690.53 for bipolar systems

[ggunn]

690.53 says that we must label the system at the DC disco with MP and max values of current and voltage. For some systems (REFUsol, for example) the only DC disco is at the inverter, and the system is bipolar. That raises some interesting questions as to how these numbers should be reported.

For example, the maximum voltage between the DC+ for the upper array and the DC- of the lower array is twice the temperature corrected string Voc, but I am afraid if I put a label on the inverter that says that the maximum DC system voltage is 932VDC, the inspector is going to freak out.

 

[GoldDigger]

690.53 says that we must label the system at the DC disco with MP and max values of current and voltage. For some systems (REFUsol, for example) the only DC disco is at the inverter, and the system is bipolar. That raises some interesting questions as to how these numbers should be reported.

For example, the maximum voltage between the DC+ for the upper array and the DC- of the lower array is twice the temperature corrected string Voc, but I am afraid if I put a label on the inverter that says that the maximum DC system voltage is 932VDC, the inspector is going to freak out.

Fpr DC, most would consider the maximum voltage to be the highest voltage from on conductor to ground or to a neutral, but I can't find a clear definition covering that.
For an ungrounded system, I would consider it to be the maximum line-to-line voltage, but again I cannot find a clear justification for that.

 

[ggunn]

Fpr DC, most would consider the maximum voltage to be the highest voltage from on conductor to ground or to a neutral, but I can't find a clear definition covering that.
For an ungrounded system, I would consider it to be the maximum line-to-line voltage, but again I cannot find a clear justification for that.

I agree; these systems are both ungrounded and bipolar, hence the 932V.

 

[SolarPro]

The manufacturer lists ?voltages, which would be ?466 in your case.

Note that the REFUsol (now AE 3TL) inverters are not technically bipolar. Here's a description from Bill Brooks (part of public comments RE: 690.41 revisions for NEC 2014):

"To be clear, I have not seen a new truly bipolar array installed in over a decade.

Neither Advanced Energy nor Refusol inverters are bipolar. Advanced Energy [he's talking about the Solaron inverters] looks like a bipolar array when it is not operating, but even then, the poles are not connected together like a typical old-school bipolar array. Refusol takes two independent circuits and connects them together as they connect through a non-isolated connection. Both of these examples are truly non-isolated arrays that receive their reference to ground through the ac transformer."

The main difference between the AE Solaron (now NX series, I believe) and the AE 3TL (formerly REFUsol) product is that the Solaron series products use monopole subarrays that are oppositely grounded, which is consistent with the new NEC 2011 bipolar definitions and requirements found in Article 690. Since the REFUsol products do not have those grounded conductors, it's not bipolar. It's just a unique ungrounded (non-isolated) array configuration.

 

[SolarPro]

This might be a better way of saying it: w/ the REFUsol every PV source circuit is fused, like you would expect in an ungrounded PV system; with the Solaron (or IPC) bipolar inverter the grounded conductors are NOT fused or switched?only the ungrounded conductors are. That's a meaningful difference from a Code perspective.

 

[ggunn]

Thanks for your help. I will show the voltages as ?.

 

[BillK-AZ]

I just installed a REFUsol_012K-024K-UL inverter. When operating, each string has 330 volts, but the inverter LCD display indicates Vdc 660V. The system uses 1,000-volt PV wire, so if there is really 660V somewhere it is not a problem. I am intending to visit the system with an oscilloscope and investigate AC and DC voltages with respect to ground.

 

[ggunn]

I just installed a REFUsol_012K-024K-UL inverter. When operating, each string has 330 volts, but the inverter LCD display indicates Vdc 660V. The system uses 1,000-volt PV wire, so if there is really 660V somewhere it is not a problem. I am intending to visit the system with an oscilloscope and investigate AC and DC voltages with respect to ground.

If the wiring is correct the only place you will see 660VDC (DC1+ to DC2-) is inside the lower compartment of the inverter. Your wiring from the upper array (DC1) and the lower array (DC2) should be kept isolated from each other everywhere, including in separate conduits all the way back to the inverter. It is an ungrounded system, so any DC line to ground measurements you take will be meaningless. On the AC side it's just like any other inverter with 480V 3P4W output; all phases will be 277V to ground.

 

[SolarPro]

I hope AE/REFUsol is providing some quality training because the installation instructions are less than ideal. I had to look hard in the installation manual to find the requirement for separate subarray conduits; the manual also requires balanced subarrays.

Separate conduits back to the inverter and subarray balancing are fundamental bipolar design principles. But the inverter isn't identified as bipolar because it doesn't fit the bipolar definition in NEC 2011. That leaves room for misinterpretation and design or installation mistakes, IMHO.

 

[ggunn]

I hope AE/REFUsol is providing some quality training because the installation instructions are less than ideal. I had to look hard in the installation manual to find the requirement for separate subarray conduits; the manual also requires balanced subarrays.

Separate conduits back to the inverter and subarray balancing are fundamental bipolar design principles. But the inverter isn't identified as bipolar because it doesn't fit the bipolar definition in NEC 2011. That leaves room for misinterpretation and design or installation mistakes, IMHO.

Yes, we ran into a spot of bother on our first installation using REFUsols because although I had it drawn correctly, our subcontractor took it upon himself to alter the design on site for the sake of expediency and combined DC1 and DC2 wiring in the same conduit. To complicate things further, the version of the installation manual they had (since corrected) made no mention of this requirement, so they had recourse to claim that in our negotiations with them to fix it (as if they would have RTFM anyway), and I hadn't put DO NOT COMBINE THESE CONDUCTORS IN A RACEWAY!!! in big red letters on the three line. Lesson learned.