"HELP" pv problems

[oddjob]

we installed recently a 3.9 system on a aliminium storage shed, from the inverter used 2.5mm to isolator switch, from isolator 50mtrs of armour cable (2.5) to Junc box joined to 2.5mm 15mtrs straight into the gen meter then from gen meter back 15mtrs to 2nd isolator beside consumer unit. i also had to put in the c/u so i split the tails to link into new c/u jus for the solar.
I have3 had the customer on saying he has only generated 6kwts today and he is south facing, but i have generated 18kwts today and mine is s/w facing in the same area. i have replaced the gen meter in the past as that did not seem to be working. can anybody give me some ideas. cheers oj

 

[Markc]

80mts on 2.5mm? Your voltdrop is going to be over 1%. A very quick cable calc and I have 25mm 3core over that distance.

This may be your problem as you are generating.

 

[BruceB]

The cables do appear to be well undersized.
Put the generation meter beside the inverter and the customer will get paid for the power lost in the long swa cable.
Regards
Bruce

 

[Markc]

The cables do appear to be well undersized.
Put the generation meter beside the inverter and the customer will get paid for the power lost in the long swa cable.
Regards
Bruce

Good call but still leaves an undersized cable.

 

[yellowvanman]

With respect to calculating volt drop, the volt drop/A/m is based on the conductor being at its rated temperature 70 or 90 degrees. With your array you'll be generating 16-17A, so that wouldn't heat up a 25mm cable to the operating temp, so I would suggest that a cable size somewhere in between maybe optimum - certainly a lot cheaper than 25mm SWA.

Think about this as the reverse of the 'rule of thumb' when measuring Zs and making sure its at least 80% of the calculated!

 

[Markc]

With respect to calculating volt drop, the volt drop/A/m is based on the conductor being at its rated temperature 70 or 90 degrees. With your array you'll be generating 16-17A, so that wouldn't heat up a 25mm cable to the operating temp, so I would suggest that a cable size somewhere in between maybe optimum - certainly a lot cheaper than 25mm SWA.

Very good point and worth considering. The only thorn in the side will come when you are asked to prove your results. Assessment or alike. I will have a look at my cable calc software to see if I can reduce the temperature rating. I agree it is very unlikely or impossible for the system to run a cable at operating temperature. There may be more on this subject to come!

 

[yellowvanman]

I had a similar problem with supplying a pump for a water spring which was over 500m from the house, software said 16mm 3 c, I said 10mm 2 core! saved £100s.

I think GN3 gives an equation to work out the resistance of a cable at temperatures between ambient and rated temp, haven't got my copy to hand at the mo.

At the end of the day though there is nothing written in stone that the VD should be 1% is there? Obviously need to keep it to a minimum to get best payback.

 

[Markc]

I had a similar problem with supplying a pump for a water spring which was over 500m from the house, software said 16mm 3 c, I said 10mm 2 core! saved £100s.

I think GN3 gives an equation to work out the resistance of a cable at temperatures between ambient and rated temp, haven't got my copy to hand at the mo.

At the end of the day though there is nothing written in stone that the VD should be 1% is there? Obviously need to keep it to a minimum to get best payback.

No it is a recommend not a reg so good working practice. If it is the case it shows from the OP what issues can arise from a small cable. I think 80mts of 2.5 is a little optimistic.

It's been to long ago since I've needed to work cable size out long hand and I prefer my software to tell me now.

 

[oddjob]

starting to panic now does this mean i should re-wire in bigger cable or by putting the gen meter beside the inverter will i get away without a re-wire

 

[TedM]

From the PV Installation Guide:

The a.c. cable connecting the inverter(s) to the consumer unit should be oversized to minimise voltage drop. A 1% drop or less is recommended. However in larger installations this may not be practicable or economic due to the very large size of cable resulting. In this case the designer should minimise voltage drop as far as possible and must remain within voltage drop limits as prescribed by BS 7671.

Note: When generating, the voltage at the inverter terminals is higher than the voltage at the consumer unit. This voltage drop must be kept to a minimum in order to prevent the inverter nuisance tripping on over voltage.

