Cable Size for Power Generated from Solar

It depends on what your voltage is going to be...I would construct a small building next to the solar collectors and convert to A/C, raise the voltage for transport...This would require minimum wire size and line loss....

Of course I'm assuming some site conditions and code issues that I may have wrong.

I believe this advice to convert at the site of production and transmit to AC is incorrect. You should move the DC current at the highest voltage you can, which in Malaysia is probably limited to 1000V. Your string sizing calculations will give you maximum power and maximum open circuit voltage. You must observe these limits when sizing conductors. Don't forget your equipment ground. You are building 6000 watt strings, at 500V you are only carrying 12 amps. If you combine those strings, you should be able to use the same conductors for each 12KW subarray. (24 amps) I would use dual three phase inverters to phase correctly, with one 12kw supply on each. If you do this, you will not have distance or phasing issues. Your solar installer, panel manufacturer, inverter manufacturer and site conditions should all be coordinated by the same person or company.

You will have significant losses, as the distance to the meter is very long for a 24KW array, in terms of feasibility. Hopefully, you will have excellent orientation and tilt at you PV site.

Conduit fill, conduit exposure, and ambient temps, distance will all need to be calculated to determine cable sizing. Inverting at the production site, 300M away, to AC at 240 will require even larger conductors and give you even more voltage losses.

Please note that I specified the highest a/c voltage possible for this situation.If phases are balanced,there will be very little inductance loss. A/C circuit breakers and switches are cheaper than DC also.

In the USA, you will need a minimum of 5 conductors(for 3Ph).A neutral must be carried with the phases whether it is used or not, and an equipment ground must also be installed, based on the size of the upstream breaker.Ambient temperature de-rating must also be applied, using the highest temperature that the conductor is exposed to.

There are many reasons why A/C is chosen for the utility grid for power transmission, which I will not go into.Suffice it to say that they do it for very good reasons of economy and efficiency.

There are cases where DC is more efficient, but it requires the use of High Temperature super conductors.

If grid power is available nearby, a previous poster's idea of selling it back to the grid is very good.You could also use the stored energy to trim your peak load for an even faster payback of your investment.(Sell high, buy low.)This would require additional investment in an energy management system, but these can usually give a very good ROI, if properly designed and installed.

The incorrect advice I was referring to was from number 1

This is an office building on the grid, with three phase grid supply. I never heard batteries mentioned. I am suggesting the placement of the inverters next to the distribution CT cab in the office building electrical room, and supply them (the two 3-phase) inverters with high voltage DC current from the array, which is 310M distant. We've built these systems, and they generally outperform site (by this I mean the site of the PV panels) inverted (DC to AC) PV systems. The output of the inverters could be delivered on 1 meter cables to the CT cab, which would provide backfed breakers (AC). This is what I call a short run.

Yes, DC disconnects are more expensive. Over 25 years, they are not.

I don't see this as being any better....In fact it would seem to be more expensive, more complicated...

Without the voltage being mentioned, no one can know the amps, nor the wire sizes.

Sorry..my silly mistakes. Its going to be AC 415V three phase.

The current more or less should be around 39.3A

3-phase? Why not just send the DC indoors as it comes off the panels and invert it there? At least your inverters will be in the dry!

I do not know what they put the inverter next to the panel in the hot sun.

Is it possible for us to send DC instead of AC? wouldnt the losses would be more?

Not with the right size cable <sigh>.

Anyway, if you're worried about losses, what are you going to do when it gets cloudy?

It's net metered, grid connected Crabby. Why would you comment when you haven't a clue?

'Coz I'm trying to find out wtf is going on.

I'm an ass, Crabby, because I missed this post, and you are exactly correct. I then told you that you haven't a clue!

We build remote arrays like this all the time. It's much more effective, but it's hard to convince some people. I frequently build open circuit PV strings that are around 550V DC, 500V at Maximum current, and move 6 or 7 amps on THWN #10 for 500 feet to my inverter. Production losses are smaller, with less expensive conductors.

That strategy sounds good, but the OP's strategy has remained largely inscrutable.

