Ground Current... Caused by motors?

[waltppiva]

I was recently called to a site to investigate problems with ground current. This site used to be a grocery store, but is now occupied by a large call center. The issues became apparent when a 10kVA UPS was installed, and when tested at startup would not consistently transfer to battery during a power outage. The UPS sales rep had a tech come to the site and he found current on the ground. When I went to the site I got similar readings. At the main neutral-ground bond I found 16.5 amps on the ground. Other readings revealed anywhere from 4.5 amps at the UPS input to 14 amps at the UPS feed panel and 18 amps at the generator ATS. From my experience I know that the only thing that will cause current to run on the ground is if the neutral and ground are bonded where they shouldn't be. They should be bonded at the main service entrance panel and any other newly derived source (generator or transformer). If N-G are bonded anywhere else in the electrical system this will allow the current from the neutral to transfer onto the ground. I have scheduled a maintenance window with the customer so I can identify and correct the source of the ground current.

As I said, this issue came to light when a new UPS was installed but would not consistently transfer to battery upon power outage. The tech from the UPS sales rep and the tech from the UPS manufacturer both stated that the reason the UPS does not transfer to battery is because there are motor loads for AC units throughout the building that when the utility power is lost, as the motors are still running they can act as a generator and backfeed power to the UPS input causing it to malfunction. I understand the theory of the motors backfeeding power, but to say that will cause the UPS to malfunction, quite frankly, sounds like a lot of BS. To correct this issue they propose an isolation transformer be installed on the primary side of the UPS. I can partially follow their logic, but it sounds like they're grasping at straws without gathering and analyzing all of the data first. Any thoughts on the ground current and/or the UPS issue?

I hope everyone has a wonderful holiday!!!

 

[jim dungar]

Re: Ground Current... Caused by motors?

Definitely grasping.

Yes motors can act as generators. They can only backfeed if they are on the load side of the UPS. Computer room Air Cond motors are common on the load side of UPS systems, so why is this now a problem? How can adding an isolation transformer to the line side prevent backfeeding by motors?

Is there a dedicated isolation/disconnect switch on the feeder to the UPS? If this is opened then interference on the line side could be ruled out.

Of course, you need to determine the cause of the ground current. Don't be suprised if some of it is not coming from the UPS.

 

[waltppiva]

Re: Ground Current... Caused by motors?

Thank you for the quick response. The UPS in question is a 10kVA that only feeds power to the server racks and phone equipment in the computer room. The AC motors are on the primary side of the UPS, not on the load side. This being the case, installing an isolation transformer on the UPS primary would not accomplish anything.

Is my logic correct regarding the probable cause of the ground current, improper neutral-ground bonds?

 

[brian john]

Re: Ground Current... Caused by motors?

1ST:

Most UPS techs (not all) are electronic techs with little or no power experience, while you should listen to them and not necessarily discount what they say, be leery, especially in this case. As for the sales person, well an insolation transformer may assist in the isolating the UPS input, remember he/she is a SALES PERSON, and selling something will help........them at least.

2nd.

Don't assume all ground current is a result of neutral boding downstream from the main service. It is possible for a resistive heater to be a source of ground current.

3rd

If there is ground current on the UPS EGC, is the UPS a separately derived system? Is the UPS neutral grounded improperly (UPS input and/or bypass has a neutral connected to the UPS output neutral)? Quite possible a neutral from the UPS distribution is connected to a non-UPS branch circuit neutral.

4th
Prior to an outage I would recommend you do a zero sequence reading of the feeders associated with the facility to isolate the panels and possible branch circuits involved with net current issues. In addition measure voltage, neutral to ground the voltage should increase (on a loaded system) as you move away from the main service or seperatly derived system. If voltage goes up and then starts back down you will be close to a neutral ground connection.

 

[waltppiva]

Re: Ground Current... Caused by motors?

Regarding the last response, the UPS is not a seperately derived system. There is no isolation transformer within the UPS. The neutral from the feed is terminated to the output neutral.

I do understand that if a neutral from a UPS distribution panel is returning back to a non-UPS panel (or any neutral returning back to a panel from which it is not fed, for that matter) will cause the voltage at the circuit to fluctuate.

