On the subject of inrush current

Steve Bailey

Lifetime Supporting Member + Moderator
Join Date
Apr 2002
Location
The boondocks of Western Massachusetts USA
Posts
8,533
The discussion in this thread about inrush current in AC vs DC circuits got me thinking.

I wasn't satified with Vic's assertion that there is no inrush in a DC inductive circuit. In the section of my college physics book where they discuss induction and derive the same equations that Vic cited, there is an assumption that there is no iron in the vicinity of the inductor. Furthermore, the same equations apply to AC and DC circuits, the difference being that in the AC circuit the current is not a constant, but a sinusoidal function.

I'm hoping someone can post citations to references that explain the source of inrush currents.

Here's my take on it, unsubstantiated by any math.

First, there is a mechanical component to inrush currents. In the inductive devices we all deal with (typically starters and solenoid valves), you're moving an armature within the coil. That armature has to be accelerated. The energy required to accelerate the armature comes from the electrical current flowing in the circuit. Thus, the current in the circuit while the armature is accelerating will be higher than the current when the armature is at a constant (including zero) velocity.

Second, we know that in an AC inductive circuit, the current waveform lags the voltage waveform by a phase angle proportional to the inductance. Prior to closing the switch that completes an inductive circuit, the voltage and current can be anywhere from zero to 180 degrees out of phase. Could the inrush current have something to do with getting the two waveforms into a steady-state phase relationship?

We also know that we can reduce inrush by putting an RC network in parallel with the coil. In a capacitive circuit, the voltage lags the current, just the opposite to an inductive circuit. It would seem that the RC network in parallel to the coil, brings the circuit closer to a purely resistive circuit, in which the current and voltage are in phase with each other. This seems to confirm my thinking that inrush is related to the establishment of the phase relationship between current and voltage.

Idle speculation for a Sunday afternoon. Does anyone care to comment, poke holes in my reasoning, or point me to the applicable math?
 
can't cite references...

But, as I remember from my AC and DC courses in college, you
have to consider what the condition of the circuit is at the
time you are interested in it.

In the case of DC (the simple case), when a coil is suddenly
turned on, there is a STEP change in voltage. The traditional
DC circuit laws do not apply. And, don't forget, you have to
look at a 'coil' in its 'DC equivalent' state...that is..
a resistor in series with a coil, and a parallel resistor
around all that.

I'd think that anyone with a really good PSPICE model of a
coil would be able to tell you the values.

An interesting subject..not well understood.

Speaking of inrush, the next good thread would surge protection
and suppression.
 
I have never read anything on DC circuits being affected by inrush currents, but I have read hundreds of articles on inrush currents concerning transformers and AC motors. Inrush currents can destroy transformers, and motors can cause havoc with entire circuits due to inrush currents, causing sags in the voltage dropping out control circuits and even causing starting problems with the motors themselves.
If anyone can locate a source discussing inrush currents in DC circuits, I would like to see it.

Roger
 
It is a matter of degree.

Any wire is an inductor at high enough frequencies, yet they allow the current to surge or in-rush. First one must define how much current is an in-rush or current surge. An inductor can be chosen to limit the in-rush to the desired value. If the inductor is not big enough then the in-rush will not be limited much.

I can do the simulation if someone really cares. I would look something like the other links I have posted using Mathcad.
 
Here is a link to the Omron LY2 relay brochure. There are examples of the effects of with and without suppression diodes or RC circuits. This may be very interesting to many people that contribute to this forum.
I am trying to find some similar graphical data on inrush Bear with me (not down).
Omron LY2 Relay
beerchug
Found what I was looking for. Here is a link to an AB IEC contactor manual showing inrush for AC coils but only pure DC current draw for DC coils.
AB Contactor Manual
🍺
 
Last edited:
Theoretical discussion of inductors is centered about one primary premise - the "IDEAL" Inductor.

WE DON' GOT NO STEEKIN' IDEAL INDUCTORS!

This discussion is not just theoretical. The effect of inrush current can be easily detected and observed.

Build a circuit using a DC power source, a switch and an inductor. Use the classic formula (the one Vic showed) and determine the maximum current that the inductor should theoretically allow. The "R"-value is the DC resistance of the inductor.

