Measuring Transient Inrush Current

Status
Not open for further replies.

philly

Senior Member
When measuring current with your average clamp on meter is the meter simply displaying the RMS current? In other words the peak of the current is 1.4 x higher then what is being displayed on the meter since the meter is taking the raw data and displaying an RMS value?

So when measuring inrush on a motor or transformer will the meter only display the RMS value of the inrush and not necessarily the asymettrical current resulting from the DC offset portion of the inrush?

When using the "peak" function of such a meter, it this capturing the peak RMS value or is this actually displaying the peak value of the current waveform symettrical or asymettrical?

I am looking to capture transient inrush currents on motors (inrush not locked rotor current) and am interested in capturing the total DC offset. I am looking at the attched meter for measurment. Will this meter be capable of measuring what I am looking for? Do I need a special type of current proble to be able to see the DC offset?

Does a molded case breaker look at RMS or peak value of current for tripping?
 

philly

Senior Member
The power meter that I am going to try to use is an Allen Bradley PowerPad Portable Power meter.
 

jim dungar

Moderator
Staff member
Location
Wisconsin
Occupation
PE (Retired) - Power Systems
Does a molded case breaker look at RMS or peak value of current for tripping?
Breakers built to UL489 are intended to protect conductors and therefore look at RMS current. There is very little general use equipment that uses, or cares about, anything other than RMS values.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
Generally, AC ammeters will tell you if they are "RMS" or not, most are. But then there is an issue of the quality of the RMS logic they use, so some provide what they call "True RMS" as a way to differentiate. Then there are those that, even if they say "True RMS" cannot accurately interpret RMS current in a harmonic rich circuit such as the output of a VFD or Soft Starter that is ramping. If you need that, look for a specific statement saying it can do that.
 

ELA

Senior Member
Occupation
Electrical Test Engineer
A look at your meters specifications.

http://literature.rockwellautomation.com/idc/groups/literature/documents/um/1412-um001_-en-p.pdf

Looks like a nice meter. Takes 256 samples per second and says it calculates a MIN,MAX every half cycle. You can get both peak and RMS values.
It appears this meter also has a graphics display and will capture in a transient mode. This might be what you are looking for to get the best picture of your inrush.
I am unsure if it is capable of calculating the DC offset you are interested in. I did not see where it listed such an ability.
You could infer a idea of the DC offset from the graphics display but I do not see an accurate calculated value being made available?
 

mivey

Senior Member
pg 27 shows the dc component of current (if you use the correct probe).

It also shows the dc component in the zero harmonic bin (pg 29 note 2).

The transient capture grabs one wave before the trigger, the trigger wave and two waves atfer the trigger.

positive and negative peak is also displayed (pg 26)
 

GeorgeB

ElectroHydraulics engineer (retired)
Location
Greenville SC
Occupation
Retired
When using the "peak" function of such a meter, it this capturing the peak RMS value or is this actually displaying the peak value of the current waveform symmetrical or asymmetrical?
My experience is that there is a minimum of 3 or 4 cycle average (usually 8 or 10) in the peak display of 60 Hz signals ... so the symmetrical or asymmetrical is moot and is averaged.

I am looking to capture transient inrush currents on motors (inrush not locked rotor current) and am interested in capturing the total DC offset.
Electronics (pretend engineer) guy here ... I've done the peak characteristics before by using a scope and displaying voltage drop on length of wire in the panel, approximately calibrating via the drop at running vs an Amprobe.

An induction motor doesn't carry DC from stator to rotor; I'm unclear of what DC offset would mean and how one could exist. If we only energized for 1 half-cycle and it was all of one polarity (I guess this could be done electronically, but not consistently via contacts), all of that 1/120 second would be "DC", but the significance would be little after several seconds of operation.

Clarify your needs further for us.
 

winnie

Senior Member
Location
Springfield, MA, USA
Occupation
Electric motor research
When an inductive load is energized with AC voltage, you can get a large and generally asymmetric current transient. This current transient has a a peak value that is well in excess of the normal peak operating current, and generally has a DC component, meaning that you have more current flow on one half of the initial cycles than the other.

The current transient dies down because of the non-linearity and resistance of the inductors; if you apply AC to a _perfect_ inductor, then you don't get a transient, but you will get a DC offset unless you switch at exactly the peak of the AC voltage.

Recently, someone here posted oscillographs of various inrush currents

-Jon
 

gadfly56

Senior Member
Location
New Jersey
Occupation
Professional Engineer, Fire & Life Safety
ELA, I'm not sure that 256 samples/sec is going to be enough to capture an asymmetric waveform. At 60Hz, each cycle is composed of only 4 data points. I would suspect that there's a whole lot of interpolatin' going on here to generate the information.

Okay, I take that back. I looked at the PDF for the meter and it says 256 samples per cycle. Plenty of resolution. I'll be quiet now.
 

ELA

Senior Member
Occupation
Electrical Test Engineer
gadfly56,

That was my miss-print, it is 256 samples per cycle as you noted.

At this point more input from the original poster would be nice.

Even though this meter can display DC current, with the appropriate probe, the manual is not stating (p27) whether that value is averaged/integrated and if so over how many cycles.

From what I have read thus far it seemed that the OP would like to have the DC component derived from just the inrush duration period?
 

philly

Senior Member
Basically I am trying to measure the maximum current seen by a breaker during motor starting. I have a theoretical number for what the LRC should be but I want to see what effect/magnitude the inrush current has on the breaker.

I want a way to measure the maximum current that the breaker sees during transient inrush. Since others mention that breakers only see RMS current then I suppose I'm interested in measuring the maximum RMS current.