For the sake of argument, and to silence the doubtful, please provide the figures for your installation using both methods.

That way no one has to take a simple claim as an undisputed truth.

I look forward to your analysis.

http://cr4.globalspec.com/search/sitesearch?do=show&sort=textmatchrank&srch=British%20Standard%207671&order=asc

http://cr4.globalspec.com/search/sitesearch?do=show&sort=textmatchrank&srch=cable%20sizing&srchobjs=t%2Cc&fs=12&order=asc

http://www.engineeringtoolbox.com/maximum-current-copper-aluminum-wire-d_1690.html
http://www.engineeringtoolbox.com/awg-wire-gauge-d_731.html
8.36mm²...cu

I am not up to date in this area.

Quote "Since it is 4 strings/block, the arrangement for the phases would be

L1 = 12kW, L2 = 6kW, L3 = 6kW"

Why is the load more balanced? Does it come in 6 kw blocks?

The solar is in 6kW block. They are going to install 4 block. That is why the power generated is arrange that way. I hope this is clear to answer your question.

Usually the 4 banks of solar cells will provide DC, the inverter will convert it to 3phase AC and (if your load is balanced) provide the same current for each phase.

If there are no local wiring regulations you must conform to the the cable size is chosen to provide an acceptable voltage drop (typicaly <5%), after all you dont want to waste the power you've just collected.

So 310m of 3C,N+E coppper cable carrying 40A you could use >16mm2

Ia=Pa/U/cos5=12kW/0.24kV/(?)=50A. So the current shall be at least 50A. Consider 0.9 Cos5, we suppose 55A current.

5% DeltaU, 25mm2 XLPE cable is enough. Al cable I think maybe 35mm2.

We do not use Al cables for long long time.

My calculation is like this.

Consider for L1,

P = 12 kW, Current (A) = (P x 1000) /(Power Factor x 3phase voltage x √3)

= 12000 / (0.85 x 415 x 1.732)

= 19.64 A

If consider voltage drop less than 2%, according to this link http://www.cablesizer.com/iec/voltdrop/, I will have this result,

For a 70 mm² Aluminium, PVC, Single Core cable of length 350 m:

The three-phase voltage drop is: 6.8405 V

(Equivalent to 1.6483 % of nominal voltage 415 V)

So i will use (70 mm2 x 3 ) and plus another cable 50mm2 for neutral.

This is suppose to be sufficient enough right?

From what I heard before, for overhead installation, we use aluminium cable. Why is it so?

You have consider for L1, Why still divide 1.732 and use 3phase voltage?

Do not consider unbalance shall be:

P = 24 kW, Current (A) = (P x 1000) /(Power Factor x 3phase voltage x √3)

= 24000 / (0.85 x 415 x 1.732)

= 39.28A

2% voltage drop:

I think steel core / Al wire (70/10) maybe OK.

Your single phase voltage is 240V, ie 3phase voltage/√3, 415/1.732

So, 12000 / (0.85 x 415/1.732) = 58A for L1

Over the distance quoted your looking at 25-35mm^2.

Check the Malaysian/Singapore standards or ask your local electricians to confirm.

Bundled aluminium conductor is the way to go cost wise with properly swaged bi-metal crimps where you are making an Al/Cu transition. The neutral conductor is normally smaller.

The neutral conductor in some cases should actually be larger than the main conductors.This is the case where there are many switch-mode power supplies that generate large amounts of harmonic currents.These currents are not detected by a normal clamp-on amp meter, because of the high frequency.Nonetheless, they create heat in the neutral that has resulted in equipment damage(open neutral situation), fires and property loss.Such currents are present in a computer-intensive building

The NEC recommends doubling the size of the neutral in data processing centers.

Newer switch-mode supplies only switch when the power is at zero(zero-crossing switching) which results in less harmonic currents being generated.

Never encountered that requirement.

I'll look into that. Thanks for the heads up.