Add a fast-fuse (quick-blow) to the circuit. The value of the fuse should be exactly equal to 1% greater than the calculated current. Insert the fuse between the high side of the source and the inductor.

Close the switch. Now, get a new fuse to replace the fuse that has just blown. Repeat as often as you like...

I'm curious to see what would happen if a fuse was inserted before and after the inductor. Would it always be the case that one particular fuse (the lead fuse) would blow before the other (the tail fuse)?
Of course, this requires that the fuses are EXACTLY the same! Ideal Inductor? Why not an Ideal Fuse?

It is precisely this effect that requires circuit fuses to be over-sized.

Now, add a resistor to this "ideal circuit". Insert it between the source and the switch. Calculate the maximum current that this resistor will allow - ignore the inductor. Install a fuse equal to 1% greater than the calculated I(R).

Close the switch. The fuse should still be intact because the current will not have exceeded the limit imposed by the resistor.

Another thing I'm curious about is... What would happen if the fuse and resistor traded places. That is, the fuse between the source and the switch, and the resistor between the switch and the inductor.

I wonder about this because, it does, in fact, take time for the electric-effect to be "felt" in a circuit.

Let's say that the circuit consists of a source, a switch, and two pieces of wire. One length of wire is connected between the source and the switch. The other length of wire is connected to the output of the switch - and that's it! The other end of the wire is dangling about in the air.

This, of course, in the normal sense, is not really a circuit because it can't be completed. But it does lead to an interesting effect. No matter how small the cross-sectional area of the wire, the end of the wire is electrically similar to a Capacitor.

The cross-sectional area will (must) certainly be large enough to accommodate at least one electron. If it can't accommodate even one electron then the cross-sectional area does not exist! (If a given mailbox is too small to accommodate even a common postcard then, can that so-called "mailbox" be acknowledged as a mailbox? No, it can't.. If it can't hold mail... how can it be a mailbox? (No, Eric... E-Mail does not count!)

So, back to the wire...
In a "rest-state, the electrons are situated as they normally are - there is no pressure (voltage) to compress them or encourage them to be in any state other than their normal rest-state.

Now, as soon as the switch is closed a "potential-difference" exists between the source and the end of the wire. Now the electrons are "encouraged" to move as quickly as possible to the "exit door". Of course, the "exit-door" is "locked shut"! Picture someone yelling FIRE in an illegal night club - or at a Rave where the doors are secured.

The initial "resistance" to flow is the physical sizes of the internal halls and doors. It is only when enough people are jammed up against the locked exit doors that the lack of a "path" to the outside is felt - all the way back to the last person trying to flee!

The crowd of people with their faces jammed against the exit doors represents the "charge" at the end of the wire - the plate of the capacitor - if you will.

The point being, before being forced to stop, the crowd ran as fast as the internal resistances allowed.

Hmmmmmmmmmm....
 
seppo-

Air is a dielectric. At such-n-such pressure, temperature and humidity, air is considered to be the "standard" dielectric. Air does not tend to support magnetic fields and electrical current flow. Certainly it can, under certain conditions, but it doesn't do so willingly. Thank God for that!

Air provides a certain amount of "reluctance" to the development of an electro-magnetic field. This reluctance is somewhat similar to electrical resistance. However, it's not exactly the same.

It takes a certain amount of energy to develop a magnetic field in any medium... air, water, gas. Energy levels must reach an "avalanche point" before a field is allowed to occur. The "avalanche point" is determined by the "reluctance" of the medium.

Once the "avalanche point" is reached, the dielectric strength of air only allows a magnetic field to exist in a particular density (according to the pressure, temperature & humidity).

For any given air-coil, you can develop a magnetic field of such-n-such strength, linearly with power, until you reach the saturation point. From that point on, all you are adding is heat (I2 x R losses).

You can not increase the strength of a magnetic field on an air-coil by cranking up the power beyond its saturation point. If you attempt to do so, you will not increase the density of the field. However, if you do attempt to do so, you will eventually exceed the power capability of the coil and burn it up!

Now, throw a hunk of iron into the picture. It just so happens to be, in terms of building a magnetic field, iron has a very, very low "reluctance". A magnetic field can develop very easily in iron. Iron is NOT a dielectric - it is a conductor!

