2. What is a Power Quality Problem ? “ Any occurrence manifested in voltage, current, or frequency deviations that results in failure or mis-operation of end-use equipment.” Dictionary
3. What does that mean? It’s dependant on your susceptibility. Given the quality of supply do I have to worry about problems with my equipment or systems?
4. Typical Financial Loss Per Event Source: The Cost of Power Quality, Copper Development Association, March, 2001 Industry Typical Loss Financial $6,000,000/event Semi-conductor mfg. $3,800,000/event Computer operations $750,000/event Telecommunications $30,000/minute Data processing $10,000/minute Steel/heavy mfg. $300,000/event Plastics $10,000-15,000/event
10. A transient power quality event has occurred on DataNode H09_5530. The event occurred at 10-16-2001 05:03:36 on phase A. Characteristics were Mag = 478.V (1.22pu), Max Deviation (Peak-to-Peak) = 271.V (0.69pu), Dur = 0.006 s (0.35 cyc.), Frequency = 1,568. Hz, Category = 3 Upstream Capacitor Switching
11. Possible Causes • PF cap energization • Lightning • Loose connection • Load or source switching • RF burst Possible Effects • Data corruption • Equipment damage • Data transmission errors • Intermittent equipment operation • Reduced equipment life • Irreproducible problems Transients
24. Possible Causes • Sudden change in load current • Fault on feeder • Fault on parallel feeder • Motor start • Undersized distribution system Possible Effects • Process interruption • Data loss • Data transmission errors • PLC or computer misoperation • Damaged product • Motor failure RMS Voltage Variations Causes and Effects
33. An integer multiple of the fundamental frequency Fundamental (1 st harmonic) = 60hz 2 nd = 120hz 3 rd = 180hz 4 th = 240hz 5 th = 300hz … What is a harmonic?
42. Harmonics (sustained) Possible Effects • Overload of neutral conductors • Overload of power sources • Low power factor • Reduced ride-through Possible Causes • Rectified inputs of power supplies • Non-symmetrical current • Intermittent electrical noise from loose connections
Very specific answer depending on what is happening within your facility. How sensitive and susceptible to power quality events is the key equipment; what is your incoming power like; how sensitive are your critical loads.
Some of the documented costs on a per event basis. Large range. 7x24 mission critical financial and/or data processing, like credit card transactions are the greatest cost of downtime and could put the company out of business. But even a standard manufacturing process like plastics have a hefty cost when you consider stopping the process, dealing with scrap, restarting and recalibrating.
Reference this discussion to point of common coupling, or service entrance to the facility. Directivity is upstream or downstream to that monitoring point. Upstream from the service entrance is generally coming from utility or other customers. Downstream is within facility. Determining directivity helps determine where problem lies. Example case study: President of our company switched from using a laptop computer to desktop and was experiencing regular reboots and other computer problems. Couldn’t figure out why, until we realized that the laptop had battery backup, so every time there was a PQ problem he self-corrected. With the desktop, he essentially changed his susceptibility (made it worse). The problem was corrected after a study and rewiring exercise.
IEEE 1159 is the prevailing US standard This list presents PQ events that vary in duration from fastest to longest.
Transients are often called “ghosts” in the system, since they come and go intermittently and may or may not impact your equipment. Transients are much less than a full cycle, oftentimes even a millisecond or less. Characterization as listed can help the skilled user determine what happened and specifically how to mitigate. Some of the above: Rise time – change in voltage over time Point on wave – cant interject itself at any point in a wave Multiple zero crossings – Can impact basic clock functions.
Lots to point out on this slide: Positive or negative transients. Means that it adds (positive) or subtracts energy from the waveform. Unipolar – goes one way (either positive or negative (both are shown above) Bipolar – goes both ways. Notching – chop up the waveform. For example, pulse drives. Case study – As an example of multiple zero crossings, one of our utility PQ engineer customers hooked up our most recently introduced instrument (the PX5) to a digital clock with an air ionizer on the same circuit. When the ionizer was off, its clock blinked on and off once per second. When you turned it on, it blinked twice as fast. Turned out to be the interjection of a transient with multiple zero crossings.
On of the most commonly understood transients is caused by a capacitor switching event. This type of transient is called oscillatory and depending upon the susceptibility of your equipment, can cause damage, especially to computers and other PLCs. The above was captured using our Signature System. Note that the AnswerModule determined that it was an upstream event, category 3. Upstream, as in upstream of that data point – likely caused by the utility switching in a capacitor. Discuss Answer Modules – intelligent algorithms built into the Signature System that determine source and cause of PQ event.
Again, transients are like ghosts in your system. Often random, can be troublesome. Can cause a range of problems, some listed here. Others include tripping breakers randomly. Can reduce life of equipment such as electrostatic discharge.
These events are longer than transients – typically one cycle or longer. Remember that 60% of all PQ events are sags, or reduction in voltage.
