Wire Sizing for High Current Loads

[modsci]

I hope this is in the right place. If not, please advise or move it.

My previous background has been Control Systems/PLC's etc. I am now involved in a project involving 4 each 4000 HP motors which will pull about 560 amps at 4160 VAC 3 phase 60 Hz.

On a previous similar project, the motors were 3 each 1000 HP at 4160 V, pulling about 140 amps each. Incoming line voltage is 25 KV.

I think the 4000 HP design is going to require individual transformers from 25 KV to 4160. My first concern was wire sizing. For this kind of current, the NEC tables (at least the way I interpret them) tell me that 1000 MCM cable has an ampacity of about 700 or so amps.

Should I instead be considering busways for additional current capacity? Should I lobby for higher voltage motors thus reducing the current, and try to design for single transformers to handle the entire 16,000 HP load? Either way, the size (and cost) is starting to get VERY LARGE to my thinking.

BTW the local utility is also putting restrictions on voltage flicker (3% max), and harmonics per IEEE 519. The voltage flicker means that softstart or reduced voltage start is going to be a must in this case.

Any thoughts out there???

 

[TestBeforeTouch]

I would recommend that you run 2 conductors per phase over the methods you mention. 2 conductors per phase is usually an easier installation and less costly for the ampacity you need.

Based on your stated background, I would also recommend that you get some help from an experianced power engineer. The issues of motor protection, starting method and harmonics should be handled by someone with experiance in those areas.

 

[rbulsara]

By you own admission you are not experienced in medium voltage systems and motor controls. This is a very specialized area which not many experienced power engineers in 600V systems, do not feel comforable dealing with. If not already, please hire some experienced professionals to design this system. Too many things to looked at besides the conductor sizing.

But as for the norm goes you do not want to use greater than 500 MCM cables as the get very hard to handle and dress. At medium voltage even 500MCM may be too stiff. Use parallel condutors if need be.

You may want to use higher voltage than 4160V, even 25kV for 4000HP motors.
A system transient analysis will be required to check if the voltage dip will be acceptable during start.

 

[EEJaime]

I agree with the others that you might want to consider getting an experienced power engineers' input. Also, not knowing your layout, you should consider that 1000 kcmil cable is not an off the shelf item for 5kV cable. Okonite cable corp. for example has a 5000' minimum manufacturing requirement for this, so it could get expensive. Parallel conductors is a serious consideration. Good luck on your project. Just out of curiosity, what are these monster motors powering?

 

[mc5w]

For 4,000 HP motors your supply transformer needs to be at least 10,000 KVA for there to be reasonable voltage drop at the transformer secondary terminals. You would need 2 such transformers plus a third smaller stepdown from 25,000 volts for your other loads so that they get good voltage regulation. You would need a 400 amp service at 25,000 volts to run your whole shebang unless you have considerable other load. You could very easily need a dedicated 25,000 volts line that is normally on a dedicated transformer where the 25,000 volt line originates.

International Exposition Center in Cleveland, Ohio has four 10 MVA transformers that step directly down from 138,000 volts to 4,800 volts ungrounded. Each transformer has an internaly tap changing voltage regulator. The place used to be a manufacturing plant. Sounds like your plant will be up in this power class rating and needs to be on the transmission system, not the subtransmission system.

 

[jraef]

I Agree with TestBeforeTouch as to using parallel conductors. Easier to handle and more readily available as pointed out by EEJaime. That is the general convention in power wiring for large motor loads.

As to the starting method, I suggest a Solid State soft starter over a conventional RV starter. They give you more flexibility as to settings and ramping, which will allow you to better tailor the application to fit the utility's voltage drop requirement. Most MV-RVSS starters also come with motor protection features equal to separate stand-alone Motor Protection Relays as well.

The major drawback to having a single transformer is a lack of redundancy. If a primary fuse were to blow on that one transformer, all of your motors will go down. On the other hand, if all of the motors MUST run at the same time and the process does not work without one anyway, then a single large transformer will be more cost effective. If you do it that way, line the MV-RVSS starters up as a Motor Control Center with bus bars distributing to their line terminals, then your conductor size issues are only on the load side to the motors.

Increasing the motor voltage just to save on conductor costs is not very practical, because you have a 25kV primary which will still require transformers to bring it down to 13.8 (the logical next step for motor voltages in the US). The control equipment alone will triple in cost even though the current is lower, because the switchgear "class" that they are built from jumps from 5kV to 15kV, an entirely different animal.

"Venditori de oleum-vipera non vigere excordis populi"

 

[modsci]

Thanks to everyone who responded to this post. We are now in the process of getting assistance from a power engineer.

Since I posted the original problem, the job scope has greatly changed for the better; now the desire is to have an individual transformer from 25 KV to 4160 V for each motor, and to design for portability so that the entire motor/ switchgear/ transformer may be moved from place to place. Probably more costly considering everything, but...

 

[mc5w]

Your transformer will still need to be about 1.5 to 3 KVA per horsepower. Unlike distribution transformers a substation transformer that steps down from 25 KV to 2400Y4160 volts is likely to be in the range of 10 to 15% impedance to control short circuit current. Smaller distribution transformers, 100 KVA single phase, 300 KVA 3-phase and smaller are usually around 2 to 2.2% impedance for good voltage regulation. 1,500 KVA 3-phase and up distribution transformers are usually 5% impedance for control of short circuit current.

You probablu be better off operating the motors in pairs off of a pair of transformers using a breaker and a half configuration on the secondary. This will give you the low impedance and 2 power sources for relaibility. An added plus is that the transformers will run cooler and be more efficient when loaded at 50% of the forced cooled rating. The forced cooled rating would only be used when 1 transformer or primary feeder is out of service. The big hydroelectric generating at Itiapu ( ? correct spelling ) that Brazil and Paraguay built uses this configuration for reliability and efficiency.

 

[jodha]

I have a 4160V, 500KVA transformer. I need Help to size its primary feeder breaker and the feeder size.

 

[ScottyUK]

jodha,

Your question has nothing to do with this thread. Start a new one for your question and I'm sure there will be some response. Mixing up topics within threads causes confusion.


----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[mc5w]

Also, about the only way that you can connect 1000 KCM conductor and have all of the wire strand conducting is to WELD the connection. You can get weldment lugs for luminum wire from

http://www.sefcor.com.

I do not know if they make weldment lugs for copper wire but they do make bronze substation lugs. The welded aluminum lugs go as small as 3/0 and as large as 3500 KCM for building wire ( AAC conductor ) and 2/0 to 2300 KCM for ACSR cable. They have a lug series that takes 1 conductor per lug and another that takes 2 conductors per lug.

About the only 2 welding processes that work on copper are exothermic welding and VERY HIGH CURRENT tungsten inert gas. It takes about 6 times as much heat to weld copper as what it takes to weld aluminum.

For what copper costs you can put in twice as much aluminum cross section and have an easier time making welded connections.