Induction heating/ Eddy currents? .. or something else?..

[PhilBob]

Hi,

I'd like to share an issue that we're currently resolving, and I'd greatly appreciate your input on potential causes and solutions.

Our team is relatively new to electrical thermography, and we've recently come across an anomaly at the substation where one of the conduits carrying a supply line for the feeder exhibits a high temperature, approximately 70°C, while the other conduits for Phase B and C maintain a normal ambient temperature.
Each conduit contains three wires and one ground wire, all with a size of 200mm².

Additional details:
• No heating or anomalies were detected on the load side. The issues appear to be concentrated upstream.
• No observed ground faults on accessible sections of the line side.

Your insights would be greatly appreciated.

WeTransfer | Send Large Files Fast

The simple, quick and secure way to send your files around the world without an account. Share your files, photos, and videos today for free.

we.tl

 

[jaggedben]

If those conduits are steel then the different phase conductors need to be distributed evenly between the conduits to avoid inductive heating.

 

[ron]

Hi,

I'd like to share an issue that we're currently resolving, and I'd greatly appreciate your input on potential causes and solutions.

Our team is relatively new to electrical thermography, and we've recently come across an anomaly at the substation where one of the conduits carrying a supply line for the feeder exhibits a high temperature, approximately 70°C, while the other conduits for Phase B and C maintain a normal ambient temperature.
Each conduit contains three wires and one ground wire, all with a size of 200mm².

Additional details:
• No heating or anomalies were detected on the load side. The issues appear to be concentrated upstream.
• No observed ground faults on accessible sections of the line side.

Your insights would be greatly appreciated.

WeTransfer | Send Large Files Fast

The simple, quick and secure way to send your files around the world without an account. Share your files, photos, and videos today for free.

we.tl

It is hard to figure out which photo is the issue. Some show the same color insulation for all conductors in a conduit (with a grounding conductor), which is a problem for inductive heating when the conduit is ferrous (as mentioned by jaggedben)

 

[synchro]

As others have said, the isolated phase (isophase) configuration of the conduits can cause heating in ferrous conduits. From the thermal images, it looks like the green equipment grounding conductors that go through the conduits are conducting a significant amount of current. I'd suggest measuring their currents with a clamp meter. Also, if you can, put the clamp around the three hot wires and ground wire going through each conduit, and compare that with just the three hot wires clamped. That way you can see if the ground current is adding or subtracting to the magnetic field. Another thing would be to put a clamp around the conduits themselves and get a current reading, if this can be done.

 

[PhilBob]

Thank you for all your responses.

Yes, I've got all the clamp meter readings before and summarized as this-

• Current readings on phases remain symmetrical, ranging from 100 to 190 Amps.
• Current reading on the ground wires is 100 Amps and above.

Clamp meter readings

7 files sent via WeTransfer, the simplest way to send your files around the world

we.tl

I'll try tomorrow if I can acquire an i-flex loop to get the other current readings.

 

[winnie]

Our team is relatively new to electrical thermography, and we've recently come across an anomaly at the substation where one of the conduits carrying a supply line for the feeder exhibits a high temperature, approximately 70°C, while the other conduits for Phase B and C maintain a normal ambient temperature.
Each conduit contains three wires and one ground wire, all with a size of 200mm².

Your insights would be greatly appreciated.

WeTransfer | Send Large Files Fast

The simple, quick and secure way to send your files around the world without an account. Share your files, photos, and videos today for free.

we.tl

Can you confirm that this is an _isolated phase_ installation, meaning phase A in one conduit, phase B in another, phase C in a third? As has been mentioned above, isolated phase installations are a serious problem with metallic conduit.

Can you use an image upload site that doesn't require registration to _view_ the images?

-Jon

 

[PhilBob]

It is hard to figure out which photo is the issue. Some show the same color insulation for all conductors in a conduit (with a grounding conductor), which is a problem for inductive heating when the conduit is ferrous (as mentioned by jaggedben)

Yes, it's a rigid steel conduit.

It's possible that they are all susceptible to inductive heating, but what's puzzling is why it appears to be affecting Phase A exclusively..

 

[PhilBob]

Can you confirm that this is an _isolated phase_ installation, meaning phase A in one conduit, phase B in another, phase C in a third? As has been mentioned above, isolated phase installations are a serious problem with metallic conduit.

Can you use an image upload site that doesn't require registration to _view_ the images?

-Jon

Indeed, it's currently isolated to a specific phase arrangement, as you mentioned. We're considering converting the cable tray to a ladder type as a potential solution to address the problem.

 

[PhilBob]

If those conduits are steel then the different phase conductors need to be distributed evenly between the conduits to avoid inductive heating.

That's precisely our plan as well, once we transition to the cable tray layout.

 

[winnie]

Ok, the root problem is the isolated phase arrangement.

The current flowing in each conduit is not balanced by return current in the same conduit, and thus there is a significant magnetic field surrounding the phase conductors and extending out into (and between) the various conduits. This is making the conduits themselves the secondary of a single turn transformer.

The only thing confusing is why only one of the 3 conduits is showing significant heating, because all 3 conduits should be subject to similar induction effects. This might simply be some high resistance preventing current flow in 2 of the conduits; I notice that the junction box is painted and that the paint extends in through the holes.

It sure looks like the ground lugs have seen some significant heating as well. The high current induced in the ground wires is another symptom of the isolated phase arrangement.

-Jon

 

[winnie]

That's precisely our plan as well, once we transition to the cable tray layout.

