Solar panel supply

A normal domestic-size install is capped at a maximum of 16A unless the DNO agrees a higher figure.  3.4kW comes in slightly under that, which is why it was a popular size install.  Unless it's over about 4kW, it shouldn't be tripping a 16A breaker.


Some inverters have a habit of tripping RCDs.  My original one tended to trip the RCD the morning after a storm.  When I looked in the manufacturer's instructions, it recommended a 300mA RCD.  Of course, mine was connected to an ordinary consumer unit with a 30mA RCD.  That inverter failed a little while back; the new one hasn't tripped the RCD yet.

Is the circuit breaker rating relevant for a solar panel supply.  I have seen ratings as low as 6 amps and up tp 16 or 20.

Indeed it is - very relevant. The current the PV inverter produces flows through cables in just the same way as any load (if in the opposite direction) - so ratings & cable sizes need to be calculated in the same way.


Shock protection generally works a little differently though - because PV inverter can continue to generate a voltage after the protective device at the CU has opened. Usually the inverter spots the lack of mains and shuts down but the total time taken can be considerable - certainly longer than the 40ms disconnection time required for additional protection. The upshot of that is if the PV a.c. circuit needs additional protection by 30mA RCD (say because the cable is concealed in a wall) then you need a double pole switching RCD so that the PV circuit becomes a separated from earth once the RCD trips - so preventing shocks to earth even in the inverter is still generating. (See reg 551.7.1 (ii)). Single pole switching RCBOs are unlikely to be suitable in such cases.


  -  Andy.

OK I will light the blue touch paper!


Given the CSA of the conductors to the inverter need to be slightly over sized for a reduced volt drop and the output of the inverter is limited by the ability of the panels to deliver their maximum output, why do your need overload protection?


Also I have never measured the short circuit current  myself on the DC side by shorting out the DC supply conductors and clamping the short, what sort of PFC do you get on say a 3.4kW panel? I do know to do this you have to measure the light falling on the panel(s) and apply a calculation. So what PFC on the DC are you getting typically?

OK I will light the blue touch paper!

:

the output of the inverter is limited by the ability of the panels to deliver their maximum output, why do your need overload protection?

 

Good point - I can see that the PV inverter could be treated as a fixed (maximum) (negative) load - so in theory overload protection could be omitted. You'd still need fault protection on the a.c. side though, and ADS - which might limit your maximum MCB size a bit though (especially if the inverter is a decent distance from the CU).


   - Andy.

Short circuit current of a solar cell is not quite like anything else.

The solar panels are just long chains of photo-diodes, albeit rather large area ones, so very slow compared to rectifiers. And diode-like they have a highly non-linear current voltage characteristic, that depends on the light level. In the region near the PN junction,  each photon that is absorbed liberates a hole- electron pair that whizz off to the electrodes to be the load current, and 1 pair per second is 1.6 E-19 of an amp - so you need a lot of photons. (hence the large area - to scale this to human experience, a twinkling but weak star is about 10 photons/second through the eye, which is a few square mm of pupil area). Note that less than a quarter of the photons are absorbed in a place where electricity is generated, some reflect and many get lost in the substrate and just warm it up.

At low loads, where the number of photons absorbed is more than the current demand,  the cells are almost constant voltage devices, and the current can be increased, up to the point where this is no longer true, and then the voltage collapses.(  A cell in the dark may have the right voltage off-load but is a high resistance - so to avoid one dark cell in the middle of a series chain  due to a badly placed bird deposit, leaves, or shadows  blocking the output of the whole string there are additional anti-parallel bypass diodes around each cell, but this is a refinement, albeit a necessary one.)

So for optimum output the current drawn needs to be increased to just under the point where the volts collapse - so short circuit is iess than 150% of whatever the optimum load would be for that particular level of illumination.  On the AC side of the inverter, if you short it you will get a couple of cycles of high current while reservoir capacitors run down, and then (if there was not grid failure  lock out) the current would drop back to a  level not much more than it was under load. In reality the electronics either blows up or shuts down gracefully depending how well it was designed and how bright the sun is on the day.

Mike


Thanks for the explanation. As I suspected the ability of both the inverter and the PV string to to deliver shed loads of fault current is not the same as our DNO transformer supplies.


I wonder what the purpose of the short circuit test on the DC conductors is designed to prove or verify? Shorting the DC conductors together and switching on to apply clamp meter seems to be counter to self survival and survival of the PV panels to me as we would not do this on a generator or transformer supply outside of a test facility with ballistic protection.

Under what conditions is this test really required ? - all it tells you is how bright the sunshine is today, and perhaps that the wiring is correct, that you can verify on-load.

No need for blast containment - as noted earlier the short circuit current will only be a bit more than the normal running. But it is not a very sensible test - unless you are at midday and optimally aligned to the sun on a clear day in June , it is quite likely that at some other time the normal running current will be higher. 

Mind you, today must be close to UK peak solar.
gridwatch  reckon 20% of UK power is solar today - so I suspect a  record soon if not already.