Electromagnetic Field Collapse Principals

The physics are about the same in both cases; i.e., the magnetic field stored in the coil and iron collapses causing an induced voltage of opposite polarity across the coil terminals. The major difference is the time it takes; simply shorting the terminals causes the field to decay at the same rate as charging the coil while reversing the polarity forces the field to collapse at approximately twice the rate. The diode accomplishes the same thing as the shorting method while saving wear and tear on the switch contacts but that would need to be one big diode!

The diode is not a problem. At a roughly 3.5 ohm internal resistance the theoretical short circuit current should not be above the normal running amps which would dictate that at 240 volts 60 amps a standard 400 volt 100 amp diode from a common welder power supply would be more than adequate.

My curiosity is that how the shorting effect causes the magnetic field to apparently reverse. I know how the electrical polarity reverses due to the collapsing field sort of having a momentum effect with the electrical input power due to inductance.

The magnetic field will NOT reverse at 'shorting'. Only the voltage polarity would, if you tried to interrupt the shorting current. The question then is what kind of short cirquit would that be? The only (insignificant) current reversal will happen at a freewheeling diode's presence, where you also have it's parasitic capacitance that will oscillate shortly when Icoil*Rcoil becomes less than diode's Vfwd. S.M.

I have a correction to make. The time constant equation for an inductor is L/R=τ. (I wished CR4 had a better font for Greek letters.) So the diode will cause a quicker collapse and a diode in series with a resister will be even quicker. The trade off will be that the back EMF voltage generated by the magnet field collapse will be larger but it now can be controlled by the value of R so that the opening contacts will not arc, as somebody else here pointed out.

your shaking cob webs loose here. And its late

I don't believe the short circuit method is the right term, it only appears that way.

It more like the principals of self induction.

And self induction occurs when the current stops flowing, which causes the magnetic field to collapses. What happens is that the lines of force are moving toward the wires from which they came which is the reverse from when the magnet buildup.

At the same time this also would means that the counter-voltage induced also will be reversed. And since its reversed, it is in the same direction as the original voltage.

The counter-voltage opposes the decrease in current during callapse and will oppose any change in voltage in the callapse. While in buildup the opposite happens where it opposes the increase in current flow.

Problem when this happens, it can cause severe arcing unless capacitors are used. Sometimes known as short circuit.

Its late, I'll read this in the morning and wonder wtf am I talking about.

I googled and here's almost word for word.

http://www.ndt-ed.org/EducationResources/CommunityCollege/EddyCurrents/Physics/selfinductance.htm

A phenomenon when a freewheeling diode is used is, a possible double-shot drop off. Due the i* dL/dt....something like this..

Generally speaking as higher the allowed back EMF, the faster the field collapses. Restricting it to one diode drop makes the system lazy but also safer. S.M.

Thanks guys, so far this has been quite helpful.

Dealing with small DC type solenoids is one thing but for the most part I don't get many opportunities to design control systems around multi kilowatt electromagnets where deliberately controlling the relatively massive collapsing magnetic field is necessary!

When working with electro magnetics you need to understand HYSTERSIS. Hystersis is the lagging of an effect behind its cause. The amount of Hystersis in electro magnetics is determined by what the core material is. When the current is shut off the magnetic field remains and then drops down. How fast it drops and far down it goes depends on the core material. I imagine if your crane is used to pick up only large heavy objects then your hystersis would not be enough to hold the steel in suspension when the power is off. If your crane is picking up all size of material and some of it is low weight items then those would hang on a while and be raining down as time went by. The reverse curent would throw all items off when you wanted them off but cost you more power consumption. So your coice of circuits would be based off the above. Are all your items heavy? Will smaller pieces dropping off after power is cut off be a nusiance? This depends on what the hystersis curve is for the material your core is made of.

The thing is that most of the crane electromagnet controllers still use the old fashioned method of using large mechanical contactors with very wide contact throws and every one has the issue of contactor point burn out due to the heavy arc flashing that goes with trying to switch a large DC power level connected to a massive inductor.

I have a working design that uses 1200 volt 600 amp IGBTs set up in an H bridge configuration to produce the short reverse of power for the fast drop but there is a catch. With a IGBT based solid state H bridge the bypass diodes in the IGBTs will let the surge of current travel backwards to the DC generator which is normally not a big deal being it can easily sink it by working as a motor for a few seconds.

The problem I am dealing with now is a AC based generator system where a 20 KW 100 Hz three phase alternator output gets rectified to a 240 volt DC output. The problem with that is that now the massive inductive kickback of power cant be dumped directly to the generator as safe load sink. It either has to be dumped across a huge surge suppressor MOV or rerouted to someplace so that the inductive surge doesn't reach arc over voltage levels.

The second electromagnet control system simply opens the main contactor and a large MOV absorbs the momentary surge until a second smaller contactor short circuits the electromagnet itself. Still this design eventually needs the two contactors replaced and whenever the big MOV eventually fails everything gets fried literally due to the uncontrolled inductive surge and resulting high powered flash overs that result.