You don't need cable tray; you simply need the wires re-arranged so that each conduit has 1 of phase A, B, C along with the EGC. It would bug the hell out of me because of the messed up color coding, but if you properly identify the cable ends, I think you could simply re-terminate the existing cables to get the proper arrangement.

There appear to be other problems with the installation; the conduit appear a bit small for the conductors, as does the junction box. But I've not done any of the appropriate calculations. The fact that only 1 conduit showed excessive heat suggests high resistance in the conduit connections. So switching to cable tray might be the best choice to fix all of the problems.

-Jon

 

[synchro]

I'll try tomorrow if I can acquire an i-flex loop to get the other current readings.

If you get the flex loop, another test would be put this loop around all three ground wires if it can be done. That would give you the net current going from the ground wires to the panel. If the current reading is small, then that would mean the currents are circulating between the ground wires.

You mentioned on another thread a PF correction bank upstream. If the capacitors are in a wye configuration, is their common connection tied to a neutral, to ground, or left open?

 

[PhilBob]

You don't need cable tray; you simply need the wires re-arranged so that each conduit has 1 of phase A, B, C along with the EGC. It would bug the hell out of me because of the messed up color coding, but if you properly identify the cable ends, I think you could simply re-terminate the existing cables to get the proper arrangement.

There appear to be other problems with the installation; the conduit appear a bit small for the conductors, as does the junction box. But I've not done any of the appropriate calculations. The fact that only 1 conduit showed excessive heat suggests high resistance in the conduit connections. So switching to cable tray might be the best choice to fix all of the problems.

-Jon

Thanks a lot sir Jon.
Your explanation has provided us with a comprehensive understanding of the situation.
We're now more confident to proceed with the corrective steps we've prepared.

 

[PhilBob]

Just to share with everyone in the group,
I've conducted again a series of current measurements on various sections of the feeder, and here's the rundown of my logs:

Current on each phases:
A - 198 Amps
B - 202 Amps
C - 195 Amps

Current on conduit:
A - 151 Amps
B - 90 Amps
C - 80 Amps

Current on the ground wires:
From conduit A- Stabilizes at 2 amps but occasionally spikes up to 170 Amps.
From conduit B- Maintains at around 85 Amps but ocassionally spikes up to 145 Amps.
From conduit C- Maintains at around 130 Amps.

These data might offer some insights into why the Phase A conduit consistently reaches a relatively high temperature.
I would appreciate it if you could share your thoughts once more.

Thank you.

 

[winnie]

These data might offer some insights into why the Phase A conduit consistently reaches a relatively high temperature.
I would appreciate it if you could share your thoughts once more.

Thank you.

My hunch is that due to small resistance differences the induced balance current is flowing more on the copper EGC in conduits B and C, more on the steel conduit on A. This is giving the different temperature readings. But I would not spend much time trying to figure out that puzzle, because the required fix is evident; however do double check the conduit continuity when you are making your wiring changes.

-Jon

 

[PhilBob]

Thanks again sir Jon,
We'll do continuity test as well on the new installation.

 

[PhilBob]

If you get the flex loop, another test would be put this loop around all three ground wires if it can be done. That would give you the net current going from the ground wires to the panel. If the current reading is small, then that would mean the currents are circulating between the ground wires.

You mentioned on another thread a PF correction bank upstream. If the capacitors are in a wye configuration, is their common connection tied to a neutral, to ground, or left open?

We obtained a set of flexible loops for current readings, and I attempted to measure the current on all three ground wires simultaneously.
And yes, the total current was quite low, ranging from 4 to 10 amps only.
What does exactly imply if the currents are just circulating between these ground wires?

I'll need to double-check the capacitor bank configuration and other connections at our plant during our upcoming activity. I'll keep you updated..
Thanks

 

[synchro]

We obtained a set of flexible loops for current readings, and I attempted to measure the current on all three ground wires simultaneously.
And yes, the total current was quite low, ranging from 4 to 10 amps only.
What does exactly imply if the currents are just circulating between these ground wires?

One way of looking at it is this: each conduit is the core of a transformer, where the phase conductor is the primary winding and the ground wire is a secondary winding. These three ground wire secondary windings are shorted together at each end on a bus or metal enclosure, kind of like a wye secondary with all three output phases shorted together. In other words, a transformer with all of the secondary output turns shorted. The fact that you measured a small current when clamping around all three wires means that there's negligible current flowing through the ground wires other than what's coupled magnetically to the ground wires from the three "hot" phase wires in the same conduits.

The conduits themselves are conductors as well as the ground wires, and so they also have currents flowing along them that are induced from the currents in "hot" wires. These three conducting conduits are also"shorted" at each end by their connection to a common box or enclosure. As Jon (winnie) has mentioned above, the resistance of the conduit and wire connections can effect how much current flows through different conduits and ground wires. The impedance of the conduits and ground wires is very low, and so it wouldn't take much series resistance in the connections to affect the balance of currents between the different ground wires and conduits.

In addition to the induced currents flowing lognitudinally through the conduits, there will also be eddy currents induced in the conduits. This produces additional heating. Hysteresis losses in ferrous metal conduits also causes heating.

 

[PhilBob]

Hi everyone,

I'm pleased to share that we have successfully completed the corrective work, which primarily involved replacing the conduits with properly sized cable trays and repositioning the wires to ensure the correct phase arrangement.
After the testing with load (nearly 200 amps), all wires, cable trays, and grounding terminals are good and are now maintaining just an ambient temperature.

I want to thank each one of you for your valuable insights and the detailed explanation. Your contributions greatly assisted us in validating our corrective plans and even incorporating additional measures.

Once again, thank you all.

 

[tom baker]

Thank you for your follow up on corrections!