Power Outage...Power Restored...Damage!

[wirenut1980]

Ok, something bothering me here that I can't explain...

I have seen this quite a few times, both single phase customers and 3 phase customers. Power outage caused by various faults and clearing of the faults is by single phase devices and 3 phase devices. Durations of outage range from seconds to hours.

Now I can understand how 3 phase equipment if unprotected can be single phased and there can be damage, and I can see if transmission lines fall into distribution lines how that can cause damage...

But what about a small commercial customer whose power is out for hours, served from a single phase distribution line, 400 amp service, power comes back on and wham, the compressor for a large refridgerator blows up and catches fire!

Or what about a 3 phase industrial customer who has an outage cleared by a 3 phase breaker, and when power comes back on, multiple motor starters fail?

These kinds of things have me scratching my head, because I just can't wrap my mind around how outages and power restored can destroy equipment in some circumstances.

Sorry for blubbering through this vague post, but it is frustrating.

 

[gar]

080501-1605 EST

wirenut1980:

Two things come to mind.

Severe overvoltages occur at the loss of power or at restoration. This could result in damaged insulation of motor windings.

Or very low voltage for a substantial time at loss or restoration.

I would tend toward high voltage transients.

When we had the great eastern blackout I was in a basement and the incandescent lamp went dim, out, back on very bright, out again and maybe one more cycle of this. Back at the shop we lost some fuses in the CNC machines, and had a failure in an Ethernet switch. None of my equipment in auto plants had any problems. In this equipment for the last 25 years I have use a Sola constant voltage transformer at the input to our gaging equipment.

With inductive circuits it is not hard to generate large voltage spikes. e = N df/dt where df/dt is the rate of change of flux. With a 8' Slimline fixture I can produce 5000 V transients when switching it on and off.

.

 

[iwire]

Rarely does the power go out 'cleanly' often the voltage may be reduced for quite sometime before it actually goes out entirely.

Low voltage can definitely take out the coils on motor starters and the windings on compressors.

 

[beanland]

Damage during the fault

Damage during the fault

I think what you may also see is that damage to equipment is caused during the fault but the power goes out so protective devices do not operate and equipment does not smoke. Then, when the power returns, the damaged equipment goes pfft. During the fault, voltage spikes can be severe, causing insulation breakdown. During restoration, there will be voltage spikes but they are generally smaller than those during the fault.

 

[mxslick]

And then there's ancient transformers...

And then there's ancient transformers...

Another reason this can happen is because old transformers on the poco's system fail upon restoration of power. One of the reasons has to do with a breakdown of the cooling oil in the transformer.

As explained in "Standard Handbook For Electrical Engineers", 13th edition, authors Fink/Beaty, Mc Graw Hill Books, Section 10-40, Part 108, and I quote: (bold emphasis added by me

"If the oil is suddenly cooled (reduction of ambient temperature or load) the oil volume contracts and and the static pressure of gas over the oil drops rapidly, allowing free gas bubbles to come out of solution throughout the insulation system. The dielectric strength of the oil and cellulose insulation system is drastically weakened when it has free gas inclusions, and this has occasionally led to electrical failure of operating transformers."

This refers to transformers with pressurized gas over the oil in the transformer tank, but the concept and failure mode applies to most oil filled transformers.

In simple terms, virtually all transformers of any age have suspended contaminants in the oil, and when the oil is cooled significantly, those contaminants come out of suspension in the oil and cause internal arcing and buildup of other gasses (including acetylene) and this causes either failure of the transformer (fire/explosion) or impression of high voltages on the normally low voltage side.

This happened a few years ago in my neighborhood.. a small plane crashed, took out a padmount, the restoration of power took over six hours, and when the power was restored, two other transformers on the same circuit failed, sending over 4kV into some buildings, catching several services on fire.

BTW, the aforementioned book is a goldmine of information on many aspects of electrical power, equipment, theory and design.

 

[wirenut1980]

080501-1605 EST

wirenut1980:

Two things come to mind.

Severe overvoltages occur at the loss of power or at restoration. This could result in damaged insulation of motor windings.

Or very low voltage for a substantial time at loss or restoration.

I would tend toward high voltage transients.

When we had the great eastern blackout I was in a basement and the incandescent lamp went dim, out, back on very bright, out again and maybe one more cycle of this. Back at the shop we lost some fuses in the CNC machines, and had a failure in an Ethernet switch. None of my equipment in auto plants had any problems. In this equipment for the last 25 years I have use a Sola constant voltage transformer at the input to our gaging equipment.

With inductive circuits it is not hard to generate large voltage spikes. e = N df/dt where df/dt is the rate of change of flux. With a 8' Slimline fixture I can produce 5000 V transients when switching it on and off.

I have seen overvoltages/transients occur at power restoration, but the ones I have seen have not been high enough in magnitude and/or not long enough in duration to cause damage according to the ITIC curve.

What circuit or load parameters "control" what the magnitude and duration of this event at restoration? In your equation e = Ndf/dt, isn't that the equation for induced voltage from a magnetic field? I'm confused on how that relates.:-?