In an inductor coil with an iron-core, the over-all reluctance to the development of a magnetic field is greatly reduced. Additionally, the field density is greatly increased. The field-density is the main factor in the strength of a magnet. (Earth has a very large magnetic field - the density of it is so small as to be "almost" un-noticeable.)

The strength of a field developed with an iron-core is many times greater than the strength of the field developed in an air-coil.

So, in terms of your drawing...

The "Without Iron" line in your drawing represents an "air-core" where the counter-emf field is relatively weak. The "With Iron" line represents an "iron-core" where the counter-emf field is relatively stronger.

The "1/2-Way Iron" is in between the two extremes.

The "Iron-Core" develops a stronger field (more field-density), more quickly. Thus current from the supply is resisted at a greater rate (more strongly and sooner). The supply emf, and the, current from the supply overtakes the coil counter-emf and reaches a steady-state later than it does in the case of the air-core.

The "Without Iron Core" develops a weaker field (less field density). The field is developed at the same rate, but the density is considerably smaller - the field is weaker. Thus current from the supply is resisted at a lesser rate (less-strongly and later). The supply emf, and the, current from the supply overtakes the coil counter-emf and reaches a steady-state sooner than it does in the case of the air-core.

The key to magnetics is, "field-density". The greater the field-density, the greater the strength of the field. In the case of inductors, the greater the strength of the field, the greater the Counter-EMF.

BUT, IN TERMS OF COUNTER-EMF, ONLY IF THE FIELD IS CHANGINGING!!!!!

Oh my, there is sooo much more! Some of the info above has been generalized.
 
Thanks Terry !

As you see, the time-constant with Iron is greater than without it.
That means in AC-circuits, that the Current must be behind of the Voltage and the current amplitude will be less than without iron. This explain that phase shift between current and voltage without mathematics theory. (I know that theory as you do, but this is for people who are not familiar with el-tech.)

Terry, this is part of the 'much mooore'....
 
I just want you guys to know how much people like me appreciate this type of information. Seeing things explained in different ways, even if you have a basic understanding, it's a real help. I love this site!


:site:
 
The IDEAL fuse

Pete Stein at Texas Christian offers a course in measurement systems engineering (great course) and one of his PET items is the COMETMAN theory. No two pieces or parts can have the SAME EXACT features, close yes but not always. The wires in the circuit are from different spools, which may have been mined from different areas, with different impurities, different insulation, all kinds of stuff that can affect a circuits performance (coiling excess wire leading to inductance, humidity, distance, crosstalk due to shielding).
BTW I seem to remember that Los Alamos conducted a study on the speed of electricity and came up with a number around 9 nano secs per 100 ft.
 
All electromagnetic apparatus with iron cores have varying degrees of magnetic remnance, which retains a substantial portion of the magnetic induction to which the core is driven until polarity is reversed at the power source. This characteristic of iron means that when magnetizing power is removed, as with opening a switch, that the core remains magnetized. When power is reapplied to the core, the magnetic state to which the core will next be driven depends on the phase angle of the source relative to the retained magnetic state of the core. If the phase angle is identical to the state when the circuit was switched off, then magnetizing current will remain well behaved, and of very small amplitude. However, if the phase angle is sufficiently out phase and at a voltage zero, a very large inrush current will result.
 
harryg,

At last, an explanation of inrush! Thank you.

Let me try to rephrase it, and tell me if I get it right.

When you shut off power to a solenoid, the armature retains some residual magnetism. If you were to take the armature out of the coil, you'd find that it would attract iron filings. The orientation of the magnetism (N - S poles) depends on the when in the AC cycle the coil was deenergized.

Sidebar: Does the strength of the residual magnetism also depend on when in the AC cycle the coil was deenergized?

When the AC power is reapplied to the coil, if the poles of the residual magnetism are opposite to the poles in the coil, there will be higher current (inrush) until the poles are in alignment.

Some questions suggest themselves:

If you use a zero-crossing switch to turn the coil on and off, do you minimize inrush?

How does an RC circuit in parallel with the coil help to reduce inrush?

This certainly explains why DC solenoids don't have as high inrush currents as AC coils.
 
