Noisy Power Supply

Not knowing where you are or the age of your building it is rather hard to tell. You might inquire as to the age of the primary transformer(s) in your building vault, and you might also ask the local electrical utility to look into the noise on the line.

O ok, sorry my mistake :)

I work on a mining site, so we manage all the power ourselves. And the UPS is currently installed underground inside a substation.

Power to the UPS is current fed from a 1000/440V TX -> 440 Sw/Bd -> UPS.

I'm guessing that I should be be investigating into the switchboard that is feeding the UPS at the moment, corrent? But what sort of load should I be looking for? Capacitive loads (lighting circuits)? or...?

I guess what my real question is that, what sort of a load could cause a noisy power source?

Older motors, have a nasty way of making excess noise. Elevators, convener belt drives, any motor that might be under a variable load. Checking the torque on motor power bus connections would also be a good place to check. Could even be an auxiliary generator. Sort of a treasure hunt to find the source.

Perhaps isolation transformers are in order.

Any help?

Stone crushers are more likely to experience impulse loads that will cause rapid line voltage variations.

Also, remote mines have long transmission lines, the voltage fluctuation could also come from the high voltage even if it is less likely. Look for impulsive loads on your site first.

The noise due to fan motor within the ups. How you are fixed the ups? In front and back how much space is provided? What is the back up sourse? Give details.I will give you solution.

So you saying that the noisy power input could be caused by the 'fan motors' on the UPS itself? I'm not too sure if this could cause noisy input voltages though...but you sounds like a man that knows the UPS design better than I do :)

Here's the details on the UPS. The UPS is bolted onto the ground, in an air conditioned open area with lots of space in front where the fans are, about 2~3m clearance around the back (lots of space on side as well 5m at least).

The UPS has input from 440V board and the bypass is also from the same board, probably not the best way idea...but no other 440V board is available...

Hi DanceWithTrees,

Definitely alot of help, I think I'll be looking at the equipments more upstream than where I'm currently looking, I guess I should be looking at equipments like vent-fans or conveyor belts off the 1000V board, and look for any sign of knocking/rumbling, correct?

Anything that could cause momentary excess loading. You may want to take a look at the output from the generators, they may be the ultimate source of the voltage fluctuation.

Good luck and please let us know what you find.

Another problem might be that the UPS itself is causing harmonics back on the power supply. See if you can identify the frequency of the harmonic components. That might give you a better idea.

Hi.. I feel it's only UPSs monitoring system that is taking pick-up from the line due to IGBT / Thyristor switching. Please check the incoming voltage externally with Fluke Power Quality Analyser and there should be nothing.If you still find the noise, check as follows :

The noise goes while UPS is transferred to battery power, it's the UPS Input Current Harmonic Distortion causing the problem.

BTW which brand & model of the UPS is this?

Rgds

Sanjeev

The problem seems to be more of a Voltage fluctuation than a noise. If the voltage is going from 266 to 271v over minutes then it is probably caused by intermittent loading from other users of that supply line. On a mining site I guess you will have a lot of machines switching on/off all the time and longish cable runs. My suggestion is to run a separate supply cable from the point where the supply comes on site to your UPS and see if this stabilises the voltage a bit. If the incoming supply has other phases try each one and choose the phase with least voltage variation.

If the problem has only started since you installed the UPS, (presumably for an office?) then you might consider getting a bigger CVT or changing the UPS for one without the monitor - it's really only a glorified battery charger on the input side.

Hey Jack,

I had a similar problem with a 150KVA UPS on a 480 volt line. I found that the monitoring voltmeter (inside the UPS) needed to be "adjusted" (calibrated?) to the middle of the range of variation. This put an end to the alarms.

The utility company will tell you that Voltage is king. This means that they attempt to regulate the voltage for any current drawn. The problem is that not all currents are sinusoidal at the fundamental frequency of 60 Hz. In particular, some older equipment (especially DC power supplies) that is still in use may take gulps of current at the peak of the sinusoidal swing. This causes harmonic distortion which can cause a large variety of problems. You can Google harmonic distortion for more information.

But the solution is to have a large step-down transformer dedicated to feeding your panel with other noisy devices on a separate feed. The odd harmonics (3rd, 5th, 7th, etc) all work together to create noisy power. A large iron core transformer will convert a lot of that energy into heat and is the usual solution short of eliminating the noisy equipment in use.

The easiest way to detect harmonic distortion is to put a current clamp on the neutral line of your power distribution panel with a frequency meter attached. If the power is fairly clean, it will see 60 Hz power. If not, it will see 180 Hz. Third harmonic current (180Hz) loves the neutral line of power distribution equipment. In fact it is so hot for the neutral that it frequently overloads (overheats) it. Look also for evidence of melted insulation at the termination of feeder lines in nearby panels.