Since the DC offset and the aysmmetrical component of the inrush current is shifted upwards how does this effect the maximum RMS current. Obviously the peak current is much greater but the breaker is measuring only RMS current not peak current. So how would this shifted waveform effect the RMS value? Would the RMS value be increased although the waveform in not symetrical.

So basically with this meter I will be able to use the transient function to see the waveform. I will be able to see the peak of the current waveform however since the breaker is not interested in peak but rather RMS current then I want a way to be able to see the maximum RMS current. I dont know if the "max" and "min" values shown on the meter will be referring to the peak values or the peak RMS values.

I guess maybe I was a little off track on the DC offset component. Although it exists, and it sounds like there is a way to measure it, I not really interested in the DC value but rather what effect it has on increaseing the RMS value of the current waveform. I'm assuming with this being said, then I do not need special probes.
 

winnie

Senior Member
Location
Springfield, MA, USA
Occupation
Electric motor research
I think that the statement that breakers 'look at RMS current' needs to be expanded.

RMS is a type of average; you take the waveform values over a period of time, you square those values, you average over that period of time, and then you take the square root of that average. The point is that an RMS measurement implies a time period over which the averaging is done.

RMS is the appropriate measurement to use when the current is changing more rapidly than the thing altered by the current. For example, the heating of a wire in normal service takes a long time, much longer than an AC cycle, so the RMS current is a useful measurement to determine conductor heating.

When you look at the trip characteristics of circuit breakers, the instantaneous trip happens on the order of a single cycle. So you need to look at the current flow over the course of a single cycle.

-Jon
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
100602-2240 EST

philly:

Suppose you had a purely thermal breaker.

There is a rate of heat loss from the trip element determined by its structure, and present temperature. The temperature of the the element will determine its deflection and thus when it reaches the trip point.

As current flows thru the thermal element it causes the temperature to rise, but this is counteracted by the heat loss from the thermal element. If more energy is going in than is being dissipated, then the thermal element temperature rises.

If you pump energy in very fast, then it can not dissipate fast enough to prevent tripping.

The trip time curves of thermal protective devices are usually plotted relative to current and time. But for short durations it is really energy that trips the breaker independent of the wave shape.

So excluding magnetic effects in a breaker and a discussion of short transients, then you want to calculate the energy introduced to the thermal element. Thus, an instantaneous plot of current, and an integration of the square of this curve is what you want.

Magnetic effects are a relatively instantaneous factor and for this you are concerned with the peak current.

.
 

philly

Senior Member
I was able to use the meter I had posted to record the inrush characteristics of a motor. I have posted one of the waveform captures although I did this several times and have several plots.

The issue I was having before was that sometimes this 150hp motor is tripping an instantaneous breaker set for an instantaneous pickup of 2500A. This happens intermitenly.

This plot is one in which the breaker did not trip. From the waveform capture it can be seen there there is indeed an asymmetrical offset involved and that the peak appears to be over 2500A. Also there were other plots that showed the current waveform peaking at about 2800A without the breaker tripping.

So this leads me back to the question of what the instantanous element is looking at? Is it looking at the peak of this current waveform? It would appear it wasn't since the breaker was not tripping at the values listed above unless there is some kind of tolerence? Or is the breaker looking for the RMS value for this current aysmetrical current to be greater than 2500A to trip? I believe the RMS value of this asymmetrical current would be signifigantly less than 2500A (I'll look, I believe meter may show value) and thus would not really indicate that current was close to trip value and thus cause nusience trips. So I guess the question is, weather or not this breaker looks at the peak of the waveform for a tripping value?

On other thing I noticed on the individual phase plots was that the magnitude of the maximum current was determined by where the voltage closed on the waveform. It appears that the phase current magnitude is greatest on the phase that has the voltage waveform closing closest to zero crossing. Is this correct?
 

ELA

Senior Member
Occupation
Electrical Test Engineer
Do you have a part number on that breaker?
Is there any settable delay time on the trip?

Looking at your waveforms I did not see any one phase inrush current lasting for one or more complete cycles at a level over 2500A? Peak only lasted 1/2 cycle > 2500A .

To your question about tripping on peak or RMS I would say that it takes a certain amount of energy to trip. Because the waveform amplitude is changing each half cycle (during inrush) it becomes more complicated to estimate the energy level.

As a first approximation I would guess that if the peak > 2500A does not last at least 1 full cycle then I would not be surprised that the breaker does not trip (without reviewing your breakers manual)?
 

philly

Senior Member
Do you have a part number on that breaker?
Is there any settable delay time on the trip?

The breaker is a Cutler-Hammer HMCP250W5C molded case breaker. It has an instantaneous adjusable setting which is currently set at 2500A.

Looking at your waveforms I did not see any one phase inrush current lasting for one or more complete cycles at a level over 2500A? Peak only lasted 1/2 cycle > 2500A .

To your question about tripping on peak or RMS I would say that it takes a certain amount of energy to trip. Because the waveform amplitude is changing each half cycle (during inrush) it becomes more complicated to estimate the energy level.

As a first approximation I would guess that if the peak > 2500A does not last at least 1 full cycle then I would not be surprised that the breaker does not trip (without reviewing your breakers manual)?

So basically you are saying that since the breaker instantaneous trip is a function of energy you would venture to guess that it could be based off of the peak current, but would have to last for longer than a cycle? How many cycles?

Attached is the cut sheet on breaker showing TCC's as well.
 

ELA

Senior Member
Occupation
Electrical Test Engineer
Not sure but from the time curve it looks to me that using the "I" setting that you can trip anywhere between 1 msec and 20 -30 msecs?

So I would expect the breaker could trip at less than one full cycle but it may also take up to 1-2 full cycles to trip.

Also keep in mind tolerances of your measuring equipment and the breaker setpoint tolerance with respect to absolute current levels.
 
Status
Not open for further replies.
Top