Zero-crossover SSRs are excellent switches for resistive capacitive, and slightly inductive loads,however,may be the worst possible method of switching on a highly inductive load. Evidence has come to light that zero-crossover turn-on of such loads can cause a surge current of perhaps 10 to 40 times the steady state current, whereas turn-on at peak voltage results in little or no surge. Surge currents of such magnitude can seriously shorten the life of the zero-crossover SSR, unless the SSR has a current rating well in excess of the load. They create EMI and RFI (all along the load line) which can destroy logic gates and cause unwanted turn-on of semiconductor switches. Additionally, these surge currents create thermal and mechanical stress on the windings of the inductance and can lead to early failure of the device. There are several component options for inrush current limiting. The two most common alternatives are the use of NTC (Negative Temperature Coefficient) thermistors (Surge Limiters) or various forms of active circuits.
 
In this thread and a previous thread, it was stated in several posts that the theoretical DC current through an inductor is one thing but in the real world where there is iron cores involved, things are different. A few days ago, I came across a fairly large DC solenoid and decided to get some empirical data and see if it followed the theoretical current curve. It was no surprise to me that the current behaved as predicted by the current equation.

The solenoid was approx. 4” long and 3 ½ “ in diameter. It had a 1 ½ “ diameter iron core with a 1” stroke. The core was supported at each end by a shaft with bushings in each end of the solenoid housing. The bushing at one end was missing which meant the core rubbed on the iron sleeve of the coil. This caused the core to stick a little as it shifted in the coil.

The solenoid was rated 24 VDC and had 18 ohms resistance. This gives a steady state current of 24/18 = 1.33 amps. I connected a 0.2 ohm resistor in series with the coil and using a digital storage scope captured the voltage across the resistor when 24 volts was applied to the solenoid.

I captured several waveforms, 3 of which are shown below. The voltage across the resistor is directly proportional to the current in the solenoid. The voltage is shown on channel 1, one major division = .25 amps. The voltage across the solenoid is shown on channel 2.

The top waveform shows the current with the core removed from the solenoid. This gave the minimum inductance and the shortest time to reach steady state. The bottom waveform is with the core fully into the coil (normal energized position). This gave the maximum inductance and a longer time to reach steady state. The middle waveform shows the core under normal operating conditions (starting part way into the coil and moving fully into the coil). This gives a lower inductance initially and an increasing inductance as the core moves further into the coil. I would have expected this curve to start similar to the top curve and transition to the bottom curve. This is approximately what happened but there were two significant notches which were not expected. My only guess is that as the core jumped ahead, (it did not shift smoothly in the coil), the sudden increase in inductance produced a rapid increase in the counter emf which momentarily caused a drop in current. The current equation supports this if the inductance is rapidly increased.

[attachment]

In all of the waveforms, there was no inrush current. The current gradually increased until it reached a steady state.


Steve

In the first post of this thread you stated
Furthermore, the same equations apply to AC and DC circuits,
This is not correct. The equation for the instantaneous current in a AC circuit is much different than for a DC circuit. Since the voltage is constantly changing the equation must also take into account the phase angle of the voltage. I would give the equation but I do not know how to show Greek characters and Calculus symbols in the message. For a detailed discussion of AC inrush current in transformers, (similar to solenoids except the inductance does not change as a result of the iron moving), look at this site.

coil traces.gif
 

Similar Topics

For a decade I've been around this forum, the new, first-time threads (in a log-in session) were display in bold typeface. The subject line...
Replies
22
Views
7,812
I know this is very far off subject, but was hoping someone here could lend a hand. I have been asked to start calibrating belt scales for...
Replies
19
Views
3,704
Working with dmroeder to get a AB PIC adapter to work under Windows using SP3 update. Installed his SP3 fix file and still can not get RSlinx to...
Replies
25
Views
6,196
Hello all FTV 8.0, RSL5K..... Is it possible to connect the visibility of a text box or a multi state indicator to a numeric input on the touch...
Replies
4
Views
2,364
We have two PLCs both are Mitsubishi-FX2N, and we want them to communicate to Weintek HMI MT8100ie. MT8100ie can communicate with Multiple PLC and...
Replies
1
Views
4,945
Back
Top Bottom