Temporary over voltage question

[wire twister]

I have an instance at my own residence where I am being hit with temporary over voltages of around 300 volts or so line to line that last several seconds. Unfortunately the POCO says that they see no way that this could be happening but my energy meter and destroyed equipment proves otherwise. My question to you folks is, what scenario can you guys come up with that could explain why this might be happening.

It seems to happen at random times of the day and does not correlate to anything I can think of in the house. I can say that this has been the coldest winter we have had since I lived at this residence. There are a lot of splices, about 1/2 a dozen, on the primaries as they make their way to my pole mounted transformer. The service goes down the pole and underground to the meter base on the house. I am about a quarter mile from the sub station. I have been informed that the transformer feeding my residence is loaded at 139% but it is permissible to load it to 180%.

Any ideas? I'm clueless about the high voltage supply and distribution side of things.

 

[JoeTroubleShoot]

How do you know the voltage that is causing the problems? Where you lucky enough to catch it or do you have some type of recording device? Either way there is no way you can get 300v and it be your problem.

It could be a number of things doing it on the POCO side, transformer, regulator, switching, substation problem. A bad splice or an overloaded transformer will not be the problem. I'm surprised that they run them that overloaded.. They must use 4/0 and nice tight connections on the leads.

You should ask for a recorder to be placed on the meter and then go from there. The only problem is that it will have to see the voltage which may damage equipment further.

 

[wire twister]

They have put a recording meter on at my request but insist that it shows nothing although I have my own recordings that prove otherwise from a whole house energy meter that has been installed since 2011. I was told there is nothing on their end that could cause voltages to elevate to that level. This problem persisted for about 3 months and went away about two weeks after they removed their recording device. I say they repaired the issue but they insist there was no issue to correct.

 

[JoeTroubleShoot]

File a claim for the loss of equipment. Unless your recorder is misreading voltage, there's nothing on your side that would give you 300v line to line, realistically anyway. Ask your neighbors if they have noticed anything as well. From my experience, most people do not report intermittent electrical problems. You think it's isolated to your house only, but when you ask around you get people who say "Oh yeah, it's been like that for years".

Once you file and claim and they deny it, then go to court with your readings. I work for a major utility and they will put up a good fight until there is a court date. They will always give you the run around because they know most people will back down. Unless they are 100% sure, they will not chance it in court.

 

[wire twister]

The meeting was because of the denial of my claim. I did mention small claims court as my next step and they decided to look into it further and told me they didn't deny the claim even though I had the denial letter in front of me. I've never been on this side of the process. Now I know what my customers feel like when they go thru this process.

 

[meternerd]

The nominal voltage of most utility distribution is 7200 volts with a 30/60:1 transformer. For your voltage to go to 300 volts the primary would have to go to 9000 volts. Have you recorded each 120 Volt leg as well? A neutral problem could cause high voltage on one leg but it would not cause high line to line voltages. Are any of your neighbors having problems? You say you have a recording meter, but what kind? Is it true RMS, high quality graphic recording, etc? If it's a basic digital meter with peak hold, how do you know it's lasting several seconds? I guess I just can't see a substation (or any other POCO equipment) being able to cause the problem. Regulators in subs are normally set to 15 second delay and then step up or down 5/8% per step to correct voltage. Max regulation is 10% high or low. 300V would be 12.5%, which would be impossible for a regulator to do.

 

[wire twister]

Neutral is in good shape. It has been visually and electrically tested even though the neutral has nothing to do with high line to line voltages. I have a revenue grade net metering device that has been in place since 2011 that outputs the min/max voltages averaged over a 15 second period as well as KWH usage, KW, KVA, & PF information. I have the smoked devices and time stamps of power events that match the damage so this is absolutely happening. Most of what I lost was able to withstand the TOVs for a couple of months before failing. That's why I didn't investigate sooner. As for the neighbors, one neighbor shares a transformer with me and they are a vacant building with no electronics operational. Everyone else up and down the line as far as I'd like to chase it are on A or C phase while I am on the B phase heading out of town. Not sure why that is or if it is even relevant. I'm sure the POCO engineers have a reason for it.

The only scenarios I could come up with would be slow switching on their end on the tap changers at the substation but it sounds like that is outside of the 10% tolerance, slow capacitor switching but I don't see any capacitors between me and the station, or a failing transformer that has windings that are partially shorting together under certain conditions causing a smaller ratio. The last one I'm not even sure could happen without having a transformer failure. That's why I'm on here asking questions.

