Uneven current ratings.

[rhamblin]

During a recent Infrared thermography scan of our plant, we came across some unusual readings from one of our main Switchgear units.

At our plant, we have a 24,900 volt phase to phase/14,400 volt phase to ground transformer dropping voltage down to 480V 3-phase. After this transformer, 11 conduits (parallel runs of 500 MCM) run into a Surge device cabinet.



From this Surge protective cabinet, another 11 parallel runs of 500 MCM feed to the 480 volt switchgear.

After finding the hot spot we decided to put a amp clamp on the hot wire, it was around 315 amps. So then we decided to measure a few of the other wires on the same phase. We couldn't check all of them because of the bundling of the wires. Here are some of the measurements that we encountered on these A phase parallel wires:

• 165 amps
• 137 amps
• 69 amps
• 105 amps
• 138 amps
• 111 amps

Now for my question, if all the wires are in parallel, shouldn't they have pretty close to identical resistances and therefore identical currents?

I don't think it would matter, but apparently the Surge protection devices in this panel, have been "burned out" since before I started 6 years ago. I'm looking for any information here. In my mind we have a lot of loose connections or bad crimps that is causing more current to be dispersed on some of the conductors.

 

[Ingenieur]

how long is the run?
the only one Inwould check is the 69

there are variables
purity of the Cu
how the ends were prepared, cleaned, compression force, etc
tightness of bolted connections

should be the same in a perfect world
+/- 15% around the avg would not concern me

 

[GoldDigger]

...
...
Now for my question, if all the wires are in parallel, shouldn't they have pretty close to identical resistances and therefore identical currents?

As long as the wires are really routed to result in a very close path length, you would expect the resistances to be substantially equal.
However, especially if the wires are short (under, say, 20 feet), the biggest part of the resistance will be the contact resistance at the terminals at both ends.
Since the overall resistance is so low even a small difference in contact resistance can lead to a large current imbalance.

Note also that current is not dispersed along the way. Instead the voltage drop increases along the way, leading to reduced current to allow the total voltage drops to be equal across all the parallel wires.

 

[rhamblin]

how long is the run?
the only one Inwould check is the 69

there are variables
purity of the Cu
how the ends were prepared, cleaned, compression force, etc
tightness of bolted connections

should be the same in a perfect world
+/- 15% around the avg would not concern me

The conduit length is 3 feet, so conductors are maybe 10-15’.

 

[Ingenieur]

The conduit length is 3 feet, so conductors are maybe 10-15’.

so even a 1 ft difference due the bending, wrangling onto lugs, location, etc is 7-10% delta
if the run is 8' even larger
and that assumes the lugs are equal, unlikely

 

[rhamblin]

...7-10% delta

delta T? Which is the difference of two things right?

 

[Ingenieur]

delta T? Which is the difference of two things right?

the difference in R between a conductor say 9' long vs 10'
the 10 has 11% more R so will flow less current
a delta or difference of 11% R between the two

 

[kwired]

The shorter the overall length, the bigger a minor difference is in relation to the whole picture.

6 inch difference in length of one conductor in a ten foot run is going to have much larger impact on resistance/current division through the parallel set then a six inch difference in an individual conductor of a 100 foot run.