First define nominal or normal voltage. This diagram uses +100 and –100 as the upper and lower (peak) limits from nominal. Your facility could be 120, 110, etc. Firs we see a 3-cycle sag (decrease in voltage), next a 3 cycle swell (increase in voltage), followed by a total loss of voltage – an outage. Mote most applications trigger on RMS but to make visualizing simple this is in PK.
Big cause of sags is the startup of a motor, or any other piece of equipment with a large inrush of current upon turning it on. A typical motor startup will draw 6-10 times the amount of power. Interesting factoid to share is that 25% of all electricity in the US is consumed by motors.
1159 not only defines terminology, but also characteristics. These are listed in time order, from shortest to longest. Remember, IEEE defines + or – 10% of nominal for swells and sags.
This is important because it helps you determine where the problem occurred. Refer to directivity as upstream (oftentimes that means source) or downstream (oftentimes, load.)
Customer called us following a sag that damaged elevator controls to help evaluate data. The customer had installed a Signature system on the input and output of UPS. As you can see, an upstream sag (from the utility) was recorded on the input side of all 3 UPS systems. Note that the Signature System sag directivity Answer Module tells the user where the sag occurred.
Here is the RMS trend and waveform data of one of the UPS system showing the waveshapes during the sag
But as you can see here, one of the UPS units had a corresponding swell at the time that the utility power went back to normal. Unfortunately, the UPS did not alarm the customer, who only learned of this potential problem by evaluating the Signature System data.
The bottom line is the Signature System not only recorded the sag impacting the elevator controls, but identified a new potential problem from the swell.
We are going to introduce a few PQ Rules of thumb that will help you understand how to interpret waveform data. In the first rule, the voltage and current both go down when there is a utility, or source-based sag.
One of many customers with permanently installed Dranetz-BMI systems impacted by last summer’s blackout that affected 8 Northeast states and Canada. This NYC office building had a complete outage at about 4 pm on Aug 14, returning again about 5:30 am.
Rule #2. When voltage and current go in opposite directions a load-generated sag is indicated. A prime example of Rule #2 is the startup of a motor.
Example symptoms of sags and swells are computer lockups, data loss, process control problems, overheating of motors, and ASD tripping
A range of options to display this critical set of PQ events. We have already shown how you can use waveform data to interpret the direction and characteristics of events. The following charts show other ways to look at this data.
The CBEMA curve (stands for Computer Business Equipment Manufacturing Association) was developed by evaluating the IBM 370 computer and its susceptibility to power quality events. The x axis is time, the y is RMS voltage. Inside the envelope are data points that are determined to have no susceptibility for the computer, while outside the envelope there may be some susceptibility and should be watched and evaluated.
The chart is an updated version of the CBEMA curve based on a broader range of computer equipment. Again, look outside the envelope for susceptibility. A custom curve can be developed for your individual facility or equipment based on its susceptibility to PQ phenomena.
Another perspective, prepared using Dranetz-BMI’s Dran-View software. This view adds the frequency of occurrence to the duration and RMS voltage of the 2 previous charts. Here you look at the “tall buildings,” as they are the events that occurred most often and should be evaluated and trended from a susceptibility perspective.
Will now introduce a case study that uses a common piece of office equipment to demonstrate how PQ problems can impact daily operations.
This simple timeline shows a voltage sag occurring every 45 seconds.
Laser printers have a heating element within that go on and off to maintain proper temperature. In this case, you can see that sag occurring when the heater is turned on as a result of the Inrush current.
As the heating element is turned on, we see an inrush and, if we remember our directivity rule, a downstream sag. We will return to this case study in a moment. But now, on to harmonics, perhaps the least understood aspect of PQ. Also note that since voltage and current are in phase it confirms the resistive heater load.
Simple definition. In the US, the fundamental frequency is 60 Hertz, so an integer would be 120, 180, etc. In Europe, that would be 50 Hz, 100 Hz, etc.
Here we see sine waves without harmonic current.
Followed by the introduction of harmonics, which are caused by non-linear loads on the system. This is a gross example of a very distorted, chopped-up current waveform. What distorts the voltage is the current being drawn against the applied voltage.
You should be concerned about harmonics when they occur over a long period of time, as they can have a cumulative effect on the power system. Electronic power supplies, ASDs and computer systems in general can add harmonics to a system and cause damage over time.
These methods extract frequency information by taking the waveforms and breaking it into individual frequency components using methods such as DFT or FFT. THD provides one number representative of all harmonic content. These parameters can provide guidance on derating factors and other compromises to equipment affected by harmonic content.
A common way of looking at harmonics, as presented by our Dran-View software is the harmonic spectrum, which depicts each individual harmonic.
This brings us to our final PQ Rule today Most power devices draw current symmetrically. For example, a customer of ours found themselves rich in even order harmonics. Upon analysis of the data and tracing back to the problem, it was discovered that the output of one of the UPS systems was only conducting electricity on the positive side, creating an imbalance of harmonics. Because they had a permanently installed system, they caught this problem before it could cause problems and remedied it.
