600 V Bus voltage fluctuation is 12% 2

[Arba]

I’m having a case where from my power monitoring on the main 4000 Amp, 600 V bus fed from a 4MVA, 44KV – 600 V Transformer I see a lot of harmonic generation from a 300 KVA compressor, and different electric welding machines of the plant. The average voltage during one week monitoring is just 1.59% lower then 346.41 V (600 V) Ref Voltage. The average load of the transformer is just above 25% of its nominal load. The average amps draining on ground wirer towards the transformer is 50 A, wile the average amp on main neuter is jus 6 A. Right now the max voltage is 616.61 V and the minimum one is 550 V. Without involving any filter, would I be able to deal with the voltage drop complain from costumer only by increasing the voltage by 2.5% and bring the max voltage to 632 V or this would stimulate more voltage spikes on the plant?

 

[DanDel]

Are the minimum and maximum voltages you mention instantaneous values?

If not, loading (or some defect in the system) must be an issue.

How is the voltage drop presented as a problem for the customer? Flicker/lights dimming, etc?

 

[waross]

The first step in problem solving is to define the problem.
The second step is to identify the source(s) of the problem.
The third step is to devise a solution or compromise to wholly or partially eliminate the problem.
It is often unproductive to skip the second step and propose a solution without first identifying the source of the problem.
Step two:-
Possible reasons for a large voltage drop at a panel.
Metering errors; Not likely if the customer is complaining about voltage dips. If there are no noticeable voltage dips, but the monitoring equipment is showing large dips youmay have a metering issue.
Feeders from the transformer to the panel:- This is a good place to look for voltage drops. The conductors may be arranged in such a way as to cause an unacceptably high impedance. some things to look for:
Feeders with insufficient ampacity.
Feeders too long for the unit impedance.
Feeder conductors arranged in such a way as to develop excessive inductive reactance. This may be something such as a duct bank with the conductors of each phase contained in separate ducts rather than being distributed so that each duct has an equal number of conductors from each phase with the neutral equally distributed among the ducts.
Measure the voltage drop on each phase conductor from the panel to the transformer.
Loose connections. Loose connections with a single conductor per phase will cause quite small voltage drops. More than a few volts drop across a connection carrying load current will develop so much heat energy that catastrophic failure will soon occur.
At 25% load you have about 1000 amps. With only a 1% voltage drop (6 Volts)across a loose connection, there will be 6000 Watts of heat developed. That much heat generated by a loose connection will quickly develop into catastrophic failure.
However, with multiple conductors per phase, poor connections on some conductors throws the load on the remaining conductors. This may cause excessive voltage drops.
Transformer regulation; Possible. Voltage drop in a transformer is caused by a combination of reactance (X) and resistance (R). (Hence X/R ratios)
At full load and unity power factor the voltage regulation is mainly caused by the resistance of the transformer windings. AS the power factor drops, more and more reactive voltage drop is developed. The full available short circuit current is only developed when the only resistance in the circuit is the transformer resistance.
What is the per/unit (%) regulation of the transformer?
Do you know the per unit (%) impedance of the transformer AND the X/R ratio?
This will give us the ability to suggest how much voltage drop to expect in the transformer.
Primary voltage drop:-
You may have a "soft" primary feed. Do you have access to primary CTs to check the primary voltages?
Many of the newer electronic kWHr meters have the ability to store a wide range of information. This may include phase voltages, phase amperages, peak voltages, peak currents, power factor, peak kW, peak KVA, peak KVAR, in addition kWHrs, KVARHrs, and KVAHrs. 15 minute intervals are common.
I have had good cooperation from some utilities regarding metering information.
The utility may be able to send out a meter man to download the information directly.
The utility may be able to send out a meter man to program the meter to provide the information in the future.
The utility may be able to send out a meter man to install a battery and program the meter.
The utility may be willing to install an electronic meter with recording capability.
You will probably be billed for some of these services. At the most, being charged for the supply and installation of a plug in meter with the wiring, Potential Transformers and Current Transformers already in place and paid for is by far the most economical way to acquire valuable data.

Your problem may be a combination of primary voltage drop and secondary drop. See if you can get more information and we can provide more assistance.
In regards to the neutral current;

The average amps draining on ground wirer towards the transformer is 50 A, wile the average amp on main neuter is jus 6 A.

I don't quite understand "main neuter is jus 6 A." Not the place to guess a meaning or make an assumption. Canyou explain a little more please?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rasevskii]

Waross has said it all, but there are a couple other items I might add--

-How does the compressor generate harmonics? Is this a straightforward induction motor using reduced voltage starting or is it a VFD? A 300kW unit must be running continuously. If it is a VFD then some of our VFD experts may want to comment.