 

[wire twister]

So with the 10% switching swing on the 7620 primary only some of my readings fall above this level. They range from 270 to 299 volts line to line. With a few in the 255-260 range which I wasn't too concerned with.

 

[gar]

140326-2020 EDT

wire twister:

Can you provide more information on your power monitor, and the data from it?

I have monitored my whole house energy use over a substantial time with a TED 1000 system. I collect 1 second data samples. You can see a sample plot at http://beta-a2.com/energy.html . If there was a large voltage spike of 1 second you would see it like the power spikes. The power spikes are motor starting pulses and are quite short.

The TED 1000 only measure one voltage and that is a hot to neutral (120 V). The TED 5000 system can read only one voltage, either line-to-line or a line-to-neutral.

If you were reading 300 V RMS line-to-line, and this lasts for 15 seconds or more, and if you do not have power generation on your side of the transformer, then the problem has to originate with the power company. Even if you had power generation capability it would have to be quite large to force the voltage that high. The power company is a pretty stiff system (meaning low impedance).

I am doubtful of power company voltage monitoring instruments. At our shop a number of years ago, late 90s, we had voltage problems. DTE put a monitor on our meter. They observed no problem monitoring over a two month period. During this time we could see large voltage changes visually with the lights, and CNC machines dropping out. Also saw a fan speed up. At that time I did not have a TED system. I set up a crude voltage monitor and saw voltages up to 140 V. So we lived with the problem. A few years later the fluctuation was large and we called DTE. Near dusk they sent a truck. This person could not do anything until another truck came. While he was sitting there he saw arcing at the transformers, three phase high leg delta with two transformers. Obviously bad connections at the transformers. They replaced one transformer, and fixed the connections. The point here is that the power company monitoring, for whatever reason, may not detect a problem.

.

 

[SG-1]

I have experienced a transformer failure where the windings were slowly shorting out. The problem went on for a year or more. The first thing we noticed was AM Radio interference. When the transformer was acting up you could not listen to the radio on that band. The voltage rise was never seen above 256 or so. I was just checking it from time to time with a multi-meter. We did not lose any electronics, although you could hear some of them making noise, until the transformer went out with a bang in a blaze of glory.

 

[wire twister]

It is a TED 5000. Unfortunately I do not have second data because of the memory limitations of the device itself. I do have the daily min/max data which is where I'm getting my numbers from. It shows a steady trend of 245 volt +/- maximums for years until November of 2013. At that point the maximums start increasing in magnitude an frequency with January seeing many events. Then about a week or so after the poco removed their meter the over voltages completely disappeared. This fact alone makes me believe there was a problem and it was corrected by the POCO.

For what it is worth I do not have any power generation equipment of my own.

 

[wire twister]

The other scenario I was thinking of is a failing neutral connection upstream of me that would cause elevated voltages in the primary feeding my transformer and possibly others. This could explain the intermittent nature of it.

 

[hurk27]

The other scenario I was thinking of is a failing neutral connection upstream of me that would cause elevated voltages in the primary feeding my transformer and possibly others. This could explain the intermittent nature of it.

Very possible, but only if you are on a single line run and not a loop feed, also the pole grounding would have to be weak as well as the balance between the phases would have to be off with being your phase (B) the least loaded phase, this is the same thing that happens when we loose a neutral connection to a house, the lighter loaded leg gets the higher voltage.

Let's say your power lines are an end of the line run down a street, you have 3 transformers with a total load of 12 amps on B phase to the MGN, 5 transformers with a total load of 30 amps on A phase to the MGN, but C phase has 4 transformers but one feeds a large service, and it has a total load of 60 amps, since a 7.2/12.4kv system is a WYE all transformers will be loading the MGN, now the MGN has a bad connection somewhere ahead of these transformers, they now try to balance themselves out across the phases, and your phase gets the higher voltage minus any current that can flow on any Earth grounding.

The problem is most POCO's try to keep the lines balanced as much as possible, and the MGN should be grounded 4 times a mile not including at each service the transformers serve, unless you have some of the highest resistance soil in the country, I can't see it happening very easily.

The other possibility is arcing faults between the primary winding turns, but you didn't mention if the POCO changed out the transformer, when a fault happens on the primary side of the windings it lowers the turn ratio which raises the voltage on the secondary, I have had this happen on a couple service calls, but it is very rare.

