Power Analyzer recording interval

Status
Not open for further replies.
A

AAEB

Member
Location
Southeast Asia
guys,

what is the recommended recording intervals to capture power system's behavior? as of the moment, we are doing it in a 1 minute interval but i am not that confident that we are capturing all of the electrical events.

thanks!
 
I don't know what the interval should be, but 1 minute is much too long. Electrical disturbances a few cycles long can effect computers.
 
110214-0836 EST

AAEB:

You have to define your goal.

Generally from a practical point of view you want to sample something in excess of 2 times the maximum frequency component you want to observe.

There are tricks to employ to get short time events over long time periods.

If you want a clean look at transformer inrush current at 60 Hz, then I would want to sample at 8/256 milliseconds, or about 100,000 samples per second. Note: 8.3 milliseconds is 1/2 cycle at 60 Hz. See my scope photo P6 at http://beta-a2.com/EE-photos.html . Sixteen samples might be sufficient for useful information, but I generally like clean waveforms.

When you collect massive quantities of data you need ways to work with the data and extract the desired information.

At home on my power monitor the maximum sample rate is about once per second. In a day this is 86,400 samples. Or for both voltage and power at 4 + 6 + 4 + 10 ASCII characters the total file size per day is about 2 megabytes. This is the maximum rate this commercial product provides (TED System). The rate is too slow to easily distinguish the startup of two different freezers. However, if your are not trying to distinguish between different motors for signature analysis, then the rate is more than adequate for most purposes.

.
 
AAEB said:
what is the recommended recording intervals to capture power system's behavior?
. . .capturing all of the electrical events
Click to expand...
I had a similar problem. I had to sample battery voltage during discharge and switch off the battery when the cell voltage dropped below the cutoff point to prevent cell damage.

It turned out what I didn't know but really needed to know was how rapidly the battery voltage dropped off near the cutoff point. I guess the ideal sampling scheme would have been to increase the sampling rate as the cell voltage dropped.

Long story short, try sampling at 1 minutes, 30 seconds, 15 seconds, 8, 4, 2. . .and so on until you capture 95% of the significant events, however you define "significant events".
Your problem sort of comes down to predicting the future but if you have historical data on this system you can narrow your sampling options.

For real sophistication, increase the sampling rate if the previous samples show increasing disturbances, i.e., an adaptive sampling scheme. There are many schemes for the smoothing of time-dependent data.
http://www.stats.gla.ac.uk/steps/glossary/time_series.html
 
Last edited:
thanks guys!

our goal is to check the incoming utility voltage activities... we are currently trouble shooting the "nuisance tripping" occuring on our switchgear.

i'll try to increase the memory capacity of the analyzer and set it to 5 seconds intervals...

thanks again..
 
110214-2331 EST

AAEB:

My suggestion is a data collection system with a 1,000,000 sample storage. Then sample at maybe 200 samples/second (5 milliseconds), and stop data collection several seconds after your trip event. Suppose it is 10 seconds afterwards that you quit. Then you have 2000 samples after the event and 1,000,000 - 2,000 before the event.

.
 
AAEB said:
thanks guys!

our goal is to check the incoming utility voltage activities... we are currently trouble shooting the "nuisance tripping" occuring on our switchgear.

i'll try to increase the memory capacity of the analyzer and set it to 5 seconds intervals...

thanks again..
Click to expand...
5 seconds is an awful long sample time for a breaker. You will probably miss most of the short time and all of the instantaneous portions of the breaker trip curve. Most system GF protection maxes out at .5 sec.

What size breaker? What is downstream from the breaker?
 
What is the manufacturer and model of the monitor you are using? Is it a permanent installed monitor, or something you set temporariliy? Does it have waveform capture capability? The best type of monitor to use for this application is something that you can set to trigger a waveform capture based on the voltage and/or current magnitudes. That way, your RMS interval period can be relatively long (such as 1 minute) to preserve memory, and you will get a close-in look at the waveforms when something abnormal happens.

