Unbalanced Parallel Conductors 7

[brhodes]

I'm a new engineer for an electric utility with limited experience and have stumbled on something that really puzzles me.

While infrared scanning a padmount transformer located outside of a plant we noticed overheating of conductors. Let me explain the setup.

This is a 500 kVA 208/120 volt transformer that feeds a 1200 amp main panel located inside the plant. There are 3 runs of 500 MCM per phase and Neutral to the main panel. The distance is approximately 30 feet from the transformer to the main panel. The cables are in 3 bundles. One cable per phase and neutral in each bundle(A,B,C,N per bundle). Each bundle is in its own underground conduit to the main panel inside the plant. The load is 80% 3-Phase.

While observing the transformer during the infrared scanning I noticed overheating of the cables on A-phase to the point of discoloring of the copper. The infrared camera showed a significant difference in temp on one conductor on each phase compared to the other two conductors on each phase. I used a clamp-on ammeter and on each phase I got an average of 200 amps more on one conductor than I did on the other two.(These results are from memory. I do not have the written values with me)

We checked for loose connections at the transformer and the main panel. The only loose connection was on one conductor on A-phase which showed signs of overheating (which probably only caused a problem because of the high unbalanced current. The connection was not significantly loose). After checking all of the connections I used a clamp-on ammeter and got the following results.

A1: 125A B1: 7A C1: 96A N1: 43A
A2: 178A B2: 395A C2: 216A N2: 158A
A3: 194A B3: 8A C3: 77A N3: 140A

Prior to making sure we had good connections A-phase showed similar unbalancing as B and C phases. The plant shut off all loads to that transformer and all conductors showed 0 current. The plant then turned on a small three phase load and the amperage was low on the conductors, but they still showed the same proportionality of unbalanced current.

I have read some threads on similar problems and the cause can be anywhere from cable positioning to loose connections. These figures just don't add up. Even the neutral current doesn't seem right. Can we have some circulating currents? Can the unbalanced conductors and high neutral current be related?

We plan on taking off the load tomorrow and verifying all connections are clean and tight on the transformer and panel.

Any help on this matter will be much appreciated!
Thank You

 

[itsmoked]

I believe you are going to be wasting your time on the connectors tomorrow if the actual cable is hot several feet from the connection points.

Have you measured the voltage on each end of the cables?

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[waross]

A phase total 497 Amps
B phase total 410 Amps
C phase total 389 Amps
Neutral total 341 Amps
It looks like you have a lot of single phase loads that are not balanced.
I would atribute some of the neutral current to phase unbalance and the rest to harmonic currents on the neutral generated by the single phase loads.
Do have a lot of lighting on ballasts?
With the current distribution on the phase cables I expect you will find dirty connections. B1 and B2 are surely not well connected but are not carrying enough current to develop heat.
The cables may be loose in a connector. When a cable is fastened into a set screw type connector, and then twisted, it will untwist the strands slightly and leave the cable loose in the connector.
When I install stranded cables I tighten the setscrews firmly. The I twist the cable clockwise and counterclockwise several times. Then I tighten the set screws (which are now loose). I repeat this until the set screws stay tight when the cable is twisted. I have seen many installations where the set screws were tightened once and the cables then inadvertently twisted and the cables left very loose in the connectors.
I have demonstrated this to electricians several times. Seven strand cable in the range of #8 to #4 (if you find it in 7 strand) can be depended on for a good demonstration. Let the electrician tighten the set screw as much as he wants. Then give the cable a slight twist and it will fall out of the connector.
I hope this helps.
I'm waiting to hear what you find.
yours

 

[itsmoked]

I may then stand corrected. Good info there waross(as usual). Now I'm very interested in what brhodes finds tomorrow.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[jghrist]

If you had a similar unbalance on A phase before fixing the loose A phase connection, and then the A phase unbalance went away, it seems reasonable to conclude that the unbalance on B and C phases are because of the connections. Probably not apparent without twisting the strands as waross suggests.

I also agree that the neutral current that can't be explained by unbalanced phases is probably caused by triplen (3rd, 6th, 9th, etc.) harmonics.

 

[wareagle]

I agree with warross regarding the unbalanced single phase loads. But look at the reading:
A1: 125A B1: 7A C1: 96A N1: 43A
A2: 178A B2: 395A C2: 216A N2: 158A
A3: 194A B3: 8A C3: 77A N3: 140A

Phase B has a problem. The conductors are connected together at the the transformer and the main switch. There has to be a bad connection. Check the connectors.
I have found corrosion under the pad that caused this type of current division. Same goes for phase C.