-What about the welders? Are these rectifiers or something else? If so a good source of harmonics as well as the VFD if that is the case.

You have 50 Amps in the Ground wire? That sounds like a harmonics issue. How are you measuring that?

Your power metering may be doing strange things if harmonics are present all around. An ordinary DVM will not be accurate, and in any case too slow to see any fast swings such as from welders.

Is there actually a flicker problem in the lighting? Then first refer to Waross procedures.

In this case it may not be possible to eliminate flicker, due the transformer impedance and voltage drops in cables and the primary supply.

Is this a new problem or was it always that way from the original installation?

Are you using temporary clamp on CTs and VT connections for the power monitoring or are there actually voltmeters on a panel one can read? How are you recording voltage and over what time period? Try some old fashoned analog meters...

Just some thoughts..rasevskii

 

[racobb]

Bill pretty well hit it on the head, and I think he was trying nicely to give you some good advice. He will corret me if I am wrong. It appears that you have taken some info, which may be good or bad, and attempted to extrapolate it to define all of your problems.

First off, it appears you have a tremendous voltage swing of 550 to 616 volts or about 12%. Strikes me as unusual. Raising the tap 2.5% will get you to about 565 to 630 volts which is still a large swing. That is assuming that these are steady state values and not some sub-cycle dip taken from a recorder during load events in your plant. If it is steady state then you are served from an extemely weak utility source, or their voltage regulating equipment is not working.....or both.

I would suggest that you look carefully at the info you have and where it comes from. If it doesn't become clearer at that point, then hire someone with the experience to help you sift through the data, or get some new data, and determine what the problem actually is and what is causing it.

Alan

 

[waross]

Thanks for the kind words friends.
If I haven't made it plain, I don't think that harmonics are your worst problem.
Arba
Harmonics may be adding to the problem maybe not, but what ever, I doubt very much if eliminating the harmonics will solve your problem. I suspect that if harmonics are part of your problem, they are a minor part.
If you are able to provide more information we can probably help find the root of your problem.
As we get more information, we will be trying to help you to determine whether the voltage dips are gradual, as a result of increased loading on the grid outside your plant or abrupt and momentary from energization surges inside the plant.
Different problems, different solutions.
Waiting to hear from you.
PS, 600V Are you a fellow Canadian?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rasevskii]

I should have also said that harmonics, if present, only would possibly affect the measurements that are being used.

Harmonics themselves would not cause the voltage fluctuations that are noted.

If flicker or no flicker, that is what matters.

regards, rasevskii

 

[LionelHutz]

That is a large voltage swing. As already pointed out, you need to figure out what is causing the voltage variation before you can take corrective actions. Hearing that the average load is only 25% of the service rated capacity raises a large red flag when you state the customer is experiencing these large voltage swings.

It sounds like someone already changed the taps to 616V to try and fix this problem. It also sounds like that solution has failed so just changing taps again will not be an acceptable solution. You will have to address the voltage swings, either by fixing the equipment causing it or by fixing the equipment affected by it.

You should not have 50A on the ground wire if there is a seperate neutral wire, regardless of harmonics. Even assuming a VFD, the harmonic distortion of the 300kW load should be about 30% measured at that load. Reflected back to the source rated current, it would only represent a few percentage of current THD. So, it's really not the big a load for the service.

 

[Arba]

Hi guys thank you very mach for your responds and time. I would try to be more specific.
Initially my costumer is complaining that at 208/110 V level, the voltage drops time to time for quite a period of time to about 95 V, 13.5% less than 110 V Ref. Further more he’s suspecting that hydro voltage variation is above its limit variation +/- 5%. 30 – 40 % of the load electronic controlled welders. The transformer impedance in % is Zt = 6.4%, but I don’t know the X/R ratio (I’m not sure if it can be approximate with a typical value of 6). This is oil filled ¾ MVA with fan ventilation. The high and low voltage on this service has been maintained regularly at once a year. This problem is not new they been experiencing for more than last couple of years, but because is happening on 208, level the production is not effected directly.
I installed the power monitoring meter on the main SWCHB. First I monitored for 30 min. the 300 KVA VFD compressor feeder, and than I installed it for a week on 4000 A, 600 V main buss. The plant is using a lot of electric welding. The summery of one week monitoring I’ve listed as follow:
Voltage, Current and Frequency Summaries
Voltage, Current and Frequency measurements on: Main 4000A Bus 11/17/09 13:43 from 11/17/09 14:43:53 through 11/24/09 13:43:53.