What I am wondering about is how the short circuit method apparently works to get the magnetic field to drop fast enough to release the load and if its that simple why there would not be any reasons that I cant simplify the H bridge design down to a simple full wave rectifier system and single AC side three phase solid state switch design and do a way with all of the apparently unnecessary and expensive surge suppression and H bridge all together and treat the electromagnets as nothing more than really big DC solenoids being driven off of a rectified AC source.

When the magnet is energized, and then you de-energize, the removal of the current is practically being resisted by the coil... Therefore, the diode in parallel allows the coil to shorted and the flow of current keeps going in the same direction through the coil until it is spent. This does not reverse the magnetism of the coil core. Therefore, some remanent magnetism will remain and will still be holding some scrap ...

To have a complete and fast collapse of the magnetic field, a reverse pulse is required to cancel the remanent magnetism.

protecting the contacts of the contactors is a different matter than the quick collopase of the field to release the load. the Diode is to protect the contacts while the reverse pulse is to quickly collapse the field.

You might not be bothered with the quick collapse or full collapse of the field (this will depend on the type of loading Material and their weight and how critical all this is).

Shorting the coil is similar to the effect of the diode but less effective since there will be a timing issue between closing one contact and opening the other..!?

Hi again. There are quick and dirty ways to get back most of the coil energy, and do it fast. Lookup 'dual swich conversion topology'. It consists of two switching devices (for this case IGBT is fine) firing synchronized, and 2 HV fast recovery diodes. With this you can get most of the power at switch-off back to DC rail, where a higher than rail voltage rated capacitor will store the power for use at the next cycle. (you can calculate it's capacitance so the voltage raise will not be too high).And don't worry it will not overload your electomagnet. It will (sadly) go to normal voltage before electromagnet goes to rated current. No need to drive power back to AC rail. Too complicated and costly. S.M.

I am not concerned about energy recovery. Its surge and voltage spike suppression that I am concerned about.

As far as the capacitor bank goes I am not sure how big it would have to be to hold the equivalent of 50 amps for roughly 5 seconds starting at a line voltage of around 240 volts and going up from there. My math suggests around 250 Farads. I have serious doubts on finding a 350 volt or higher rated 250 farad capacitor bank for less that what the whole electromagnet system costs.

the whole point of this is to find the simplest cheapest most reliable method. Obviously the short circuit method works being many electromagnet crane controllers are designed that way to begin with.

Did some quick math for 450v readily made electrolytic capacitors and came up with [C]= [L]/13. C in Farads and L in Henries (with safety margin the conversion percent loss that would be bigger than needed). So is your coil estimation 3250(!) Henry? . S.M.

Well, as I said in my comment before, your interest is really the protection and /or extension of the life of the contacts of the contactors.

You are also concerned about the contacts on the shorting contactor, as you mentioned before.

Suppressing the arching to protect these contacts is a compromise between the cost of the suppressors and there efficacy AND the changing of the contacts, life cycle etc...

Switching on the AC side of the rectifier is a good idea but you still need to protect the rectifier diodes from voltage surge by having a fly weel diode and some anti-surge device like MOVs. You can also have an electronic circuit that cuts off the AC at the correct part of the sinusoid cycle when the surge will be minimal or zero... Need someone in that field to elaborate.

Hi Tcmtech, in your OP you said that the inductance is quite large and that the supply voltage is 240 VDC. I was under the impression that if you operate a circuit on DC the inductance would not come into effect, but if the DC is of a poor quality then it may have some effect.

If the DC power is just rectified AC, wouldn't the easiest way to control the switch Off and On be to switch the AC Supply, this way the arching of the contactor contacts would only have the effect of switching a resistive ratting, then the only other problem you would have is how to deal with the back emf which you could implement the fly-wheel diode across the magnet, this would have the same effect as the shorting contactor only it would be quicker.

Cheers

Joe

Correct the inductance is large but during normal operation its not an issue. The main power limiter is the electromagnets own winding resistance of around 3.5 - 4 ohms.

The curiosity I had was that why some factory built controllers like the Ohio Magnet Co units reverse the polarity for a short burst of a few seconds but others just open the main power contactor and let it flash over, only limited by what the MOV dissipates, until a second smaller contactor short circuits the electromagent for the remaing 2 - 3 seconds that it takes to disipate the magnetic field energy and cancle the remaining stored energy that the inductance was holding.

Both designs have the same basic load drop times but two different methods of operation.

The magnetic field action was what I was mostly curious about between the reversing polarity system and the short circuit system designs. In real life apications both work fine.

What's missleading you and I think I can clarify, is that you think that reversing the polarity the DC rail feeds the electromagnet reversely. NO. By reversing the polarity connection the generated back EMF that is already reverse than the charging voltage and being as higher as neded is fed BACK to DC rail, and this is only applicable if the DC rail CAN sink it i.e not in your case that you have AC rectified to DC and you can't sink any back charge. S.M.

Yes I know that part. Thats why the old style of reversing works with the brush type DC generator systems and not the AC alternator based ones which is what I am now working with.