Thanks again for the quick response. Hope everyone has a wonderful holiday.

 

[ron]

Re: Ground Current... Caused by motors?

Could you share why the UPS is not a seperately derivrd system, per the definition in the NEC? Why isn't the neutral bonded to the ground at the output of the UPS?

 

[dereckbc]

Re: Ground Current... Caused by motors?

waltppiva, a UPS may or may not be a SDS. It depends on the type of UPS. If it is a standby type then it is probable not SDS. If it is a dual conversion or interactive type found in most commercial and higher end applications it is a SDS.

 

[waltppiva]

Re: Ground Current... Caused by motors?

The UPS in question is a dual conversion "online" system, but because it does not contain an isolation transformer, which would create a neutral on the secondary side, it is not a separately derived source per NEC definition. Therefore the neutral from the input is tied directly to the output neutral, or essentially passed straight through. I agree that most larger industrial UPS systems are SDS's because they contain an isolation transformer. However, I have worked with a number of smaller systems (<20kVA) that may not always contain an iso xfmr. In this case the input neutral is passed through to the output.

 

[dereckbc]

Re: Ground Current... Caused by motors?

waltppiva, this sounds like a standby UPS. Is it a single-phase input?

 

[waltppiva]

Re: Ground Current... Caused by motors?

DerekBC, this is a 10kVA 3 phase UPS. The isolation transformer is an option with this model unit, but it was not ordered with this particular unit.

 

[charlie tuna]

Re: Ground Current... Caused by motors?

we had a large law firm call us in on a new ups installation (by others) that had this same problem. we found that the ups system was not bonded to the supply source. the installing contractor had bonded the computer floor to the ups system and that was it. i can't remember the manufacturer (started with a "d") but it caused the ups to be very slow transfering to the battery during a loss of power. i think it is normal to read current on the main grounding jumper and if you had field connections between the neutral and the ground this would reduce this current reading at the main grounding jumper. if the system is installed correctly, it really doesn't care what current is flowing through the main grounding jumper. it should see a common system ground potential as a reference point and consider this potential as "zero". i agree with others, most of the techs associated with this equipment know only the electronics end and when there's a problem and they see current on a grounding jumper they go nuts!

 

[dereckbc]

Re: Ground Current... Caused by motors?

Originally posted by waltppiva:
DerekBC, this is a 10kVA 3 phase UPS. The isolation transformer is an option with this model unit, but it was not ordered with this particular unit.

OK, do you know the manufacture and model number?

If it is 3-phase,(I assume 208) what does the neutral have to do with it? There should not be one on the input.

 

[hurk27]

Re: Ground Current... Caused by motors?

Sounds like the neutral is bonded at the UPS and at the panel and this would cause a paralleled neutral if the neutral feeds straight through this ups.

 

[waltppiva]

Re: Ground Current... Caused by motors?

Charlie Tuna: If there are field connections from neutral-ground, this will create parallel paths for the neutral current to flow onto the ground. I agree that the UPS will use a common system ground potential as a reference, but if that ground reference has current which is fluctuating (when equipment in the facility is turned on and off) then the ground reference for UPS controls and power supplies will be fluctuating as well. I have seen this cause abnormal behaviour with other UPSs in the past.

Derekbc: The unit in question is a Powerware
9330-20 10kVA 3 phase 208 in/208/120 out. Because there is no neutral created on the output (as it would be on the secondary of an isolation transformer), the neutral is pulled with the input conductors and passed through the UPS to the load panel.

Hurk27: I'm not sure if the neutral is bonded in the load panel. I will be getting a maintenance window after Jan 1 when I will troubleshoot their entire site wiring for improper bonds.

Thanks to all for the exchange.

