Three Phase site - multiple electrical issues.

Status
Not open for further replies.

Sparty D

Member
Location
Kansas
Hello, I am a new member and have enjoyed this site as a resource in the past. I have site that has me baffelled. And open to suggestions.
The site is a 2 year old complexe with three buildings. A memory care facility, this complex is fed by a Wye transformer to a 1600 amp main. There are multiple sub panels in each of the buildings. In the first year the building expierenced multiple failures and it was not helped by the poor wiring done by a now bankrupt contractor. As example, covers were pulled off and labled breakers were not in use so possible unbalanced loading was possible early on. Everything is Romex.

Since the start; half the electronic ballast lights have been replaced, the Arc Fault breakers trip without anyone in rooms, emergency lights register a power failure, UPS units do the same, currenly about 90% of the electronic thermostates on PTAC units have been replaced, two fire alarm power supplies, low voltage lights go out 5-6 at a time. The part we work on is the elopment system and the tag receiver by the doors scramble and need to be reset. (RS485 and RF) Other low voltage units with simple logic have needed power cycles or replaced. When entering the building a radio in the 120mhz area breaks squealch.

Suspecting Harmonics a power quality test was done and ten min. averages were as high as 19.9%THD, but the testing company does not think that is out of the norm considering the electronic lights. The complex max current draw for the testing period was 524amps and the neutral wire had 123amps. Again, they don't think that is abnormal or an issue.

Any Three Phase experts or Harmonic experts agree? Thanks for input.
 

JDBrown

Senior Member
Location
California
Occupation
Electrical Engineer
This is just a guess, but have you measured the voltage at the main service? A lot of the problems you describe sound like undervoltage issues to me. When you measure, make sure you use a True RMS meter, since with ~20% THD a reading with the other kind will be totally unreliable.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
This is just a guess, but have you measured the voltage at the main service? A lot of the problems you describe sound like undervoltage issues to me. When you measure, make sure you use a True RMS meter, since with ~20% THD a reading with the other kind will be totally unreliable.

I would definitely recommend a true RMS meter for the current measurements, but I doubt that the harmonic level in the voltage at the service is anywhere near that high. If it is, then there is a serious problem with the impedance of the POCO feed to the building.
 

texie

Senior Member
Location
Fort Collins, Colorado
Occupation
Electrician, Contractor, Inspector
Hello, I am a new member and have enjoyed this site as a resource in the past. I have site that has me baffelled. And open to suggestions.
The site is a 2 year old complexe with three buildings. A memory care facility, this complex is fed by a Wye transformer to a 1600 amp main. There are multiple sub panels in each of the buildings. In the first year the building expierenced multiple failures and it was not helped by the poor wiring done by a now bankrupt contractor. As example, covers were pulled off and labled breakers were not in use so possible unbalanced loading was possible early on. Everything is Romex.

Since the start; half the electronic ballast lights have been replaced, the Arc Fault breakers trip without anyone in rooms, emergency lights register a power failure, UPS units do the same, currenly about 90% of the electronic thermostates on PTAC units have been replaced, two fire alarm power supplies, low voltage lights go out 5-6 at a time. The part we work on is the elopment system and the tag receiver by the doors scramble and need to be reset. (RS485 and RF) Other low voltage units with simple logic have needed power cycles or replaced. When entering the building a radio in the 120mhz area breaks squealch.

Suspecting Harmonics a power quality test was done and ten min. averages were as high as 19.9%THD, but the testing company does not think that is out of the norm considering the electronic lights. The complex max current draw for the testing period was 524amps and the neutral wire had 123amps. Again, they don't think that is abnormal or an issue.

Any Three Phase experts or Harmonic experts agree? Thanks for input.

Pure guessing from the comfort of my desk without knowing all the facts and particulars, this sounds like a possible intermittent neutral connection. You probably don't have many incandescent lights to notice some bright or some dim in a place like this. I would setup a voltage monitoring/recording test over at least a 7 day period to see how that looks. When you look at the data if you see one phase go high and another phase go low at the same point in time I would look for a weak or intermittent neutral connection.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
Pure guessing from the comfort of my desk without knowing all the facts and particulars, this sounds like a possible intermittent neutral connection. You probably don't have many incandescent lights to notice some bright or some dim in a place like this.
Is the lighting 120 volt or 208 volt? If the latter, it would not be directly affected by an unreliable neutral. (But I would not expect you to be blowing ballasts on such fixtures either.)

If it is a high resistance or intermittent neutral, it could be made worse by single phase loads with a high inrush current like large motors. Motors which are on steadily and are wired in wye may actually have a stabilizing effect on the neutral voltage.
 