RMS. Voltages Value Date and Time
Phase A minimum 317.5V Nov 24 2009 10:23:53
Phase A average 340.4V
Phase A maximum 355.0V Nov 22 2009 03:58:53

Phase B minimum 317.2V Nov 24 2009 10:23:53
Phase B average 341.1V
Phase B maximum 356.3V Nov 22 2009 03:58:53

Phase C minimum 317.2V Nov 24 2009 10:23:53
Phase C average 340.7V
Phase C maximum 355.4V Nov 22 2009 03:58:53

Neutral minimum 366.2mV Nov 17 2009 20:38:53
Neutral average 521.1mV
Neutral maximum 1.342V Nov 17 2009 14:43:53

RMS. Currents Value Date and Time
Phase A minimum 531.0A Nov 22 2009 04:13:53
Phase A average 986.2A
Phase A maximum 1.984kA Nov 19 2009 16:08:53

Phase B minimum 566.7A Nov 22 2009 05:43:53
Phase B average 1.063kA
Phase B maximum 2.086kA Nov 18 2009 12:13:53

Phase C minimum 546.6A Nov 22 2009 05:43:53
Phase C average 1.024kA
Phase C maximum 2.012kA Nov 19 2009 12:43:53

Neutral minimum 2.014A Nov 21 2009 07:58:53
Neutral average 5.762A
Neutral maximum 42.83A Nov 18 2009 07:13:53

Ground minimum 28.64A Nov 17 2009 16:48:53
Ground average 50.52A
Ground maximum 174.89A Nov 17 2009 16:53:53

Voltage and Current Distortion Summaries
Voltage and Current harmonic distortion measurements on: Main 4000A Bus 11/17/09 13:43 from 11/17/09 14:43:53 through 11/24/09 13:43:53.

Voltage Distortion Value Date and Time
Phase A minimum 0.71% Nov 24 2009 10:58:53
Phase A average 1.18%
Phase A maximum 1.7% Nov 23 2009 22:28:53

Phase B minimum 0.73% Nov 22 2009 10:33:53
Phase B average 1.177%
Phase B maximum 1.66% Nov 24 2009 03:23:53

Phase C minimum 0.82% Nov 22 2009 10:38:53
Phase C average 1.349%
Phase C maximum 1.86% Nov 21 2009 22:13:53


Current Distortion Value Date and Time
Phase A minimum 3.05% Nov 19 2009 13:08:53
Phase A average 4.573%
Phase A maximum 7.42% Nov 21 2009 02:53:53

Phase B minimum 2.81% Nov 19 2009 17:58:53
Phase B average 4.284%
Phase B maximum 7.07% Nov 21 2009 02:53:53

Phase C minimum 2.71% Nov 22 2009 12:33:53
Phase C average 4.454%
Phase C maximum 7.38% Nov 21 2009 03:13:53


Voltage Flicker Value Date and Time
Phase A minimum 0.756 Nov 24 2009 13:13:53
Phase A average 1.252
Phase A maximum 1.739 Nov 21 2009 22:33:53

Phase B minimum 0.499 Nov 22 2009 05:23:53
Phase B average 0.905
Phase B maximum 1.520 Nov 22 2009 22:53:53

Phase C minimum 0.772 Nov 23 2009 14:33:53
Phase C average 1.332
Phase C maximum 1.883 Nov 21 2009 22:23:53

Power Factor Value Date and Time
Phase A minimum 0.917 Lag Nov 23 2009 01:33:53
Phase A average 0.846 Lag
Phase A maximum 0.454 Lag Nov 22 2009 06:48:53

Phase B minimum 0.932 Lag Nov 24 2009 04:53:53
Phase B average 0.863 Lag
Phase B maximum 0.431 Lag Nov 22 2009 11:48:53

Phase C minimum 0.919 Lag Nov 23 2009 01:13:53
Phase C average 0.840 Lag
Phase C maximum 0.36 Lag Nov 22 2009 11:48:53

Total minimum 0.899 Nov 24 2009 04:58:53
Total average 0.850
Total maximum 0.785 Nov 22 2009 11:53:53

Harmonics Values

The first 16 harmonics for the Phase A current are shown below (RMS Values):
Harmonic Max Amplitude Avg Amplitude
1 1930.592 986.661
2 190.987 3.237
3 80.279 11.653
4 56.325 1.295
5 84.811 29.781
6 35.608 1.295
7 62.799 23.307
8 31.076 0.647
9 23.307 2.590
10 22.012 0.647
11 28.486 12.301
12 20.070 0.647
13 21.365 8.416
14 15.538 0.647
15 15.538 0.647
16 13.596 0.647