Also everytime I have had the POCO put a recorder on a customers meter they received a printout of the recording with their account number, meter number, and name and address on the printout, did you receive this? if not then I would suspect their hiding something.

 

[gar]

140327-1223 EDT

wire twister:

The TED 5000 is not a revenue grade power and energy monitor. It has major power factor problems, and there are many reports of erroneous readings. Major software problems. Also appears that line voltage variations, probably meaning momentary loss of power, can cause failure to communicate or data changes. Although I have a TED 5000 I do not used it other than to see how it works.

In some ways the the 1000 system is a more reliable system.

Is your 5000 system wired to both phases? If so then your single voltage measurement ( TED can not make two voltage measurements, nor two current measurements ) is a line to line measurement. In the TED 1000 and 5000 systems the two apparent input voltages are immediately added at the input to form one summed voltage. The same for the two possible current inputs.

With the 1000 system the design is such that I write both voltage and power measurements to hard disk once every second. So those short average measurements recorded once per second have no cummulative error.

As negative as I may sound I can get reliable useful information from the TED 1000 system.

See my notes at http://beta-a2.com/engry_b.html .

.

 

[gar]

140327-9891 EDT`

wire twister:

Some more TED comments.

I am willing to put some credibility in the TED voltage data for your problem. Especially because it time correlates with your failures. Whether the voltage values are accurate at that level I don't know. The level might be high enough to saturate the input voltage circuit. I have not tested TED for its maximum voltage saturation level. This means voltages could have been higher. Also not having shorter time measurements and how the voltage varied during the 15 second periods does not help identify the cause of the problem.

Although TED says that the 5000 samples once per second that is not true. At their fast rate it is once per two seconds, and the data for every other second is simply a duplicate of the previous second data.

Sometimes the 5000 model samples only occur at about a 5 second period.

The TED 1000 and 5000 systems derive power from the black and white 120 V input leads. 150 V or higher for a moderate time might damage the small internal power transformer and linear 5 V regulator. I think you are lucky that your overvoltage condition did not damage TED.

I don't know how well your data would hold up in court, but I think it is good evidence of power company problems.

What were the maximum durations of the high voltage condition, I don't remember that you mention this?

.

 

[kwired]

It is a TED 5000. Unfortunately I do not have second data because of the memory limitations of the device itself. I do have the daily min/max data which is where I'm getting my numbers from. It shows a steady trend of 245 volt +/- maximums for years until November of 2013. At that point the maximums start increasing in magnitude an frequency with January seeing many events. Then about a week or so after the poco removed their meter the over voltages completely disappeared. This fact alone makes me believe there was a problem and it was corrected by the POCO.

For what it is worth I do not have any power generation equipment of my own.

It does seem a little suspicious that problems went away as soon as they removed their monitoring equipment. They probably figure they are not going to give you any information on what happened there as it will only be fuel for you to use against them in any pending lawsuit. Any further course of action involving what is going on with the POCO side of things really needs to come from legal counsel IMO, as they almost have to be hiding information to cover themselves.

140327-1223 EDT

wire twister:

The TED 5000 is not a revenue grade power and energy monitor. It has major power factor problems, and there are many reports of erroneous readings. Major software problems. Also appears that line voltage variations, probably meaning momentary loss of power, can cause failure to communicate or data changes. Although I have a TED 5000 I do not used it other than to see how it works.

In some ways the the 1000 system is a more reliable system.

Is your 5000 system wired to both phases? If so then your single voltage measurement ( TED can not make two voltage measurements, nor two current measurements ) is a line to line measurement. In the TED 1000 and 5000 systems the two apparent input voltages are immediately added at the input to form one summed voltage. The same for the two possible current inputs.

With the 1000 system the design is such that I write both voltage and power measurements to hard disk once every second. So those short average measurements recorded once per second have no cummulative error.

As negative as I may sound I can get reliable useful information from the TED 1000 system.

See my notes at http://beta-a2.com/engry_b.html .

.

