Utility Scale Battery Storage

[bkg73123]

New to this industry. Not my job to size the equipment but from time to time I am required to come up with indicative pricing without proper engineering. I do not have an electrical degree but do have some knowledge. My questions are related to wire sizing related to parallel runs hooking up a Xfmr to an Inverter that connects to the battery storage containers. The example one line diagram below shows a Xfmr supplying 4 battery containers with nine 750kcmil conductors and one 350kcmil gnd conductor per each battery. It is my understanding that the gnd conductor must be sized to handle the same amp rating as the OCPD, but that doesnt seem to be the case here possibly due to derating which I am not that familiar with. Assuming 90deg CU with a size of 350kcmil the gnd can handle 350A. Yet with the battery having a 1,632kva rating (1,963FLA), I calculate nine 750kcmil conductors (3 parallel 3 phase runs) would only handle 1,605A (without derating). So, I am confused because it looks to me like 3 runs of 750kcmil is not adequate and neither is the single 350kcmil gnd conductor. I am also foggy on if these parallel runs are multi taps from a single CB or if each run has its own CB. Also, I am pretty sure going parallel isnt as simple as dividing the amps because you may only get 75-80% additionally ampacity on parallel runs. Some insight into this would be very much appreciated. Thank you

 

[bkg73123]

Looking at it from another perspective if the Xfmr is 4.4mva that means it can only provide ~5,293A divided by 4 battery containers would be 1,594FLA per battery container in which case 3 parallel runs of 750kcmil would be adequate before any derating calcs.
But I am not sure if this is the correct way to look at it and still dont understand why a single 350kcmil gnd is spec'd unless that gnd is Aluminum. A 350kcmil aluminum would be correctly sized for 1600Amp OCPD. But would that be correct to use an Aluminum ground with CU conductors?

 

[winnie]

Lots of stuff going on in the question. I'm just going to focus on the parallel conductor issue.

Under NEC rules, parallel runs are calculated using simple addition of capacity of individual runs. If you have 3 conductors with a 475A rating in parallel, then then net ampacity is 1425A.

Adjustment rules might apply if all of these conductors are in a single conduit. When you have 9 conductors in a single conduit you have to adjust the ampacity by 0.7. To complicate this, inside the conduit you are permitted to start your adjustment from the maximum conductor temperature rating.

The 750 kcmil conductors have a 75C rating of 475A, and a 90C rating of 535A. In your application you probably have 75C terminals, and have to use the 75C rating of 475A per conductor or 1425A net.

If all these conductors are in a single conduit, you have to apply the 0.7 factor, but can work from the 90C value, or 3 * 535 * 0.7 = 1124A net.

What is the OCPD on the cable between the battery containers and the transformer?

-Jon

 

[bkg73123]

Lots of stuff going on in the question. I'm just going to focus on the parallel conductor issue.

Under NEC rules, parallel runs are calculated using simple addition of capacity of individual runs. If you have 3 conductors with a 475A rating in parallel, then then net ampacity is 1425A.

Adjustment rules might apply if all of these conductors are in a single conduit. When you have 9 conductors in a single conduit you have to adjust the ampacity by 0.7. To complicate this, inside the conduit you are permitted to start your adjustment from the maximum conductor temperature rating.

The 750 kcmil conductors have a 75C rating of 475A, and a 90C rating of 535A. In your application you probably have 75C terminals, and have to use the 75C rating of 475A per conductor or 1425A net.

If all these conductors are in a single conduit, you have to apply the 0.7 factor, but can work from the 90C value, or 3 * 535 * 0.7 = 1124A net.

What is the OCPD on the cable between the battery containers and the transformer?

-Jon

Thank you Jon. The cables run in an above ground cable tray between the Xfmr and Batteries. This is all I can find in regards to OCPD between the Xfmr and Battery. It is on the Battery inverter. Is this saying the breaker should be sized at +/- 5% of 1,632kVA? Since 1,632kVA is the apparent power of the Battery/Inverter Container?