 

[charlie tuna]

Re: Ground Current... Caused by motors?

walt,
what i was thinking with field "neutral to ground" connections would reduce the main ground jumper reading ? right? it will really be hard to isolate and find any connection in the field between neutrals and ground???? how are you going about doing this????
i agree the potential on the ground will fluxtuate with current flow - but the problem we had was the ground seen by the ups - was not the ground seen by the power source. the ups was monitoring it's ground/neutral potential to it's phase voltage and was slow to respond to a power outage.. i remember the cost of the ups was $30,000. because thats what the difference in my price and the contractor who got the job and i thought "he forgot to include the ups?"---wrong! he had included the ups.. where he figured he'd save was letting two residential helpers wire an office space. man, what a mess! one two inch pipe out of the branch circuit service panel with 84 wires in it---whats the deration factor??? 50 amp copy outlet fed with #12 wire. high hats in the confrence room wired fixture to fixture with peices of extention cords-orange-yellow-round flat-you name it! exposed 277 volt/120 volt stepdown transformers just screw off onto metal studs to feed emergency high hats. the law firm hired me to supervise the repair without liability. i would inspect from 7:00am till 9:00 am and they would have to work at night to correct my items. first couple of days they just half a$$ed it - and i caught every short cut - they worked for three months - my bill to the firm was about $10,000....

 

[don_resqcapt19]

Re: Ground Current... Caused by motors?

Walt,
Are you sure about the information on this UPS?? All "double conversion" units are SDS. The spec sheet for that unit shows that the input is 3 phase 3 wire and that the output is 3 phase 4 wire. There is no connection to the supply grounded conductor unless you are using the bypass switch to transfer the load to utility power. The following quote is from the manual for this UPS. Note the manual is 262 pages and will take some time to download.

The UPS system functions automatically to supply AC electrical power to the critical
load. There are four standard operation modes:
? In Normal mode, the critical load is supplied by the inverter, which derives its power
from rectified utility AC power. In this mode, the battery charger also provides charging
current for the battery, if needed.
? In Battery mode, the battery provides DC power, which maintains inverter operation.
The battery supports the critical load.
? In Bypass and High Efficiency mode, the critical load is directly supported by utility
power.
? In Maintenance Bypass mode, the critical load is directly supported by utility power, but
the UPS can be shut down for maintenance. This is accomplished by the operator
moving the Maintenance Bypass Switch to Bypass.

The critical load is always supplied from the inverter. There must be a grounded conductor in the UPS output. The is no way to use the hots from the UPS inverter and the grounded conductor from the utility supply.
Don

[ December 29, 2003, 09:18 AM: Message edited by: don_resqcapt19 ]

 

[waltppiva]

Re: Ground Current... Caused by motors?

Charlie Tuna: My plan to isolate the N-G bonds in the electrical system is first to unplug all PCs and equipment from 120V receptacles. Then starting at the Main Feed panel and/or generator ATS measure continuity between N-G. At the distribution panels if there is continuity from N-G, I will disconnect the ground wires from the gnd buss bar one at a time until the continuity clears. Then we know what circuit is bonded. I will have an electrician with me to correct the issues as I find them. I've done this a number of times at Lowe's stores that would have multiple N-G bonds causing lots of abnormal behavior with the UPS and PC loads locking up, etc.


Don: I should have mentioned before, the UPS does have an options cabinet with a maint bypass switch. I read the installation manual and it does specify to feed the unit with "phase A B C and neutral mains and bypass input wiring from source to the bypass input terminals in the options cabinet"... "After wiring the options cabinet to the facility power and critical load be sure to ground the system according to local and/or NEC". This way when the unit is in maint bypass there will be a neutral passed through to the load distribution panel.

 

[dereckbc]

Re: Ground Current... Caused by motors?

Don we must be working on this around the same time. I just got the spec sheet and printed it out and concur the unit has to be SDS. It is 3-wire in, 4-wire output.

With a dual conversion UPS (rectifier-battery-inverter), the output has to be SDS, no way around it. That is why I asked what kind of UPS the unit is.

 

[hurk27]

Re: Ground Current... Caused by motors?

Walt I'll bet bottom dollar that this unit's switch is In Bypass and High Efficiency mode. This would explain the longer time for transfer and the current on the grounding conductor. When the switch is in normal mode the load is already on the inverter and there is no switching but In Bypass and High Efficiency mode. It has to transfer to the inverter.

 

[don_resqcapt19]

Re: Ground Current... Caused by motors?

Wayne,
When the switch is in "bypass" it can't transfer to UPS power. The bypass switch disconnects the load from the UPS output and connects it directly to the utility supply.
Don