Surplus solar diverter + battery storage issue?

[lakelandfolk]

Hello, Octopus recently completed our ASHP installation (£700 after deducting bus) which is performing very well. We wanted to use a solar diverter with off peak grid if required, to heat the DHW cylinder. We purchased a PV MATE solar diverter and Octopus have connected it to the cylinder for us.  I am having problems getting the surplus solar function to operate and wonder, because we have battery storage, if the batteries need to be 100% charged first before the diverter can come into action.

[JohnMo]

Ours is similar, only once battery is fully charged and the PV has no where to go but export, does the CT clamp pick up and divert to immersion.

[lakelandfolk]

Many thanks John, I thought that might be the possible problem.  We should get to 100% charge later to day so will keep an eye on the PV Mate

[lakelandfolk]

Hi John, today batteries are 100% and PV MATE is showing 1.5kwh + is being diverted to immersion, however test meter on immersion heater shows only circa 140volts AC is being delivered. I note that page 3 of instructions says "installation must be in the immediate vicinity of the water heater"  Our PV MATE is about 15 meters from the cylinder via a 2.5 cable, then spur outlet then .5 metres of 2.5 heat resistant flex. Could the 15.5 metre cable run be causing such a big voltage drop??  When we use the boost function then the normal 230v AC is delivered.  Any ideas/observations would be much appreciated

[JohnMo]

56 minutes ago, lakelandfolk said:

circa 140volts AC is being delivered

Not really sure how they modulate the power to match what's available? Maybe they vary volts to keep things in balance - could be talking rubbish. @ProDave maybe able to shed some light on it as he built his own diverter.

[lakelandfolk]

 Thanks John,  I'm still confused, despite the voltage at the immersion heater appearing to be 140v ac the heater appears to be performing OK and the water temp has increased in 1 hour from 43 to 55 degrees. I will try asking ProDave as you suggest

[dpmiller]

it's like a dimmer switch. it turns the heat up and down and so you will see the voltage vary, yes.

[JamesPa]

On 16/05/2024 at 12:47, lakelandfolk said:

 Thanks John,  I'm still confused, despite the voltage at the immersion heater appearing to be 140v ac the heater appears to be performing OK and the water temp has increased in 1 hour from 43 to 55 degrees. I will try asking ProDave as you suggest

The immersion heater is just a resistor, it will get hot whatever voltage is applied, just not as hot if the voltage is lower.

 

 

On 16/05/2024 at 17:14, dpmiller said:

it's like a dimmer switch. it turns the heat up and down and so you will see the voltage vary, yes.

Both actually work by cutting off the ac waveform at a point in each half cycle, the chosen point depending on the dimmer setting/amount of PV power available.  This reduces the average/RMS voltage without needing anything other than an (electronic) on/off switch, and without having to divert power to a resistor (which would get hot in the dimmer case and wouldn't achieve the desired result in the PV diverter case).  See explanation here.

 

Your voltmeter registers this as a reduction in voltage because it is designed to return the RMS (root mean square - essentially average) value.

[ProDave]

28 minutes ago, JamesPa said:

Both actually work by cutting off the ac waveform at a point in each half cycle, the chosen point depending on the dimmer setting/amount of PV power available. 

Many (including my home made one) work by burst firing.  Most electricity meters work on an "energy bucket" and only register power consumed when one whole watt hour has been consumed.  That takes a little under 1 second with a 3kW immersion heater.

 

So I set my burst firing time to be half a second and the immersion is on anything from 0% to 100% of that time.  So say there is 1.5kW surplus, then my immersion heater would be on for half of that half second period.

 

As long as energy both ways does not exceed 1Wh the meter will not register, it will count up a bit then count down a bit.

[JamesPa]

2 hours ago, ProDave said:

Many (including my home made one) work by burst firing.  Most electricity meters work on an "energy bucket" and only register power consumed when one whole watt hour has been consumed.  That takes a little under 1 second with a 3kW immersion heater.

 

So I set my burst firing time to be half a second and the immersion is on anything from 0% to 100% of that time.  So say there is 1.5kW surplus, then my immersion heater would be on for half of that half second period.

 

As long as energy both ways does not exceed 1Wh the meter will not register, it will count up a bit then count down a bit.

Interesting, I guess every day is a school day!

 

Do inverters integrate the input power for as long as half a second.  Its plausible that they do, but if not then burst firing surely wont actually match the load from the solar PV.  Interesting that it fools meters though, is that legal, or does it only work because generation is approx the same as consumption so there is both import and export taking place but they both balance and aren't registered?

Edited May 31 by JamesPa

[ProDave]

The solar PV will try and produce a sine wave output at whatever power is available.

 

Burst firing of a PV diverter is easier and less likely to cause RFI problems, Just think of the meter as averaging it's reading over a short time period, hence the shorter higher bursts of dump load get evened out as far as the meter is concerned.

[JamesPa]

28 minutes ago, ProDave said:

The solar PV will try and produce a sine wave output at whatever power is available.

 

Burst firing of a PV diverter is easier and less likely to cause RFI problems, Just think of the meter as averaging it's reading over a short time period, hence the shorter higher bursts of dump load get evened out as far as the meter is concerned.

OK so, if I understand you correctly, the way it works is that, in the 'on' periods of the burst mode there is import, in the 'off periods' there is export, and the meter averages the two out (or thereabouts) rather than recording both import and export separately.  Furthermore is fortuitous that meters work this way because it makes it possible to design a diverter which doesn't generate as much interference (but still registers 'correctly' on the meter) whereas if they did separately record import and export the burst mode design would result in an unwanted billing.  Interesting.

[SteamyTea]

6 hours ago, JamesPa said:

RMS (root mean square - essentially average) value.

If you root a square, you get rid of the negative components.

So.

√(230²) = √(-230²) = 230

 

[JamesPa]

14 hours ago, SteamyTea said:

If you root a square, you get rid of the negative components.

So.

√(230²) = √(-230²) = 230

 

Yeah I know, I was just trying to simplify the description to the point where even people who aren't familiar with AC (but who may nevertheless read the thread) will understand it.  The simplification is probably not necessary for this forum though even if emphasising, from time to time, the difference between kWh and kW is!

Edited June 1 by JamesPa

[Tetrarch]

On 15/05/2024 at 12:10, JohnMo said:

Ours is similar, only once battery is fully charged and the PV has no where to go but export, does the CT clamp pick up and divert to immersion.

 

This behaviour seems less than optimal.

 

I love this Elon Musk explanation:

 

 

 

I would have thought that there should be some kind of configuration that would enable you to choose how you want your diverter to behave.....

 

Regards

 

Tet

 

[JohnMo]

1 hour ago, Tetrarch said:

This behaviour seems less than optimal.

I am not charging a car and wanting to do it fast, the 80% figure is not relevant - I may only be charging  the battery at 3kW via a 6kW inverter (which is way below what the batteries could accept).  So the behaviour is just fine.