Another reason this can happen is because old transformers on the poco's system fail upon restoration of power. One of the reasons has to do with a breakdown of the cooling oil in the transformer.

The cases I am talking about the transformer did not fail, at least not that I know of...I guess it is possible for a transformer to fail and still keep working in that it does not blow.


Beanland and iwire, you may be on to something there. But again, I am not sure why this damage only occurs at one customer out of hundreds on a distribution circuit. What is special about the circuit/load parameters that causes the one customer to have damage and no others?

Thanks for the replies everyone!

 

[zog]

Ok, something bothering me here that I can't explain...

I have seen this quite a few times, both single phase customers and 3 phase customers. Power outage caused by various faults and clearing of the faults is by single phase devices and 3 phase devices. Durations of outage range from seconds to hours.

Now I can understand how 3 phase equipment if unprotected can be single phased and there can be damage, and I can see if transmission lines fall into distribution lines how that can cause damage...

But what about a small commercial customer whose power is out for hours, served from a single phase distribution line, 400 amp service, power comes back on and wham, the compressor for a large refridgerator blows up and catches fire!

Or what about a 3 phase industrial customer who has an outage cleared by a 3 phase breaker, and when power comes back on, multiple motor starters fail?

These kinds of things have me scratching my head, because I just can't wrap my mind around how outages and power restored can destroy equipment in some circumstances.

Sorry for blubbering through this vague post, but it is frustrating.

Thats what all of those "extras" are in the protective relays and switchgear are for, you know the ones that are often not purchased or maintained in order to "save money".

There are about 90 different types of protective devices for power systems (I could post them all if requested) most plants use 2 or 3 of them.

 

[zog]

"If the oil is suddenly cooled (reduction of ambient temperature or load) the oil volume contracts and and the static pressure of gas over the oil drops rapidly, allowing free gas bubbles to come out of solution throughout the insulation system. The dielectric strength of the oil and cellulose insulation system is drastically weakened when it has free gas inclusions, and this has occasionally led to electrical failure of operating transformers."

thats what sudden pressure relays are for

 

[brian john]

Any time the power is turned on of off there is a line transient, and in the case you mention the transients are generally much worse (I would assume), due the the extent of the outage, and the nature of the outage.

And as mentioned low voltage, extended over voltage, major transient due to lightning.

Then there is the customer lie, Oh it worked before the power went out. Oh I had 1000.00 of dollars of steak in the refirg. Heard and seen both.

 

[cschmid]

When you loose power you need to shut down your huge loads right away..because when the power comes back on the draw is so huge it causes a brown out condition until the load stablizes..By leaving your motors and such on the line, causes the problem to be acerbated.. when the load is to huge at start up your motors have the opportunity fail..this can cause permanent damage to motors and is more prevalent than damage due to power failure..Because power failure is like hitting the stop button..except on huge motor loads that have large inertia loads..this in its self can be damaging to motor and equipment..

 

[gar]

080502-1008 EST

wirenut1980:

In a simple battery, switch, resistor, and inductor circuit used in elementary transient analysis you generally look at the change of current following switch closure.

If this circuit is analyzed at turn off we are dealing with quite different results. An inductor opposes a change of current. This is why the exponetial change in current on closing the switch above. If we have current flowing in an inductor and open the switch supplying the current, then the inductor will try to maintain that current in the same diirection and with the same magnitude as before the switch was opened.

Thus, after the switch opens the polarity across the inductor reverses and this becomes the source of energy instead of the battery and the current remains the same. If the switch resistance goes to infinity, then the voltage goes to infinity. The voltage does not go to infinity because the air gap between the switch contacts breaks down or something else does. Very large voltages are generated this way. For example in your automotive ignition system.

Now consider a transformer. If the primary line is broken at the instant of maximum magnetizing current in the primary, then we have the potential of a very large voltage generated at the primary. This is reflected to the secondary and any load on the secondary. Any effective impedances in the transformer, the magnitude of the load, and the stored energy in the transformer at this interruption will determine the transient voltage subjected to the load.

The df/dt is the source of the transient voltage from the transformer core.

I doubt that in the notheast blackout that my voltage exceeded double the nominal voltage, but someone much closer to the orgin probably saw much larger spikes. Note: to some extent a transmission line may look like a low pass filter. A low pass filter will take an impulse of energy, lower its peak magnitude, and spread it over time.

,

 

[djohns6]

thats what sudden pressure relays are for

Explain please .

 

[zog]

Explain please .

A sudden pressure relay is a realy driven by the sudden pressure relief device that senses a Delta P/time and trips the transformer off line if the pre set parameters are met. I think it is a IEEE device #68 but I need to look that up on Monday. Here is a pic of one.

 

[mxslick]

thats what sudden pressure relays are for

Quite true..but, as was explained to me by the poco's troubleman, sudden pressure relays aren't commonly installed on distribution transformers. They are used heavily on substation transformers though....and work quite well as an internal arcing fault will generate an abrupt rise in pressure.

BTW, do those relays also react to a sudden DROP of pressure? For example, if the tank were to rupture or some other sudden loss of oil happens?