In my R/D I have found that, "power factor" voltage-vs-current does weard things to circuits. mabe that phase correction capators can help.

In his type of applications, a standard capacitor bank is more likely to cause troubles than fixing anything. There are fast voltage variations that will not be tracked by the standard slow controllers and contractors.

If he is going to install any compensation, he must look at a fast capacitor switch and controller that react within a line cycle. Send me an email through this site if you want more details. We have solved this type of problems before for oil rigs.

Hard to analyze issues like this from a distance.

You might want to check and verify that the Thyristor/IGBT gate and biasing leads (usually red & white or black & white) are properly twisted together. If these leads (wires) are not properly twisted together along their entire length, your system will suffer severe "hammering" (noise) and damaging vibration due to the "on - off - on" transition being too abrupt. I have solved many site noise/vibration issues such as this where the factory assembly personnel simply failed to complete the equipment assembly correctly.

Also be aware that connection of a UPS to a voltage source that is being supplied by a "constant-voltage-transformer' and/or line filtering equipment often results in the devices (UPS vs. CVT vs. Filter) "fighting for control" which can cause a lot of noise and can cause failure of any or all components depending on the magnitude of the voltage the disturbance creates. Basically the system dynamics in this type of case can mimic voltage "doubler" or "tripler" circuitry (diode + cap + resistance) and the induction caused by rapid collapse of the system conductor magnetic fields when de-energized will significantly add to this destructive scenario.

If none of the above solve you issue, it may be necessary to add inductors to your power supply in order to stabilize and prevent rapid changes in systen conductor current. The size and rating of the inductors should be decided by a reputible Power Quality Service Field Engineer. A power quality study for a small system is relatively inexpensive and sometimes "money well spent". In some cases the vendor will perform the study for free in order to make a sale. Contact GE, ABB, Siemens, Westinghouse, or other vendor of your choice and they will provide good information and/or a solution.

Hi guys,

Just want to say, thank you all for your inputs first. I'm glad I found this forum, you guys have been very helpful in providing different ways to tackle this problem.

At this moment, I'm going to put a temp metering unit at the switchboard to isolate the problem to see if the voltage variation is occurring inside the UPS (switching of IGBT inside the UPS) or if the problem is coming from an upstream source (eg. motor/fans/crushing units).

Also we've found that the voltage variation appears to be intermediate as well. So, I think this could probably suggest that there's something affecting the input voltage when it's turned on, it's just that we don't know which exact equipment is actually affecting it. But having made this assumption, I don't think I'll be ruling out the problem the UPS rectifier circuitry just yet. Once we get the metering unit hook up we will know where to start looking :)

And, some details on the UPS, it is an Eaton/Powerware 9355, 30kVA, 415V/50Hz unit.

I notice that none of our Australians have responded. Could it be that your noise is in your Mains? They are notoriously noisy. We had a good discussion on this about 6 months ago.

Search all of CR4 for "Australian power". This will lead you to a discussion involving American equipment in Australia "exploding" (along with many other discussions). That is not the exact title, but you should see it. If that does not help, you might send a PM (private message) to Sparkgap. I would say HE is our resident expert on Aussie power.

If that doesn't work, come back to us.

Sincerely

Bill

I just looked a this thread. You have a Wye 415/240 volt system. If you are getting voltage readings to neutral above 240 as you have reported from your logger, it means you have some problem with the neutral. That is the only way for higher voltages to occur. The worse the neutral connection, the closer to 415 you will see, depending on where the loose neutral connection is - you may have a voltage dividing issue across the lines through a neutral junction where the junction's path back to the neutral buss is challenged.

Hope you haven't spent too much money on this so far. CJM

Hi guys,

Just thought I come back and follow up on the problem that I posted earlier.

As stated before, there was an over-voltage problem and noise on input supply. We solve the over-voltage by replacing the supply transformer to the UPS, with a multi-tap transformer. And carefully tap it so that it will stay within the operating voltage with the UPS.

As for the noise from the input, I found out there were multiple motors use for crushing and conveyor on the supply end which was causing problem. The main reason that I realised this was becasue I found the noise to by 'kinda' periodical...and further research found this to be our main problem. The way we thought about solving this was to purchase a power conditioner between the new TX and the UPS, but we end up not doing it. Because with the capacity that we were looking for, needed to be specially manufactured/engineered...=$$$.

The system is working at the moment without the power conditioner, as we're using the UPS power inverter as the power conditioner. Spoke to the manufacturer before, they confirmed that it shouldn't cause a problem, as the UPS was design to accept 'whatever' voltage (as long with range), it will produce a smooth output.