Sparty D

Member
Location
Kansas
The building did have a Power Quality study for three weeks at the main. Nothing failed during the test so it could be considered inconclusive.
A loose neutral has been suggested, but the local power company said their equipment is fine. (expected reply). We are the low voltage contractor and the high voltage contractor hired after the first went bankrupt was supposed to check things thoroughly. Especially after they found that breakers were labeled and not used.
During the three weeks there were momentary (<1sec.) dips to 97, 106 and 114. Prior to the outside company doing the test I also had a 106 captured by my Fluke 289. All three phases dipped simultaneously when the low reading was recorded. The outside company felt it was the utility side that caused the dip.
The lighting is 120VAC and mostly florescent. The harmonics start like clock work around 8am as the current ramps up too. All three buildings are affected, but they all do get power from the main.
This was the last building built in the area and I am not sure how the utility ran the feeds, but the substation is just across the four lane street.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
Just to make sure we are not missing something, can you confirm whether the high harmonic content is measured in the current only or in both current and voltage?

If the fluorescent ballasts are low power factor and the lighting is the majority of the load, 30% is not difficult to accept. More so if they are electronic raonics.ther than magnetic ballasts. A magnetic ballast should produce phase distortion without as high a level of harmonics.
 

Sparty D

Member
Location
Kansas
Current = Voltage Harmonics with stepdown transformer?

Current = Voltage Harmonics with stepdown transformer?

The Fluke used gave 10 minute avg ranges.
Phase Voltage 118.8-124.5
Phase Current 86.1-524.0
Neutral Current 28.0-123.4
Power Factor .866-.943
Voltage Harmonics 1.55-2.38
Current Harmonics 4.95-19.90

It is hard to tell what the major loads would be. Each room is like a hotel and has a PTAC and the halls have a total of 150 ele. ballast lights. There is kitchen and I think 6 roof top units (two each for the three buildings). At the time the test started, it was a nice 68 degrees and none/little of the HVAC was running. The current harmonics at that moment were 15%.

Since all of the low voltage systems have seen some type of problem or damage I asked the tech setting up the analyzer what the high current harmonics would do since low voltage systems use a step down transformer. He said that current harmonics become voltage harmonics through step down transformers. Anyone know if that is correct or where I can go to see what the output results would be?
 

Sparty D

Member
Location
Kansas
Still acting up

Still acting up

The Fluke used gave 10 minute avg ranges.
Phase Voltage 118.8-124.5
Phase Current 86.1-524.0
Neutral Current 28.0-123.4
Power Factor .866-.943
Voltage Harmonics 1.55-2.38
Current Harmonics 4.95-19.90

It is hard to tell what the major loads would be. Each room is like a hotel and has a PTAC and the halls have a total of 150 ele. ballast lights. There is kitchen and I think 6 roof top units (two each for the three buildings). At the time the test started, it was a nice 68 degrees and none/little of the HVAC was running. The current harmonics at that moment were 15%.

Since all of the low voltage systems have seen some type of problem or damage I asked the tech setting up the analyzer what the high current harmonics would do since low voltage systems use a step down transformer. He said that current harmonics become voltage harmonics through step down transformers. Anyone know if that is correct or where I can go to see what the output results would be?

The site had more activity and this is after I installed a power conditioner. I just installed a power conditioner from Tripp Lite and the over a weekend it had multiple anomalies. If you believe the conditioner is working properly, the only thing left is Radio Interference or harmonics from ground or transients coming through the ground. This also occurred with a excessive amount of rain....3 inches. I still wonder if the main substation across the street could be a factor since the worsted case of lock ups occurred during a period of super ground saturation.
Any thoughts is welcome!! Everyone here is baffled.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
The site had more activity and this is after I installed a power conditioner. I just installed a power conditioner from Tripp Lite and the over a weekend it had multiple anomalies. If you believe the conditioner is working properly, the only thing left is Radio Interference or harmonics from ground or transients coming through the ground. This also occurred with a excessive amount of rain....3 inches. I still wonder if the main substation across the street could be a factor since the worsted case of lock ups occurred during a period of super ground saturation.
Any thoughts is welcome!! Everyone here is baffled.
The main symptom which did not initially fit into the realm of possible symptoms of neutral problems is the 120MHz radios breaking squelch.
That by itself is an indication of high stray RF power or high levels of power from a permitted service.
I would investigate that separately from the rest of your power conditioning problems.

Is there a high power commercial or governmental radio transmitter in the area? It need not be in the 120 MHz range itself, since you might be seeing harmonics of the fundamental RF frequency.
Changes in ground conductivity can certainly affect stray power from a radio transmitter.

You did not mention initially the main substation across the street. That can cause very high AC currents to be present in the ground under the building, with the surface voltage gradient even rising as the soil gets saturated because the primary neutral currents are not forced into the deeper layers of earth. Such currents could cause "grounded" equipment in different parts of your building or building complex to be at different voltages, with associated damage to communication equipment whose wiring runs all around the building.
I would get careful measurements made looking for voltage differences between the EGC wires at different points in the building. Difference in neutral voltage are expected because the neutrals carry current normally. The same does NOT apply to the EGCs, pipes, structural steel, etc.