The first 16 harmonics for the Phase B current are shown below (RMS Values):
Harmonic Max Amplitude Avg Amplitude
1 2058.133 1063.703
2 200.051 3.884
3 89.991 4.532
4 66.036 1.295
5 88.048 31.076
6 45.319 1.295
7 72.510 24.602
8 26.544 0.647
9 27.839 1.295
10 26.544 0.647
11 37.550 12.301
12 20.070 0.647
13 25.897 9.064
14 14.243 0.647
15 14.891 1.295
16 16.185 0.647

The first 16 harmonics for the Phase C current are shown below (RMS Values):
Harmonic Max Amplitude Avg Amplitude
1 1983.680 1024.858
2 282.920 3.237
3 148.258 10.359
4 113.298 1.295
5 103.586 31.076
6 66.036 0.647
7 73.158 22.660
8 49.851 0.647
9 46.614 1.295
10 44.672 0.647
11 53.088 12.948
12 33.018 0.647
13 41.435 9.064
14 27.191 0.647
15 28.486 0.647
16 27.191 0.647


The first 16 harmonics for the neutral current are shown below (RMS Values):
Harmonic Max Amplitude Avg Amplitude
1 22.206 5.568
2 1.748 0.039
3 2.279 1.398
4 0.609 0.013
5 0.958 0.298
6 0.440 0.013
7 0.557 0.194
8 0.233 0.000
9 0.220 0.129
10 0.194 0.000
11 0.246 0.129
12 0.207 0.000
13 0.194 0.091
14 0.168 0.000
15 0.155 0.052
16 0.129 0.000

The meter report doesn’t list the ground harmonic report. If I refer each of max, min, phase voltages to the average voltage value I see an approximately total voltage variation of (+4%, -6%). Based on those data the max level of the average variation is 11.25%. By low the max voltage variation from Hydro is 10%, (+/- 5%). Means hydro is just 1.25% above the limit. What about the harmonic contribution in the case? I don’t think this 6% voltage drop alone in 600 V would cause 13.5% voltage drop on 208 levels. This is based on IEEE Std. 519-1992, paragraph 5.3.1, page 29. Would any body gave me an opinion about that?

 

[waross]

From the use of 600 Volts and 208 volts I am going to guess that you are in Canada.

Initially my costumer is complaining that at 208/110 V level, the voltage drops time to time for quite a period of time to about 95 V, 13.5% less than 110 V Ref

The standard voltage and the voltage corresponding to 208 Volts is 120 Volts.
That would make your 95 Volts almost 21% low.
I suspect that you have at least two issues;
1> variable and often low utility voltages.
2> As serious voltage drop issue somewhere between the 600 Volt panel and the 120 Volt point of utilization.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rasevskii]

Is the 120/208 supply fed from a separate 44kV/120/208 oil transformer or is it from a 600/120/208 transformer? It sounds like an outdoor unit actually. If that is from the 44KV then it has to be a weak supply from the utility.

If the utility metering is on the 44KV side, there must be a set of PTs that a recorder could be connected to see what sort of voltages the 44Kv has over a time period. That would be a first step.

If in fact the 120/208 is fed from the 600V then other problems exist as Waross has said in his initial post. A check of connection conditions at the main panel is strongly indicated, especially if deterioration over time
has occurred.

There are obviously harmonics, but that would only affect the metering you are using, not the voltage drop, in my opinion.

Is your power analyzer giving only digital readouts or is there a possibility of an analog style chart record over a time period, like say 24h, when the plant is in production?
That would say a lot more.

Would the utility agree to set up their own recorder at the site? That would be another good move.

What was the original complaint, and how long was that after the plant was originally built? In other words was it a problem right from the original startup or later? Some detective work is indicated...

regards, rasevskii

 

[racobb]

I still have to ask about your recorder data. The average voltage looks pretty good. Over what time period are your max and min values taken? How often do they occur? If they are sub-cycle or a few cycles in duration, then it is probably from the plant loads and doesn't mean much other than the load is changing maybe when the compressor starts or a welder draws the initial arc.

If it is happening often enough, you may well notice an irritating flicker. Changing taps does nothing for the flicker though.

Are you chasing a flicker problem or long duration voltage sags? What exactly is your recorder reporting in your posted flicker data?

The harmonics while not too bad are probably from the welders. There is also the issue of their duration as well. Ask the welder Mfg if they produce 5th and 7th order harmonics.



Alan