 

[bkg73123]

Spinning my wheels a little here while I wait for reply... If 1,632kVA is apparent power equates to ~1,963FLA with 2,000A OCPD. Would require 5 parallel runs of 75deg 750kcmil CU totaling 2,375 ampacity. However, since all are running in same cable tray I need to derate by using the 90deg ampacity which is 5*535*.7= ~1,873A which is not enough so I would need to go to 6 parallel runs to achieve 2,247A.

 

[winnie]

1) I don't really know the rules for ampacity adjustment in cable trays. The factors depend on the design of the cable tray (eg ventilation) and the way the cables are installed and spaced apart.

2) The trip ratings look like they are for inverter electronic self protection, and don't say anything about OCPD.

Jon

 

[ron]

Double check if you can get away with primary only protection for the Delta-wye transformer, because if you need secondary protection, that will help you understand what ampacity the feeders to the inverters get sized at.

 

[bkg73123]

Do the current carrying conductors have to be sized to the OCPD or the FLA? Sometimes there can be a big difference when getting up into higher amp systems.

 

[jaggedben]

If covered by the NEC, current carrying conductors have to be protected by the OCPD. Both need to be sized to 125% of the BESS output or input, whichever is greater. Are the BESS being set to charge and output at full nameplate or is there a design decision to limit their output?

I presume that the BESS are listed Power Control Systems. The code allows conductors to be sized to a PCS setting. (705.13) Otherwise, if this install is covered by the NEC, I don't see how the ampacity of the output can be less than the rating of the BESS.

Note: Technically the kVA rating is irrelevant. The rated continuous output current is what the code goes by. I was wondering if perhaps the equipment has a lower amp rating than the max kVA implies, but that could typically only explain about a 10% difference assuming the actual voltage was at the max above nominal and the current stayed at the max. That would be rare, but with a certain brand name on the equipment I wouldn't be surprised if someone is pushing the envelope...

I will add that designing this sort of installation is well above my pay grade, but the code still says what it says.

 

[bkg73123]

If covered by the NEC, current carrying conductors have to be protected by the OCPD. Both need to be sized to 125% of the BESS output or input, whichever is greater. Are the BESS being set to charge and output at full nameplate or is there a design decision to limit their output?

I presume that the BESS are listed Power Control Systems. The code allows conductors to be sized to a PCS setting. (705.13) Otherwise, if this install is covered by the NEC, I don't see how the ampacity of the output can be less than the rating of the BESS.

Note: Technically the kVA rating is irrelevant. The rated continuous output current is what the code goes by. I was wondering if perhaps the equipment has a lower amp rating than the max kVA implies, but that could typically only explain about a 10% difference assuming the actual voltage was at the max above nominal and the current stayed at the max. That would be rare, but with a certain brand name on the equipment I wouldn't be surprised if someone is pushing the envelope...

I will add that designing this sort of installation is well above my pay grade, but the code still says what it says.

Will assume full nameplate. So for example, if the BESS is rated at 1,323 FLA, the amperage I would use to size OPCD and conductors would be 1,323 x 1.25 = 1,654 Amps. Now this will require multiple parallel runs so does each run get its own OCPD typically or would this be multi tapping a single OCPD?

 

[ron]

For the projects like this that I've done, I've provided a single main breaker for the transformer secondary, with distribution breakers to each BESS inverter.

Note 2 from table 450.3(B) makes the opportunity of multiple breakers without a single main, a Pain.

 

[bkg73123]

If covered by the NEC, current carrying conductors have to be protected by the OCPD. Both need to be sized to 125% of the BESS output or input, whichever is greater. Are the BESS being set to charge and output at full nameplate or is there a design decision to limit their output?

I presume that the BESS are listed Power Control Systems. The code allows conductors to be sized to a PCS setting. (705.13) Otherwise, if this install is covered by the NEC, I don't see how the ampacity of the output can be less than the rating of the BESS.

Note: Technically the kVA rating is irrelevant. The rated continuous output current is what the code goes by. I was wondering if perhaps the equipment has a lower amp rating than the max kVA implies, but that could typically only explain about a 10% difference assuming the actual voltage was at the max above nominal and the current stayed at the max. That would be rare, but with a certain brand name on the equipment I wouldn't be surprised if someone is pushing the envelope...

I will add that designing this sort of installation is well above my pay grade, but the code still says what it says.