Do some or all of the failed electronic components have wired interconnections between buildings each of which has its own ground electrodes?
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
Excuses me but a 1600a service and Romex I find it hard to believe are used in the same paragraph.

I agree.

What is a memory care facility? Like a nursing home of some sort?

I think there is a significant clue in the radio breaking squelch. I don't know what it means, but it is rather unusual and needs to be run down.
 
Last edited:

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
What is a memory care facility? Like a nursing home of some sort?
A more general and kinder term for a place for Alzheimer's or other dementia patients who can't remember where they are or what they are doing.
Some have a tendency to wander off (euphemism = elope) and so need to be monitored to make sure they do not leave the facility. In the OP's case the guests are apparently fitted with RFId tags to enable them to be found more easily.
They may or may not have other medical problems to complicate things, but that would usually require a skilled nursing environment rather than just a memory care facility.
 

Sparty D

Member
Location
Kansas
The main symptom which did not initially fit into the realm of possible symptoms of neutral problems is the 120MHz radios breaking squelch.
That by itself is an indication of high stray RF power or high levels of power from a permitted service.
I would investigate that separately from the rest of your power conditioning problems.

Is there a high power commercial or governmental radio transmitter in the area? It need not be in the 120 MHz range itself, since you might be seeing harmonics of the fundamental RF frequency.
Changes in ground conductivity can certainly affect stray power from a radio transmitter.

You did not mention initially the main substation across the street. That can cause very high AC currents to be present in the ground under the building, with the surface voltage gradient even rising as the soil gets saturated because the primary neutral currents are not forced into the deeper layers of earth. Such currents could cause "grounded" equipment in different parts of your building or building complex to be at different voltages, with associated damage to communication equipment whose wiring runs all around the building.
I would get careful measurements made looking for voltage differences between the EGC wires at different points in the building. Difference in neutral voltage are expected because the neutrals carry current normally. The same does NOT apply to the EGCs, pipes, structural steel, etc.

Do some or all of the failed electronic components have wired interconnections between buildings each of which has its own ground electrodes?

First, each building is isolated from each other and there is a fiber optic connection for RS232 communication to a PC. So each building, other than the same power source is independent. The units can be affected by RF, RS485 and power. The AC is stepped down to 12VDC. I installed a power conditioner so between that and the power supplies I would hope the DC is not the source. The units are also earth grounded and is now the focus for me.

I have driven around the area and don't know of any radio or cell towers/transmitters in the immediate area. There is a over abundance of HAM licenses though. The latest problems arose after a 3" rain so the ground was very saturated. Since the electrical company that did the install went under and apparently had other projects that had issues I do wonder if the grounding is correct. It was not correct to start and the GC had a new company sink three rods and pull a new ground after a year of operation.

I am trying to get a journeyman who I think knows he stuff in the door and look at it. I think a fresh look is needed. I have to wonder with the three phase service, backup generator, and multiple subpanel's in the three buildings if grounding was done right. I have read that improper grounding can be a source of EMI.

The power quality study that was done showed 19.9 % THD on Current. They didn't think that was out of the norm. The neutral had 123 amps with a max current of 523 amps. That seemed high for the amount of current.

And yes. 1600 amp main and romex (from the subpanels). This is a "residential" complex and built with wood studs....like the old days.
 

templdl

Senior Member
Location
Wisconsin
The site had more activity and this is after I installed a power conditioner. I just installed a power conditioner from Tripp Lite and the over a weekend it had multiple anomalies. If you believe the conditioner is working properly, the only thing left is Radio Interference or harmonics from ground or transients coming through the ground. This also occurred with a excessive amount of rain....3 inches. I still wonder if the main substation across the street could be a factor since the worsted case of lock ups occurred during a period of super ground saturation.
Any thoughts is welcome!! Everyone here is baffled.

You wouldn't think that harmonics would be an issue on 1ph branch circuits. Harmonics are an issue with 3ph where the harmonic end up as neutral currents which are in addition to the unbalanced current that are carried in the neutral.
Hence, panels with 200% neutrals and k factor transformers built with extra iron to deal the extra heating because if the harmonics along with other design factors too include a 200% neutral. The transformers also include electrostatic shields to isolate noise.
 

hurk27

Senior Member
while the electronic ballast can be a great source of RFI since the radio is in the 120mhz which is the 6000th harmonic of 20khz which some electronic ballast work at, by the way what kind of radios are you using that is in the 120mhz band as that is in the aircraft band which would be unusual to have a hand held using that frequency?

I good RDF noise finder is an old transistor AM radio and the flat iron core antenna can be used to zero in on the direction the noise is coming from, 1200khz on the AM broadcast band would be the 60th harmonic of 20khz.