The presence of triplen (multiples of 3) harmonics are cause for concern as they can add to the neutral phase. Example problems caused by this are overheated distribution neutrals and transformers or nuisance tripping of key equipment.
Note the pictorial addition of the 3 rd harmonic .
Non-linear loads, single phase computers, electronic power supplies and adjustable speed drives are all known culprits of harmonics. Plus, the addition of computers, servers and other microprocessor-based equipment can cause harmonic buildup. Harmonic studies are important exercises for facility managers to conduct on a fairly regular basis.,
Part 2 of our laser printer case study, this time focusing on the impact of harmonics.
Again, here is the waveform with the heater turned on. Not a perfect sinewave but very high in harmonic content
Here is the harmonic distortion displayed as a spectrum while the heater is turned on. Note the presence of the odd order harmonics, with a total harmonic distortion of 2.8% for voltage and 5% for current. IEEE recommends THD below 5%, so the situation looks okay.
Next we look at the waveform with the heating element turned off. Non-linear, rich in harmonics.
And more important, the harmonic content in the idle mode. Now we see that the THD for voltage is still within the IEEE recommended range at 3.1%, but the THD for current has escalated to 140%.
To summarize what we have learned. Note the final slide showing the flat topping of voltage when the printer is in idle mode. This phenomena can cause problems to other equipment on the circuit.
Frequency problems often occur with cogeneration units, when the generator is out of phase with the utility. Flicker is an important parameter in Europe and is defined as very small perturbations in magnitude of the voltage that manifests itself in lighting.
1 st peak on blue cycle shows positive transient, immediately followed by sag. Notching in 1 st waveform of black cycle. Frequency change toward end, prior to interruption.
2 basic approaches to monitoring with different instrument configurations to match. Many customers integrate the two, installing a continuous monitoring system on critical circuits, like pre-and post-UPS or the service entrance, and use a portable for troubleshooting or installing new equipment.
Dranetz-BMI’s newest portable instruments for troubleshooting, performing surveys, assessing load distortion/imbalance, etc. Widely used instrument. Key challenge with portables is that you measure after the problem occurred, hoping it will recur so you can analyze and mitigate.
Continuous monitoring-one system, multiple applications: Power quality monitoring – provides total picture 7x24 of system for proactive management Reliability-based monitoring – To set maintenance schedules, trend and evaluate continuous health of key circuits and equipment. Energy management – Understand load profile, shed or shift loads, eliminate high demand charge, control power factor Process and production management – Correlate power information with process parameters to maximize output.
From single phase clamp-on meters to sophisticated 3-phase portable monitors that capture every event.
Equipped with 8 independent channels, the 3-phase PowerGuide 4400 is the only advanced power monitoring instrument to incorporate a color touch screen into its lightweight design. Automated setups provide instant detection of circuits and configurations, ensuring that the instrument is ready to successfully collect data. The PowerGuide 4400’s unique annunciator “report card” provides instant power quality answers in the field. A wide range of power monitoring data is collected, analyzed and tabulated in color-coded categories to quickly identify areas of concern, which are identified in red. Drill down for more detailed information by simply touching the intuitive touch screen to locate the source and pinpoint the root cause of power quality disturbances.
Users can select the length and mode of data collection, including troubleshooting, data logging, power quality surveys, energy and load balancing. The PowerGuide collects data at 256 samples/cycle/channel, offers remote communications using RS-232, Ethernet or USB options, and meets IEEE 1159, IEC 61000-4-30 and EN50160 standards.
The PowerXplorer integrates the most advanced feature set available in a power monitoring instrument with an easy-to-navigate, color graphical user interface. With high-speed sampling and data capture (1 microsecond/channel), this 8-channel workhorse simultaneously captures and characterizes thousands of parameters, using a range of standard and customizable operating modes. The unique measurement capabilities of the PowerXplorer include capture of low-medium-high frequency transients through peak, waveshape, rms duration and adaptive high-speed sampling, as well as power measurements to clearly characterize non-sinusoidal and unbalanced systems.
The PowerXplorer uses digitized high-speed sampling to capture and analyze microsecond-wide transients (Dranetz 658-like and BMI 8800-like). Transients, generated by fast-switching electronics, medical diagnostic equipment, capacitor switching, transformer energization, research & development equipment and load shifting, are immediately characterized as impulsive or oscillatory and detailed for further analysis
Use this slide to describe the Signature System architecture. DataNodes – data gathering measurement instruments. PQ, Energy, Legacy and 3 rd party nodes such as Advantech Adam Modules and GE kV meter InfoNode – web server and analysis device. Can also talk about InfoNode on a PQ software in lieu of hardware option. AnswerModules – intelligent algorithms for determining the source and cause of PQ problems. Cap switch, radial line to fault, benchmarking, sag directivity, UPS performance, etc. Enterprise level Software – PQView for looking at multiple sites.
Conclude with simplicity of SS – How it collects data 7x24 and delvers answers to pager, email or contact closure. Stress user interface, web-browser access, multiple users, real time, trending, etc.