Here is a link that you might find useful:

http://ecmweb.com/mag/electric_solve_neutral_overload/

and an excerpt form the NEC:

Neutral tip. Sec. 310-15(b)(4)(c) requires you to increase the size of the neutral supplied by a 3-phase, 4-wire, wye-connected system if a major portion of the load consists of nonlinear loads saturated with harmonic currents. FPN 2 of Art. 220-Sec. 22 warns of this problem. Here, you may have to double the neutral in size or harmonic filters installed to correct the problem.

Both those URLs were to the same WC&M article mate. That was interesting.

The URL or the NEC link would be much appreciated if you have it at your finger tips.

Could the higher H3 neutral current manifest or present itself as an apparent over voltage episode in the reticulation network? Thinking aloud here....

Here are some links that explain things a little better than the NEC,which is not designed to teach, merely to state the rules.

http://business.highbeam.com/2095/article-1G1-15231760/adhoc-subcommittee-nonlinear-loads-reports

http://www.iaei.org/magazine/2000/03/wire-temperature-ratings-and-terminations/

Thanks for that.

I would not use the minimum calculated size .The larger cable you can afford, the better for overall efficiency(less loss).

The problem with Aluminum conductors used to be the softness of the metal.When it expanded due to heat, it would leave an indentation,causing the connection to loosen more and more each thermal cycle.Final result was excessive heat and fire hazards.Metallurgists have addressed the problems and a harder alloy is now used.

Ariel conductors use a special aluminum crimp sleeve or CAD welding to terminate or splice the wires.Ariel wires are also less susceptible to temperature gradients than enclosed conductors, so the thermal expansion problems seldom occurred in these applications even with the old aluminum wires.

The new aluminum alloys are much better than in the past.If proper cleaning,anti-oxidation coating, and torque specifications are followed, there will be no more problems than with copper.Use Cu/Al rated connectors when transitioning to/from copper wire to aluminum.

Do not use any underground splices unless CAD welded, and I don't particularly like these.(Emergency repairs only.)

What type storage media are you using?

Batteries I presume?

These should be as close to the solar panels as possible.Then invert to A/C.The higher the A/C voltage output the better (smaller cables,less losses), you can always step it back down at the destination.

A higher initial cost can be offset by higher system efficiency in the long term.

All phases leaving the inverter(A/C) should be closely balanced, or you will get inductive heating of the conductors and increased losses.

Hope this helps.

Thank you.

I calculated the aluminium cable with allowable voltage drop less 2%. We can't afford to have larger cables than the required loss.

We do not have any media storage. This system is for showcase and testing for solar panel. We just use whatever being produce so, we don't need any storage. More over the batteries will cost a lot.

I have already highlighted that the phases should be closely balance. But I'm not sure why they want to do it that way.

Yesterday, a solar installer highlight that we need to have lightning arrestor. Is this a must? since our system have its own SPD (surge protection device) unit.

A surge protector will not disperse a high-energy lightning strike.It will, however,help prevent over-voltage conditions due to spikes in the incoming power.

Lightning is very unpredictable,and is notorious for breaking the rules,but it does not hurt to put the odds in your favor.If lightning is common in your area,then I would recommend lightning rods.If there are tall structures near the array, the structures will give some protection.A rough rule of thumb is to extend a 45 degree cone from the top of the structure to ground, and whatever falls within the circle is "usually"provided some protection.However, to play it safe, install lightning rods.Make certain that the installers are fully qualified or the results may be inadequate or even worse than nothing at all.Do not let them sell you ion emitters to prevent strikes:They have not been proven to be effective.

Keep in mind that even the best lightning rod system cannot protect from the electromagnetic field (EMP) damage from a severe nearby strike, and I am sure you cannot afford a Faraday shield on everything.

Good luck with your venture.

I have studied local installations in Northeast USA and find the trend is to make solar conversion as efficient as practical (>17%), forget about energy storage, feed unused energy back to the grid during peak solar ingress, and to use power from the grid during low solar ingress. Local inverters are positioned under each panel and feed AC back to the mains. Local inverters are small, efficient, and are guaranteed for 25 years. All this permits the use of smaller cables than running DC from the panels.