Solar Freaks

Cabling between the Panels is via the leads with plugs, But then via a J box etc to the inverter, and inverter to the SWB, the size of cable can be 2.5mm or 4 mm in size depending on voltage drop etc. OK, but what type of cabling is it, is it normal domestic power point cabling 2*2.5 +e cable or 2*4mm+e cable? and does it make any difference as long as its 2.5 or 4 mm copper conductors and not special twin cabling?
please advise?

Cable type DOES matter. Obviously the 240VAC cabling from the inverter to the switchboard can be commonly-available "TPS" (preferably 4mm, possibly larger depending on length of cable and size of inverter(s), but many installers use 2.5mm). The insulation voltage rating of the cabling from junction box or isolator (in/on roof) to inverter needs to be capable of "1.2 times" the open-circuit array voltage. This catches MANY installers out, as they typically use the same 2.5mm or 4mm TPS cable (and sometimes twin instead of twin-and-earth) which is usually rated at 500/1000V but I've seen as low as 450V (single-insulated = 500V, double-insulated = 1000V - but remember when using twin, there is no second layer of insulation between the two cores within the cable). When using 9 or more panels in series, the open-circuit voltage of the array usually exceeds the "1.2 times" rating of the cable insulation.

The cabling also needs to be labelled at regular intervals to prevent it being confused with "regular" wiring.


This "1.2 times voltage rating" problem is also applicable to the DC isolators which are rated at either 500V, 800V or 1000V. Usually the same (dodgy/careless/inexperienced (cross out non-applicable)) installers use the cheap 500V $40 isolator on every job. The 800V isolators are $300+, no idea of the 1000V ones, but you can start to get an idea of how some installers try to save considerable $$$ by taking shortcuts, but unfortunately it's not entirely their fault as some systems are delivered to site as a pre-packaged kit (so much for the "design" of the installation).

I've come across this problem with every SMA2500 inverter that's been installed with 12 panels in series (regardless of installer)... SunnyDesign software gives you all the details, but assumes the installer is smart enough to realise he needs to use suitably-rated cabling and isolators...

The installers that know their stuff (and don't skimp) use single-core double-insulated RADOX cable, or the specially-designed solar cable made to withstand UV as well as all the other requirements (I can't recall the brand).


It can become quite an expensive "defect" to fix when found by the inspector!

Inspector,

My understanding is that the '1.2x' factor is been reduced to '1.1x'. Can you confirm this is true?

Also, im not really sure why we have 1.2x anyways. If the manufactures state the goods are safe to that level, why do we need to factor in 20%?? Im sure the manufactures have considered some type of safe leeway as well.

Also, i know may system running 9 panels in series using standard cabling (2.5mm twin). Are these all defects? If so, how are they still allowed to operate?

TonyB wrote:Inspector,

My understanding is that the '1.2x' factor is been reduced to '1.1x'. Can you confirm this is true?

Not at this stage, although this is the first I have heard of that change, but I know the AS5033 is up for review.

Also, im not really sure why we have 1.2x anyways. If the manufactures state the goods are safe to that level, why do we need to factor in 20%?? Im sure the manufactures have considered some type of safe leeway as well.

I would think so too. Manufacturer's spec's tend to err on the side of safety - take the Orion VS CMS2000 for example - if they are exactly identical internally, why the difference in operating voltages?

Also, i know may system running 9 panels in series using standard cabling (2.5mm twin). Are these all defects? If so, how are they still allowed to operate?

I'd need to know the panel spec's (Voc) and it's possible the inspector may have overlooked it. I only became aware of the issue earlier this year, so there's possibly a few I overlooked last year (although the vast majority of systems I inspected last year were with the old $8k rebate - ie: 1kw systems).

Inspector, how do you connect say, an Orion to 9 panels. Common sense would say all in series, because it cannot be divided in 2 strings, and also the working voltage of the panels would be around 300 volts, which is the best operating point for that inverter.

You would also do the EXACT same for the CMS2000 inverters, and they are identical, just spec'd differently. Now, AS5033 would only allow us to legally operate the CMS2000, because 45V x 9 panels = 405V, 405V x 1.2 safety factor = 486V. 486V now exceeds the 450V of the Orion. We know our voltage is never ever gonna be that high, but you still dis allow it.

For the overlooked systems last year. I wouldn't worry about it anyways, the Voc is almost Always 10% less than what the label says anyways. Check this for yourself and see.

At the end of the day, we have Panel manufactures increasing their labeled Voc for safety reasons. We have Inverter manufactures narrowing the operating voltage windows for safety reasons, and we have a 1.2 safety factor. All this combined makes it to insensible and hard for the installer.

I didn't mention inverter operating voltage ranges; I only referred to AS5033 which mentions isolator and cabling voltage ratings. I only made reference to the Orion VS CMS2000 as I am aware they are the same internals, but the manufacturers specify different maximum voltage inputs. AS5033 makes no mention of inverter "multiplication factor" - only isolator and cabling.

For the inverters, I would only be concerned if the Voc was higher than the inverter's rated maximum input, and would not apply the "1.2 times" calculation.

Would you be concerned if the actual Voc was lower than rating on the inverter, but higher than the rating when the 1.2 comes into play?

Not at all Tony. So long as the Voc of the array is within the operating limits of the inverter, then the inverter will be OK. I honestly don't know why the "1.2 multiplier" requirement is applicable to isolators and cable insulation in AS5033 - when I come across such a system, technically I am required to isolate it for safety - however doing so will create the Voc to occur and potentially damage the isolator and cable insulation (although the cable insulation would be susceptible to this daily, and the isolator only when being switched).

Very interesting ..
That brings up another point....

Do you think itll actually be better to turn panels off by using actually shorting them? rather than leaving them open circuit.
Whilst racing solar cars, we found that the safest way to turn the power off is to use the switch to actually short the panels, rather than opening the circuit. At short circuit, the panels will have 0 volts across them as they are current limited. This eliminates the potentially hazardous open circuit voltages.

Also, the Isc is usually only 5% higher than Imp(where the system operates), so there will be no issues to cables heating etc.

Personally, Ild rather have a 5A Isc flowing via my turned off solar system as opposed to a potentially dangerous Voc.

Wat do you think?

Tony, I have thought about this for some time and am undecided which would be the safest option. If current flow contributes to maintaining an arc during short-circuit fault conditions then shorting the array may not be a good idea, but from an isolation perspective, I agree with your methodused for your solar cars.