wired into the bathroom between the spur and the immersion heater.

grahamc2003

The manual for my Sunnyboy specifically states that if wired with two strings, both strings must be the same size, slope, direction etc - i.e. looks like the two strings are managed by a single mppt. If the strings are a different slope etc, then the single mppt will have trouble getting the optimum voltage (or rather the optimum voltage for 2 strings combined will probably be off the optimum voltage for either string).

I expect this is why my installers said they needed a 'special' inverter (which they couldn't source in time) for my planned system on two direction (SW and SE) and different number of panels (7 and 4). Hence why I've got a small system on one roof only (to meet the march deadline). I now think the inverter I needed was a twin mppt system - are they that 'special'?

Most of these problems can be cured by microinverters or, probably better, by a solaredge type system, where just the tracker is on each panel, and not the inverter too. Of course, those systems may produce reliability problems in the longer term, with more electronic boxes (outside) to go wrong.

Casper55

Have purchased a heat pump to heat my koi pond ( Normally heated by gas boiler) and run it during the day. I have estimated that once ambient temps get above 10 degs the Heat pump has a cop of 5 or more so cheaper than gas. My Soalr panels are now producing in excess of 2.8 kw from mid day which I can use 1.2 kw free on the heatpump so none of the energy produced by the panels is going into the grid. Even though the koi pond is only heated to 24 degs in the summer, once the covers have been removed it has been quite expensive to heat using gas . The heat pump Cost me £ 1000 so it will be interesting to see how long it would take to get my money back if I ever do

larkim

Wouldn't it be more "money saving" to eat the fish?

don0301

15.1 kWh today

47.1 kWh for last 3 days

198.9 kWh so far for month

wish these damn things would work in Winter

edwink

Martyn1981 wrote: »

Thanks Z. I think it's fair to say that I've run out of at talent at this point.

The 4 cables into the isolation switch, then 2 into the inverter, seems a little odd. Still no better than an absolute beginner on Sunny Design, but it wasn't happy with 5+5 or 6+6. In fact 7+7 was still too low a voltage, thought this did at least get a suitability score sadly it was 45%. 8+8 got 93%.

Like you, I also suspect that the isolation switch is being used to connect the two sets of panels as one string of 10. In order to get a high enough voltage to operate the inverter. The less suitably aligned panels then acting as a brake on the system.

Ed, at this stage, depending on your relationship with the installer, it might be worth just waiting and getting some more time under your belt. Though not fun, the system is working, and it may take a month to get enough comparative data. Unless of course you can find a nearby system with a similar orientation. Size doesn't matter (stop sniggering!), as you can just proportionalise output.

Anyone with recommendations?
BDPV, PVOutput etc?

Hopefully, there is absolutely nothing wrong, but I'm struggling to understand how a single string can operate across two differing roof angles (let alone potentially differing shading). Fingers crossed the fault lies entirely with my limited knowledge, not with the system.

So give it a bit more time, I'll keep digging, as I'm enjoying learning more about strings, inverters etc. If things still look wrong then it's time to have a chat with the installer.

You could also in the meantime, contact SMA, here's the site, send them an e-mail and ask for their opinion on the suitability of a single inverter for 2 roof pitches.



Mart.

Hello Mart

Thanks for all your help. I have contacted sma.de as you suggested to ask their opinion regarding the inverter.
Will keep you posted. Thanks again!!!