I have a TED 5000, and have noticed it has issues at times. I don't use so much for monitoring incoming power problems though it may help uncover some things. I usually install it at a customer premises when they are complaining about high electric bills. If you make daily reports and graphs for a few days (if you wait longer then that you don't get enough hourly details in the reports) you can often look at the graphs and tell them when they left for work, when they came home, when the kitchen activity slowed down after they came home, etc. each day. You can often pick up a power change signature from specific items and tell them a particular spike is likely the water heater, the well, the heating, etc. Many times they come to realize they are using more energy then they thought and nothing is wrong with anything.

I haven't used it recently, but I have heard many complain about high energy bills the last couple months, and have to remind them it was pretty cold and they are likely using more then they realize, plus energy costs only go up 2-5% each year most years but after 5-10 years it starts to add up to more then one realizes.

 

[ghostbuster]

Here is a video that demonstrates a primary voltage problem (electrical arcing) can be transferred to the secondary side wiring and burn down the entire building.Less serious utility voltage problems if left uncorrected can also be "transferred" to the cutomer's secondary side.

http://www.youtube.com/watch?v=37W-X_c6amY

 

[kwired]

It is a TED 5000. Unfortunately I do not have second data because of the memory limitations of the device itself. I do have the daily min/max data which is where I'm getting my numbers from. It shows a steady trend of 245 volt +/- maximums for years until November of 2013. At that point the maximums start increasing in magnitude an frequency with January seeing many events. Then about a week or so after the poco removed their meter the over voltages completely disappeared. This fact alone makes me believe there was a problem and it was corrected by the POCO.

For what it is worth I do not have any power generation equipment of my own.

To add to what I said before, if you are close to the substation unless there is a problem specifically with your transformer, other customers about have to be effected as well, they may be witholding information to help prevent everyone effected from filing claims against them.

 

[gar]

140327-1029 EDT

kwired:

With the TED1000 system and computer collection of the data I can get over 1 day's worth of 1 second data before I need to save the file. One day of 1 second data is about 3,000,000 bytes. A considerably longer time period of 1 minute data can be stored before a file save is required. With this raw data it is possible to post-process for any desired information.

The TED systems have a major problem with non-unity power factor. Using the TED system I can "prove" that a whole house power factor correction capacitor will reduce the amount of energy used, but this is not true.

Your use of the 5000 serves a very useful function. I would think more electricians would find it a useful troubleshooting tool.

Using two 1000 MTUs would provide a useful means to study neutral problems. With the original 1000 MTU resolution was 1/64 W and about 1/57200 V. I do not know what is the noise level on these values. These ratios come from the binary data for a specific absolute change in the input signal. The 1/57200 seems large, but at some time I measured the hexadecimal change for 0.1 V and got a decimal value of 5720.

I believe it is possible with simple circuitry to provide battery backup of the MTU to maintain power to the MTU electronics during loss of line voltage, and for short durations to see what happens during the on-off transisitions. Of course this means that you need a different receiver of the data than the RDU or Gateway. For the 5000 system a more complex receiver-transnitter system is required.

.

 

[kwired]

140327-1029 EDT

kwired:

With the TED1000 system and computer collection of the data I can get over 1 day's worth of 1 second data before I need to save the file. One day of 1 second data is about 3,000,000 bytes. A considerably longer time period of 1 minute data can be stored before a file save is required. With this raw data it is possible to post-process for any desired information.

The TED systems have a major problem with non-unity power factor. Using the TED system I can "prove" that a whole house power factor correction capacitor will reduce the amount of energy used, but this is not true.

Your use of the 5000 serves a very useful function. I would think more electricians would find it a useful troubleshooting tool.

Using two 1000 MTUs would provide a useful means to study neutral problems. With the original 1000 MTU resolution was 1/64 W and about 1/57200 V. I do not know what is the noise level on these values. These ratios come from the binary data for a specific absolute change in the input signal. The 1/57200 seems large, but at some time I measured the hexadecimal change for 0.1 V and got a decimal value of 5720.

I believe it is possible with simple circuitry to provide battery backup of the MTU to maintain power to the MTU electronics during loss of line voltage, and for short durations to see what happens during the on-off transisitions. Of course this means that you need a different receiver of the data than the RDU or Gateway. For the 5000 system a more complex receiver-transnitter system is required.

.

I have 2 MTU's with my system. Often a customer will want to blame a high electric bill on a specific piece of equipment, making it perfect to monitor the main feeder with one MTU and the questionable circuit with the other MTU - you do have to remember that the main feeder readings includes what is recorded on the second MTU and the NET readings will appear larger then what is really there.