If you are stepping outside these guidelines then you have to have a clear reason for doing so.

 

[Markc]

starting to panic now does this mean i should re-wire in bigger cable or by putting the gen meter beside the inverter will i get away without a re-wire

What inverter is it? A good majority show on the inverter generation totals either in days, weeks, years or total and a actual reading. How does the total generation rate of the inverter match the gen meter? If there's a big difference between them then you have your answer. Nothing in the regs states a maximum of 1%Vd but it is good working practice. It will be your choice on how to over come the problem.

 

[claypole]

http://download.sma.de/smaprosa/dateien/5692/SB30TL_40TL_50TL-IEN103630.pdf
Don't know what inverter is used here but for the Sunny Boy TL's page 26 of the install manual above would suggest even 10mm on the AC side is insufficient. Presumably the low yield is down to tripping on the inverter whenever power is high? Possible to verify this on inverter monitoring software? Is the problem worsened by the grid monitoring on the inverter already reading very low due to cable loss supplying the inverter, and then loss again on the output side?

 

[oddjob]

Inverter is a Fonius 3.6tl. I'm new to the pv game and have only installed 8 jobs all relatively straight forward domestic installs apart from this one - I did the NICEIC pv course and they only suggested 2.5mm cable. Feeling a bit of a pratt now, but I've put in so many hours trying to get everything in place I can't think straight and I'm not sure how to put it right ..... happy to sort it for the customer but not sure where to start. Haven't got cabling software - maybe should have got that first???

 

[oddjob]

What inverter is it? A good majority show on the inverter generation totals either in days, weeks, years or total and a actual reading. How does the total generation rate of the inverter match the gen meter? If there's a big difference between them then you have your answer. Nothing in the regs states a maximum of 1%Vd but it is good working practice. It will be your choice on how to over come the problem.

It seems that generation on the inverter is greater than generation meter. I'll need to check this is the morning but from what the custmer has said it sounds as though that's what's happening.

 

[Markc]

Inverter is a Fonius 3.6tl. I'm new to the pv game and have only installed 8 jobs all relatively straight forward domestic installs apart from this one - I did the NICEIC pv course and they only suggested 2.5mm cable. Feeling a bit of a pratt now, but I've put in so many hours trying to get everything in place I can't think straight and I'm not sure how to put it right ..... happy to sort it for the customer but not sure where to start. Haven't got cabling software - maybe should have got that first???

I think you have to make the decision on what to do.

 

[BruceB]

starting to panic now does this mean i should re-wire in bigger cable or by putting the gen meter beside the inverter will i get away without a re-wire

Putting the meter by the inverter hides some of the problem, but my tongue was firmly in my cheek when I said that.
I suggest you have to fix the design.

Regards
Bruce

 

[JulianC]

Another possiblity (not done any calcs so don't know how likely this might be) is that the volt drop in your cable is enough to send the output of your inverter up to its G83 trip limit under periods of high irradiation. It therefore could be tripping out when the sun comes out, hence the low generation figure.

 

[edexlab]

Isn't there something in the regs which says equipment installed should be to manufacturers instructions ,as long as this does'nt compromise BS7671
In most inverter manuals there's a reccomendation on cable csa and length of run ,some ie Sma sb 4000tl have a minimum cable size of 6.0mm

Cable cross section Max. cable length
SB 3000TL-20 SB 4000TL-20 SB 5000TL-20​

4.0 mm² 23.5 m not permitted not permitted
6.0 mm² 35.2 m 23.3 m 18.6 m​

10.0 mm² 58.7 m 38.8 m 31.1 m

 

[BruceB]

Isn't there something in the regs which says equipment installed should be to manufacturers instructions

It's stricter than that. 134.1.1 says "Electrical equipment shall be installed in accordance with the instructions provided by the manufacturer of the equipment."

Regards
Bruce

 

[DH Electrical]

My opinion is you will need to rewire the cable run in larger cable, as if the generation meter is moved closer to inverter it will cause meter to possibly be more accurate but you may still have volt drop