Sizing DC cables, PV Array to Inverter. Confused. Help please!

[Mark42]

My brain is overheating trying to make sense of PV panel parameters and calculate the CSA of a DC feed cable to the inverter.

For my 16 HL Solar 250W panels, the spec is:

Imp 4.94A
Vmp 50.6V
Isc 5.35A
Voc 60.5V

Two strings of eight panels in parallel.

Therefore:

Total current at max power = 4.94 x 2 x 1.25 = 12.35A

Total voltage at max power = 50.6 x 8 x 1.15 = 466V

[Note 4.94A x 50.6V = 250W per panel, as spec, but allowing for max tolerance, 4.94 x 50.6 x 1.25 x 1.15 = 359W (ie unlikely!)]

However the DTI guide (Sec 2.1.6) uses Voc and Isc for their main dc cable size calculations:

Total Isc = 5.35 x 2 x 1.25 = 13.4A

Total Voc =60.5 x 8 x 1.15 = 560V

[Note 5.35A x 60.5V = 234W per panel, but allowing for max tolerances, 5.35 x 60.5 x 1.25 x 1.15 = 465W (ie ridiculous!)]

I believe the de-rating factors of 25% and 15% are to allow for possibly over-spec panels.

But surely the panels cannot be SIMULTANEOUSLY over-spec on both parameters of voltage AND current? And ALL of them, ALL of the time?

But the DTI guidance appears to indicate that they can be. Is this right, or am I misunderstanding something?

Also, surely Voc cannot simply be added up to give a 'real-world' figure for series strings can it? Are there not resistive losses?

So which voltage/current figures does one use to calculate DC volt drop? And why is only 1% recommended?

[*** EDIT: this is wrong. It's less than 3%, Ref: DTI Guide, Para 2.1.4.1 ***]

If calcs are made using the above absolute best performance figures, surely we don’t have to engineer the cabling to give less that 1% drop on the very few UK days when the panels are producing at the ABSOLUTE MAXIMUM possible? Why bother? If there is a few percent loss on those fantastic days, then so what?

I don’t get this at all.

The upshot is, what is the minimum CSA for a 150m DC feed cable, XLPE SWA, laid to rest in the cold Norfolk earth?

 

[infinity]

Is DC not 3% Volt drop allowed? 1 % on AC!

 

[BruceB]

You are applying numbers to the wrong situation.
- the max Voc is important because the cable has to be able to take that from an insulation perspective (current will be zero).
- the max Isc is important because the cables must be able to take this current for a sustained period without melting the insulation and sheath (voltage will be lower than normal).
- the max Vmpp/Impp are important as they are the max operating point of the panels and the conditions at which you need to design the voltage drop to be an appropriate percentage, normally 1%.
So they are three different situations not one.

Sunny design or similar will give you an answer to your sizing question

Regards
Bruce

 

[yellowvanman]

Is DC not 3% Volt drop allowed? 1 % on AC!

Recommended is 1% on ac and dc.

 

[screwdriver]

allowed volt drop per string is worked out under mpp which is normal operating conditions, it 3% of 466 which is just under 14v you would use 12.35a for the amps to calculate volt drop if you were using 2 core, it would be 10mm which gives volt drop of 8.7v (12.35a*150m*4.7mv)

you would now double check using the voc and isc (corrected) that the cable can take 460vdc and 13.4amps , I should hope you don't need to refer to the regs to see if 10mm takes 13.4amps!!!!

The panels can't be open circuit and short circuit at the same time!

 

[screwdriver]

To be totally clear, just found this!

2.1.4.1 Cable sizing
Cables must be rated, as a minimum, to the voltage and current ratings derived using the multiplication factors in 2.1.2.
Standard de-rating factors must also be applied (BS 7671).
Cables should be sized such that overall voltage drop at stc between the array and the inverter is <3%.

 

[Mark42]

You are applying numbers to the wrong situation.
- the max Voc is important because the cable has to be able to take that from an insulation perspective (current will be zero).
- the max Isc is important because the cables must be able to take this current for a sustained period without melting the insulation and sheath (voltage will be lower than normal).
- the max Vmpp/Impp are important as they are the max operating point of the panels and the conditions at which you need to design the voltage drop to be an appropriate percentage, normally 1%.
So they are three different situation not one.

Sunny design or similar will give you an answer to your sizing question

Regards
Bruce

Thanks, Bruce, that’s great stuff!

But I want to understand the “Why?” of all this, not just plug numbers into someone else’s computer program and bow down, unquestioningly, to its sacred pronouncement.

Maybe the program is based on false premises, or ridiculous worst-case scenarios not applicable in every case.

That’s what engineering is all about, knowing when you don’t have to follow general rules like a parrot, and be able to state, robustly when necessarily, why that is.

Note that the DTI guide, in section 2.1.6, specifically states use Voc and Isc for cable calcs; it’s double–ticked too, indicating it’s a REGULATION, that mere mortals must follow.

Do you therefore agree that these are the wrong figures to use, in that Voc and Isc can never occur simultaneously in a real circuit?

I knew nothing about PV a week ago so hesitate to argue with ‘experts’ who write books so soon, but there it is …

 

[BruceB]

I quite agree with you - understand what tools are doing and their limitations.

On the double tick, I suggest you are over-interpreting:
- you do use Voc, but only to check insulation levels;
- you do use Isc, but only to check it will not melt;
- and you use mpp voltage and curent for the voltage drop calc.

Edited to add look at an output curve for a panel and you will see Voc and Isc can never occur together. Where would the power come from?

 

[Mark42]

Thanks screwdriver, that's great stuff too. So it’s 3% 'allowed' for DC and I was wrong. But my comment about why engineer for only the very best days still applies.

My gut feeling always was that 10mm would be absolutely fine, considering the total ENERGY being transferred (regardless of voltage, current, day of week or whatever).

But I do try not to think with my gut.

 

[screwdriver]

You use voc and isc to calculate the maximum current and voltage the cable needs to take as a bare minimum.

the vmpp and Impp are there for you to work out everyday running conditions for volt drop which would generally be what I start with on a cable calc as it is quickest to get the right size for longer runs.

Same lines as BS7671 calcs as you work out volt drop on the load and not the rating of the mcb.