 

[brhodes]

Thank you all for your response!
To answer itsmoked's question, I have checked the voltage at both ends of the cable and its fine. Also, the cable is only significantly hot at the A-Phase connector on the transformer. I do believe there was a bad connection there. As you move away from that connector the cable becomes cooler, only warm to the touch due to high amps. Same is true for the other conductors; they are only warm due to high amps.

To respond to waross: the customer informed me that there was not much single phase load on this panel. As for how much is "not much", I will find out tomorrow. On the other hand, the single phase load is lighting, more than likely lighting on ballasts. Also, we checked the screw connectors on the panel and they were tight. We will take a closer look tomorrow to verify. As for the connectors on the transformer, they crimp on the cable and bolt on the transformer plate. We checked the connections and all seemed tight except for one cable on A-phase which would explain the discoloring, but as wareagle noted, they may be corroded. I will find out tomorrow.

waross and jghrist--
The harmonics could possibly be the cause of the high neutral current in addition to the single phase load. I may attach a meter and record the load a few days to measure the harmonics.

Correct me if I'm wrong, but even with single phase load, the overall phase amps don't seem extremely out of balance. Can harmonics make that much difference to put 341 amps on the neutral?

Thanks for the help!

 

[waross]

Hello brhodes
We had a thread recently that may have some bearing. It was a problem with crimp-on motor lugs that had been dipped and baked before the lugs were installed. The varnish inside the lugs was causing failures.
The point is, in this instance, the cables may not have been cleaned properly before the lugs were installed.
Twisting a cable that is connected to a breaker is sometimes a challenge, but with bolt-on lugs, you can use a wrench on the connector body to twist it back and forth to get the strands seated and the set screws tight. I would enter the cable and tighten the set screws the first time and then unbolt the lug from the bus to twist it. Try one to see how much difference it makes and then you decide whether to do them all that way.
I often use penetrox on the set screws, the cables and the connections.
Re the current; The maximum phase difference is 108 amps. A quick and dirty estimate is that about 2/3 of your neutral current may be harmonic.
respectfully

 

[LionelHutz]

Here's an interesting and somewhat relevent thing I read in an Electrical Apparatus article. In a paper plant they found that if they only re-tightened suspect connections that 1 in 5 would show an improvement while if they took the joint apart and cleaned it that 92% would show an improvement.

 

[waross]

Hello LionelHutz
Thanks for the info. I have never kept statistics but that sounds reasonable.
respectfully

 

[brhodes]

LioneHutz,
Thanks for the information. I plan on having all of the connections taken apart and cleaned. I will let you all know the outcome.
Thanks!

 

[brhodes]

Readings prior to cleaning connections:

A1:420A B1:552A C1:350A N1: 73A
A2: 13A B2: 7A C2:130A N2:189A
A3: 80A B3: 10A C3:235A N3:195A

All voltages looked good.
The 1, 2, & 3 notations on the phases do not associate with a particular conductor, they are just used for representation in this discussion.

We started cleaning the connections on A-phase. After disconnecting the bolt-on lugs the connections showed a lot of pitting, I would assume would be due to overheating. We replaced the bad portion of the three cables on A-phase. During the process of taking loose the B-phase connections we noticed to of the connections very loose which would make sense why the other cable was carrying most of the load. C-phase connections seemed okay, but we disconnected and cleaned them as well. The connecting pads on the transformer seemed very rough like they had a textured finish on them. I not sure if that had any thing to do with this problem, but we smoothed out the rough surface. After tightening all connections, we staged in the load and checked the loading of the cables during that process. After all loads were connected, including the entire 1-phase load, these were the final readings.

A1:200A B1:219A C1:183A N1:13A
A2:206A B2:170A C2:158A N2:26A
A3:196A B3: ? A C3: ? A N3:26A

We were unable to check B3 and C3 safely. The majority of the neutral amps picked up after the single phase load was added.

I can see how the connection problems solved the unbalanced conductor issue, but it seems very unusual that it would also solve the high neutral current issue as well.

I want to thank everyone who replied, each of you were a big help! Thanks! The connection issue seemed to have solved the problem. If any of you has an explanation for my last statement, I would love to here you thoughts.

