Power Loss Terminology?

[BrekOne]

I am an Electrical Engineering intern currently employed at a small Biotech facility.
I am writing to ask about whether there are any specific terms that you would use for describing power loss. For example I mean something like a dip in the power where one would notice the lights dim quickly for a second or two but not affect anything powered by the UPS. Also is there any standard terminology as far as certain amount of time that power would have to be gone before it is called a complete power failure. Would the gap in time, here at the facility it is 8 seconds, to transition from normal power to the back-up generator?
Thanks in advance.

 

[itsmoked]

The "dip" is considered a "transient" or "voltage transient".

If the voltage drops to zero for any length of time at all, then you have had a "power failure" or an "outage" or a "power loss".

If the voltage has dropped 10% below normal you have had a "brown out" or a "voltage sag".

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Zogzog]

"If the voltage drops to zero for any length of time at all, then you have had a "power failure" or an "outage" or a "power loss"."

Also sometimes refered to as "Dark"

"Would the gap in time, here at the facility it is 8 seconds, to transition from normal power to the back-up generator?"

You would have to check the manufactures info on the transfer time but typically the generator should be on line in 10 seconds, so your gut is about right. You should have some records of this acceptance test, somewhere.

 

[dpc]

http://grouper.ieee.org/groups/sag/IEEEP1564_00_09.pdf

If you can access this presentation, it should give you some leads for additional Googling.

The ITIC (CBEMA) curve is often referenced as the standard for short-term voltage fluctuations.

10 second loss of power on transfer to engine-generator is a common requirement. It's achievable if equipment is properly maintained and tested.

 

[BrekOne]

Thanks for the responses guy, they were very helpful. I have another question though. Do you think that NSTAR would have data available for these occurences even if it were just a voltage transient and not a total power loss or brown out?

 

[dpc]

What's NSTAR?

 

[BrekOne]

NSTAR is a electric and gas utility company. They provide the power for the building. For more info on NSTAR, you can visit:

http://www.NSTARonline.com

But to rephrase the question, would the company that provides to power for the facility be able to provide data for a voltage sag, power loss, etc if one occuried? I was just wondering is there a way to obtain that easily.

 

[davidbeach]

They probably could not. For the most part the utility metering is not sophisticated enough to record such things. If there is a repeating problem, they can often come out and make a temporary installation of additional metering equipment to capture and record such things.

 

[jghrist]

Information on outages may be available from the utility. It depends on the extent of their SCADA (Supervisory Control and Data Acquisition) system, their outage analysis system, and their willingness to spend the time to compile the data for a particular customer. The information that pertains to a single customer may not exist.

Normally information on voltage sags will not be available from the utility. This data usually comes from customer-owned metering, unless there is a special project set up for utility measurements. Voltage sag data gathering is an option on some newer utility substation relays and meters, so it wouldn't hurt to ask.

I would avoid the term "power loss" in this context. It may be confused with losses that are associated with current flowing through resistance (I²R losses).

 

[rbulsara]

It depends whether you are a user or supplier of electricity. For users the definition depends on the nature of your loads. If any power outage (transient or maintained) causes the key equipment to shutdown then it is considered an outage.

Typically computer type equipment will continue to work if the outage is less than one cycle. Lighting on the other hand may sustain longer outage.

If you are a user Refer to ITIC or perviously known CBEMA curve that defines duation of a power outage that is accetable for computer equipment. Since most facilities now a days have them as critical equipment, this 'definition' will apply to most users.

Most utility compnay will only call it an outage if they fail to restore power by automatic switching. So if a reclosure closed and holds in third attempt to clear a temporary fault, all of user equipment may be down, but utility compnay might say it was not an outage.

So the answer is it depends.....

 

[itsmoked]

A better question would be, "what you need this info for?".

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[jraef]

Generally that is usually something you negotiate up front with the PoCo (shorthand for Power Company). Most will offer "guaranteed" power delivery contracts where they will specify a list of acceptable (or you a list of unacceptable) transients and disturbances... for a price of course. If you want the cheapest power rate, unreliability is the hidden cost; it's up to the user to determine the real costs based on downtime losses etc., so the utilities only collect and store data on those disturbances if it is going to cost them revenue. Otherwise it is up to you to do your own monitoring and recording.

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[DanDel]

BrekOne, I've sat across the table from NStar reps for much the same issues of which you speak. The recordings that they provide are averaged over 15 minutes and across three phases, and are totally worthless for any short outages.
The only option is to use a sophisticated recording instrument yourself (RPM, Dranetz/BMI, etc) to show them what is happening at you facility.

BTW, the word 'transient' is typically used (in my experience, anyway) for a short (micro-to-milli-second) disruption, also called a 'spike', and typically caused by switching events.
Words used for longer-term events are 'sags' or 'surges', which describe voltage decreases and increases, respectively, over time spans of several cycles to minutes or hours.
An 'outage' has, I believe, been explained already.

 

[BrekOne]

Thanks DanDel,thats exactly what I was asking about. Do you think costwise it is worth getting the sophisticated instrument to monitor 'sags' and 'surges'?

Also thanks to everyone else who posted.

 

[jghrist]

Have you had production losses or equipment damage that you suspect were caused by sags or surges? You have to weigh the cost of these against the cost of the instrumentation. If the sags and surges are frequent, you could hire someone to monitor the system with portable instruments for a short period of time and analyze the situation.

 

[DanDel]

If it's causing significant problems at you facility, then, yes.
It may make more sense to hire a testing compnay with this type of equipment to do the monitoring and write the report and help you present it to NStar with your concerns.

 

[stevenal]

The utility may not have ready data they can pull up, but they may have portable instrumentation they can install for monitoring. Call and ask to speak to someone regarding power quality.

 

[BrekOne]

Thanks for the help everyone. I will look into this and get back with results or anymore questions.

 

[wfowfo]

The instrumentation may not be as exotic as you might imagine. Most modern solid state revenue meters have more functions than you can imagine (power quality being one of them) and can be purchased for less than $1000 depending on the required need of sophistication. For instance, if you just want to see Minimum/maximum levels and outage counts, it might just be a few hundred dollars. If you want to look at wave shapes for the duration of the anamoly, it will cost more.

Also, many of the newer relays (Schweitzer for example) have "event capture" capabilities that give sequence of events as well as magnitude and duration values for viewing whatever occurence you program it to look for. If your POCO doesn't have these, the relays themselves aren't that expensive either (particularly if you can justify the expense by adding in their protective capabilities).