Ediit to add: My monitor has the capability to write over old data when the memory fills up. If you can't trigger waveforms, you could set the interval small and when the memory fills up, keep overwriting it until something does happen, then quickly download it to get the good stuff.
 
Last edited:
My recommendation is to have two logs - one periodic and another one trigger based. This should be easy to set up with most devices (which one you have)?

Both logging should be as fast as possible and should have as much information. How often can you check the device? Divide this time with the memory to have your log period and set the trigger threshold to a level that will provide you as much information as possible.

If you can access the meter immediately after an event occurs, log at least every second and set the threshold for THD level (this will help you detecting transients). Some meters have a possibility to log a sequence and to stop a logging based on condition. In such cases, you can stop the logging once the current is zero (assuming you still have voltage from battery or UPS).

A list of my personally recommended equipment appears in my website. It also include some pros and cons for each one.
 
I agree with the OPs that continuous monitoring will likely use too much storage memory. A PQ monitor with a "report-by exception" feature (settable trigger thresholds) may be the best way to perform your investigation.
 
So this becomes a balance of the cost of having a monitor babysit a breaker that trips at random intervals for no discernible reason
and
the value of finding the cause of the problem rather than just replace parts starting with that part most likely judged to be the cause.
 
How about a Ideal VPM? Mine caught a drop to 51V that lasted 33.5 cycles yesterday.

Hmm, how come the "Ideal VPM" doesn't show up as the hyperlink that it is unless you run the mouse over it???
 
Last edited:
AAEB said:
guys,

what is the recommended recording intervals to capture power system's behavior? as of the moment, we are doing it in a 1 minute interval but i am not that confident that we are capturing all of the electrical events.

thanks!
Click to expand...

when we are looking for the steady-state numbers, we usually do 10 minute intervals on our recording equipment. when we want to capture voltage disturbances, we set the triggers to above 10% or below 10% the voltage we are monitoring. since our equipment takes 256 samples per cycle, we are confident that all voltage events are captured.

we use a bmi dranetz 4400 for our monitoring
 
There's a device called drive-cam meant for commercial drivers. The camera is recording every minute of driving, however it doesn't have a huge amount of memory. It is continuously over-writing the old portion. When the on-board accelerometer senses an impact, it will retain a minute or so before the collision then continue to record for another minute.

Incident logging type PQAs should have event +/- xx minutes logging.
Even the basic logging on Fluke 189/289 can do this. If you set the logging for 5 minutes, then set for "15%" as event threshold, a deviation of 15 percent will trigger an instant log point. Granted, it doesn't have the ability to perform harmonics, swell, sag, dip visual analysis that extremely expensive Dranetz can, but it is useful in many applications if you know how to use it.
 
Electric-Light said:
There's a device called drive-cam meant for commercial drivers. The camera is recording every minute of driving, however it doesn't have a huge amount of memory. It is continuously over-writing the old portion. When the on-board accelerometer senses an impact, it will retain a minute or so before the collision then continue to record for another minute.

Incident logging type PQAs should have event +/- xx minutes logging.
Even the basic logging on Fluke 189/289 can do this. If you set the logging for 5 minutes, then set for "15%" as event threshold, a deviation of 15 percent will trigger an instant log point. Granted, it doesn't have the ability to perform harmonics, swell, sag, dip visual analysis that extremely expensive Dranetz can, but it is useful in many applications if you know how to use it.
Click to expand...
A Fluke VR1710 would work also.
 
robbietan said:
when we are looking for the steady-state numbers, we usually do 10 minute intervals on our recording equipment. when we want to capture voltage disturbances, we set the triggers to above 10% or below 10% the voltage we are monitoring. since our equipment takes 256 samples per cycle, we are confident that all voltage events are captured.

we use a bmi dranetz 4400 for our monitoring
Click to expand...

I also do 10 minute intervals for my tests with voltage trigger parameters set at +6% and -10%. I use a Dranetz Power Visa and an Amprobe DM-III multitest.

Some pics of the meters, this is my first time posting on this forum so if the pictures do not show up please let me know and I will change the soure code.

IMG_0161.jpg


photo.jpg
 
Status
Not open for further replies.
Top