View attachment 2567001

It was my understanding that this table was for calculating the MAX OCPD allowable, but you could elect to use an OCPD equal to the FLA if you wanted? Is the 125% jaggedben referring to the 125% in this table or is it based on another requirements somewhere else in the NEC?

 

[wwhitney]

It was my understanding that this table was for calculating the MAX OCPD allowable, but you could elect to use an OCPD equal to the FLA if you wanted? Is the 125% jaggedben referring to the 125% in this table or is it based on another requirements somewhere else in the NEC?

See NEC 705.12.

Cheers, Wayne

 

[ron]

It was my understanding that this table was for calculating the MAX OCPD allowable, but you could elect to use an OCPD equal to the FLA if you wanted? Is the 125% jaggedben referring to the 125% in this table or is it based on another requirements somewhere else in the NEC?

The table establishes the maximum. You can put any size equal to / smaller than the maximum

 

[jaggedben]

See NEC 705.12.

Cheers, Wayne

Or 706.20(B) and 706.21(B).

 

[jaggedben]

Or 706.20(B) and 706.21(B).

Or rather, for 2023 or 2020 NEC references (not 2017), see 706.30 and .31 respectively.

 

[GoldDigger]

Will assume full nameplate. So for example, if the BESS is rated at 1,323 FLA, the amperage I would use to size OPCD and conductors would be 1,323 x 1.25 = 1,654 Amps. Now this will require multiple parallel runs so does each run get its own OCPD typically or would this be multi tapping a single OCPD?

The code recognizes strictly parallel conductor (single common end point at each end, therefore no individual OCPD) as being able to carry the sum of the rated ampacities. The very strict limitation on identical length and material, including thermal characteristics, for each individual conductor minimize the possibilities for unequal sharing as long as you pay attention to termination resistance at the ends. But if one conductor fails for some reason, the single OCPD will not prevent a cascading failure of all the others, with potentially severe consequences along the way.

Giving each run its own OCPD would make them NOT parallel conductors in the NEC sense, but might be seen as violating the rule against paralleling OCPDs to get a single larger amperage OCPD. I have not seen this particular situation discussed in the context of building wiring. It may occur more commonly within a listed device, in which case UL rather than NEC would apply.

 

[pv_n00b]

The conductors from the XFMR to the BESS are all XFMR secondary conductors that have to be sized using 240.21(C) as well as sized to the BESS output. The 240.21(C) requirements are going to depend on the length and location of the conductors and the rating of the OCPD in the BESS.
Since the XFMR represents a separately derived system (NEC 250.30) the grounding conductor is a supply-side bonding jumper and is not an EGC sized to an OCPD rating. It is based on the size of the current carrying conductors from the XFMR to the BESS. The XFMR being much larger than a single BESS is going to dominate a lot of the calculation. Then there is the additive component. Each circuit from the XFMR to a BESS is exposed to fault current from the XFMR plus (n-1) times the BESS fault current.

 

[pv_n00b]

For the projects like this that I've done, I've provided a single main breaker for the transformer secondary, with distribution breakers to each BESS inverter.

Note 2 from table 450.3(B) makes the opportunity of multiple breakers without a single main, a Pain.

View attachment 2567001

This is an MV XFMR so we need to use the MV table (NEC 450.3(A)). The MV XFMR table is more liberal with the protection requirements. The notes are pretty much the same though.

 

[bkg73123]

Taking into account what pv_n00b stated I am thinking this may be getting too muddy for me. I have made a calculator to use to get me in the ballpark of sizing these conductors for indicative pricing when engineering isn't available. I took the above one line provided by a partner engineering firm and plugged in the transformer specs, etc. and I come up with slightly different wire sizing than they did. If anyone cares to look it over it would be much appreciated. The white cells are user input. Currently the GND wire on the secondary is based on OCPD and if there is a single GND wire that must mean they are using a shared conduit which according to my calcs would reduce the ampacity of the conductors by 50% making 9-750kcmil inadequate. So maybe I shouldn't be applying that? I think the whole basis of what I have built is probably wrong but here it is if you wanna take a look.

https://rapidshare.io/2mIv/BESS_Calc.xlsx