 

[itsmoked]

Good on you brhodes. I do stand corrected! And thanks for the great feedback...

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[jghrist]

Maybe the bad connections caused enough of a voltage unbalance to result in a large phase unbalance. A small voltage unbalance on motors will result in a large current unbalance.

 

[brhodes]

jghrist,
While the load was unbalanced I checked the voltage at the secondary terminals on the transformer and they were within 1-1.5 volts of each other. After we corrected the connection problems they are within a .5 volt of each other.
Correct me if I'm wrong jghrist, but it wouldn't seem that 1-1.5 volts would make that large of a current unbalance on motors?
Thanks!

 

[jghrist]

The 1-1.5 volt (~1%) voltage unbalance could result in 8% current unbalance.

 

[bigbillnky]

brhodes

I would like to add to the point waross has made concerning terminations. It may/will be difficult to twist large conductors in order to properly seat the conductor strands in the lugs. I have found that tightening the lugs and simply working the cable back and forth lightly works well. Repeat the process a few times until the strands have fully separated and seated in the lug, then torque to spec.

As waross pointed out, almost all field electricians are unaware of this problem.

Bigbillnky,C.E.F.....(Chief Electrical Flunky)

 

[edison123]

An useful thread to learn from.

brhodes - a star for starting this and taking it to a logical end.


* Basically, I would like a full-time job on part-time basis *

 

[waross]

Congratulations on a successful excersise.
Reality check;
Before cleaning the neutral current was 457 amps.
After cleaning the neutral current was 65 amps.
The load before cleaning and after cleaning was significant ly different. Possibly the computer department was online before and didn't go back online immediately after the current was restored.
Before cleaning the current on "C" phase was 202 amps more than the current on "A" phase.
Before cleaning it looks like you had an extra 200 amps on "C" phase and over 450 amps on the neutral.
There is a large single phase load that was not present after cleaning. This single phase load is generating harmonics.
Reality check #2 Unbalanced voltages will cause disproportionate current unbalances, but only on the phases. Three phase motors have no connection to the neutral and have no effect on the neutral current. Only loads connected line to neutral affect the neutral current. The 60 hz. component of the neutral currents cancel each other out and the 3rd. harmonic component of the neutral currents adds.
Again an indication that the loads were different before and after cleaning.
These statements are based on plant equipment in good condition. If you have a motor or transformer with an internal fault to ground it may cause heavy neutral currents.
If the large unbalanced neutral current returns, start checking motor currents. If the voltages are unbalanced the currents will be unbalanced but in the same pattern. The same phase will be high on all motors. If you find a motor with currents that don't fit the pattern, check it for internal problems. First step, turn it off and see if the neutral current is reduced.
Look at the grounding for signs of unusual connections and ground currents flowing in the neutral. Is there any cathodic protection equipment nearby?

Reality check #3 When suspecting that large current differences or large voltage differences are caused by bad connections, ask "How much voltage and current is involved. How much energy (watts) is that. Is the heating reasonable considering the amount of watts I think I'm loosing?"
If a single conductor has a bad connection it will generate a lot of heat and do a lot of damage. If one or two cables of a parallel feed have bad connections it will divert the current to the cable with the good connection and it will over load but little or no damage will be done at the bad connections.
And BTW brhodes you deserve a star for a job well done.
respectfully

 

[brhodes]

waross,
Thanks for the responses.
To respond to your reality checks:
I agree with the loads being different and there being a single phase load that was not on after cleaning. There may be a logical reason for the difference in loading. As they put their panel back online, a lot of motors had to be manually started meaning that some of the motors may not have been running when I took these measurements. As for the single phase load, I also agree that there seems to be a single phase load that was not on. The plants electricians assured me that all of the single phase loads were on. I am suspicious about that. I plan on installing a recorder to monitor for load changes, high neutral current and harmonics.
Reality check #2:
Thanks for the input on this. One thing I failed to mention earlier was that while measuring the neutral current before cleaning I measured N2 to be 189A and N3 to be 195A, but when I clamped around both conductors (N2 & N3) I got 74A. Almost like circulating current.
Not sure about the cathodic protection equipment. What could this equipment generate?
Reality check #3:
I completely agree. One or two cables did have a bad connection and diverted the current to the cable with the good connection, but there was little damage to the cable carrying most of the current due to it being close to operable limits.

I will keep you all posted with my findings.
Thank you!