VFD drives if used can also be a source of RFI depending upon the output frequency it is set to run at as we found with some drives at work getting into a wireless alarm system.

As for the other problems, as others have said, my first though was bad neutral connection at the first disconnect or ahead but you ruled that out with the voltage fluctuation on all three phases which would lead me to think a closer look at the POCO's substation might be in order.

A partial loss of the primary MV neutral to the utilities transformer could cause the transformer to return the primary transformer current back through the buildings grounding system which could cause all kinds of havoc, ha the EC do a voltage test from the service equipment cases to remote earth to see if there is a voltage above 10 to 15 volts, if so then have the utility check all the neutral connections on the primary side the the service transformer.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
130928-0832 EDT

Sparty D:

It appears that there may be several different types of problems. A more accurate description of the system is needed.

For the moment I will only comment on several items.

You mentioned RS232 with a fiber optic link. You did not indicated if this path had any problems. It should not.

Then you referenced RS485, bit no mention of isolation. If this path does not have isolation you will have data errors and/or equipment damage.

Equipment that locks up and does not recover implies badly designed circuits. Primarily meaning no watch dog timer to provide automatic reboot. But also means the equipment is not adequately designed to handle transients. Total loss of power for an extended time vs voltage spikes are different problems.

What is the 12 V DC stuff, and how does this relate to the problem?

What are the transformers that you referenced? If you have a transformer with a high load current harmonic content, then the voltage at the output of said transformer will have a larger harmonic content than at said transformer input because of the series impedance between input and output of said transformer. This is only logical as a conclusion.

If you pick some reference point, like the EGC termination at the main panel, then you can not assume that the voltage between the reference point and any other point on the EGC is zero, or even close to zero. This is why any communication paths need isolation.

.
 

Sparty D

Member
Location
Kansas
130928-0832 EDT

Sparty D:

It appears that there may be several different types of problems. A more accurate description of the system is needed.

For the moment I will only comment on several items.

You mentioned RS232 with a fiber optic link. You did not indicated if this path had any problems. It should not.

Then you referenced RS485, bit no mention of isolation. If this path does not have isolation you will have data errors and/or equipment damage.

Equipment that locks up and does not recover implies badly designed circuits. Primarily meaning no watch dog timer to provide automatic reboot. But also means the equipment is not adequately designed to handle transients. Total loss of power for an extended time vs voltage spikes are different problems.

What is the 12 V DC stuff, and how does this relate to the problem?

What are the transformers that you referenced? If you have a transformer with a high load current harmonic content, then the voltage at the output of said transformer will have a larger harmonic content than at said transformer input because of the series impedance between input and output of said transformer. This is only logical as a conclusion.

If you pick some reference point, like the EGC termination at the main panel, then you can not assume that the voltage between the reference point and any other point on the EGC is zero, or even close to zero. This is why any communication paths need isolation.

.

The complex is 3 separate buildings and each door unit (w/built in RFID tag receiver) used RS485 to relay messages to the Main Controller of that building. The Main Controller then relays that buildings messages to the PC via RS232. Two or the three use fiber optics converters so that should be electrically isolated.

The Door and Main units are 12VDC and to get the 12VDC I use 120VAC to 18AVC open frame transformers that power Linear Power Supplies outputting the 12VDC. That output has been measured and very stable as measured by a Fluke 289. I have never been there at the right spot at the right time.

The units can handle a total loss of power, but with batteries on the Linear Power Supplies it should not happen

If you did not catch it earlier in the thread, in the first year the site had 1/2 the electronic ballast replaced, 75% of the electronic thermostats on PTAC units go bad, two fire panels go out, low voltage light bulbs go out 6 at a time and arc fault breakers trip in empty rooms.

I am a pilot and thought a possibility could be a transmission from a plane on approach to a local small airport. I was surprised when I walked in and the radio broke squelch. Step outside it stops. Walk back in it starts.....in all three buildings.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
A partial loss of the primary MV neutral to the utilities transformer could cause the transformer to return the primary transformer current back through the buildings grounding system which could cause all kinds of havoc, ha the EC do a voltage test from the service equipment cases to remote earth to see if there is a voltage above 10 to 15 volts, if so then have the utility check all the neutral connections on the primary side the the service transformer.

In addition to measuring voltage to remote earth (and a valid "remote earth" may be hard to find across the street from a substation!), the EC should look for high current in any EGC or GES conductors. These will be indications of a high gradient between local surface earth potential across the building site.
It would take as little as 1A to cause a 10 volt differential across a 10 ohm ground electrode resistance, but a 10A GEC current could cause only a fraction of a volt difference within the GES wiring but correspond to a difference in local earth potential between the two electrodes of 100V, a real life safety hazard as well as potentially damaging connected equipment.
 
Status
Not open for further replies.
Top