And in the case I had mentioned, the problem wasn't a sudden rise or drop in oil pressure that was the issue, it was as the oil cooled the suspended gases and contaminants came out of suspension in the oil, attracted to the insulation and that caused the breakdown.

Had I been in contact with that tech then I would have asked for the failed transformer(s) so I could open them up and check out the damage..

Oh, and thanks for posting the link to the pic!!

 

[zog]

Quite true..but, as was explained to me by the poco's troubleman, sudden pressure relays aren't commonly installed on distribution transformers. They are used heavily on substation transformers though....and work quite well as an internal arcing fault will generate an abrupt rise in pressure.

BTW, do those relays also react to a sudden DROP of pressure? For example, if the tank were to rupture or some other sudden loss of oil happens?

And in the case I had mentioned, the problem wasn't a sudden rise or drop in oil pressure that was the issue, it was as the oil cooled the suspended gases and contaminants came out of suspension in the oil, attracted to the insulation and that caused the breakdown.

Had I been in contact with that tech then I would have asked for the failed transformer(s) so I could open them up and check out the damage..

Oh, and thanks for posting the link to the pic!!

I believe they can be set up either way.

 

[djohns6]

A sudden pressure relay is a realy driven by the sudden pressure relief device that senses a Delta P/time and trips the transformer off line if the pre set parameters are met. I think it is a IEEE device #68 but I need to look that up on Monday. Here is a pic of one.

OK , I just wasn't clear on your original post and how it it related to the topic . I see what you were driving at now . I'm familiar with sudden pressure relays and the one you showed is a QUALITROL brand with it's associated seal-in relay . The device is a 63 , right ? Never saw one set up for a drop in pressure though . Anything is possibe I guess .

 

[zog]

OK , I just wasn't clear on your original post and how it it related to the topic . I see what you were driving at now . I'm familiar with sudden pressure relays and the one you showed is a QUALITROL brand with it's associated seal-in relay . The device is a 63 , right ? Never saw one set up for a drop in pressure though . Anything is possibe I guess .

Yeah, 63. Thanks.

 

[hillbilly]

080502-1008 EST

Thus, after the switch opens the polarity across the inductor reverses and this becomes the source of energy instead of the battery and the current remains the same. If the switch resistance goes to infinity, then the voltage goes to infinity. The voltage does not go to infinity because the air gap between the switch contacts breaks down or something else does. Very large voltages are generated this way. For example in your automotive ignition system.

Now consider a transformer. If the primary line is broken at the instant of maximum magnetizing current in the primary, then we have the potential of a very large voltage generated at the primary. This is reflected to the secondary and any load on the secondary. Any effective impedances in the transformer, the magnitude of the load, and the stored energy in the transformer at this interruption will determine the transient voltage subjected to the load.
,

A light just went off in my head:smile: .

I now understand something that has bothered me for a while....which is what the OP was talking about.

Your reference to a automobile ignition system was the clue.

On a automobile system, when the primary supply to the coil (which is really just a transformer) is broken (opened), the field in the primary winding collapses and induces a super high voltage in the coil secondary....somewhere around 20KV potential in reference to the negative side of the auto electrical system..

This voltage induces a spark across a air gap (spark plug) as the positive current is discharged to the system negative.
This happens 66 times per second on a 4 cylinder engine running 4000RPM.

This is apparently the same thing that happens when a service transformer with a saturated coil is suddenly disconnected.
It induces a high voltage on the transformer secondary....ie...the house panel.

I am by no means a expert on transformers and my answer is merely speculation, and the terminology that I use is my own, although I do understand ignition systems on automobiles.

Does this sound correct....in any way?

If so, it answers a few questions in my mind.

steve

 

[gar]

080503-1216 EST

steve (hillbilly):

You are correct.

In the automotive ignition application the secondary capacitance (from both the leads and coil distributed capacitance), and the primary capacitance from the capacitor across the contacts have a substantial effect on the waveform. This slows down the rate of rise.

In the days, mid 50s, of 12 to 1 compression ratio and leaded gas there was considerable buildup of deposits on the plug insulator and this shunt resistance and the moderately slow rate of rise of the secondary waveform resulted in early failure. A solution to this was to use a capacitor discharge system providing a much faster rate of rise of the secondary voltage.

In the development of a capacitor discharge system we found that we could fire the fouled plugs, but there was less energy in the spark and this resulted in slightly less fuel economy at road load conditions. Then the task was to maintain the fast rate of rise and increase the energy imparted to the spark. We did this and put more energy into the spark than the standard ignition system. This was 1958, a major recession hit, and the funding was cut.

.

 

[gar]

080403-1301 EST

I need to add further clarification. In the automotive ignition coil the tansformer further increases the voltage from the inductive kick in the primary. Note: the DC excitation to the coil may be less that 6 V when you subtract out the voltage drop from external resistance. The coil does not multiple this voltage. The voltage across the primary of the ignition coil is a result of the stored energy in the core and the rate of collapse of the magnetic field.

In the case of a distribution transformer it is a step down transformer so the inductively induced voltage from the primary to secondary will be reduced, but the voltage on the primary from an inductive kick can be very large relative to the normal input primary voltage so we still can get a very large secondary voltage relative to the nominal value.

.