Electric Arc Furnace Tripping

Your problem seems to be that you are drawing "short-circuit" currents for too long when the electrodes dip into the metal. It is entirely correct for the overcurrent protection to work if this goes on for seconds. I would imagine that the control system should limit the initial current to values well below "short-circuit" and any tripping "time x current squared" limits - maybe it is not doing this! You have not made it clear if you have a problem with that - or with an electrode lifting mechanism which now operates more slowly than it did, due to an "amplidyne" fault. If you wreck your transformer, OCB or amplidyne you will go from "production at any cost..." to no production at all for a long time. Please advise.

When was the last time your furnace controls were tuned? 67model is pretty much correct with answer he has given. I would be looking at the reversing control circuit for mast to raise operations. It appears that response time is improper. I am not sure if your using DC motors or hydraulic actuators for R/L operations, but are you mast rollers and bearings in good shape. A lot of times there is mechanical issues rather than electrical as well.

Well i believe that the amplidyne control is tunied in routine by authorised department.I think the arcing should be done with a 6inch gap with melted scrap.So there won't be any problem with the controls, but when we checked the metal-diode bridge(which gives supply to the field of amplidyne generator) with a multimeter it showed some abnormal values but the shift-engineer seems to be satisfied with the results.I guess the control system is normal because we have a thyristor controlled furnace also showing same behaviour.The engineer brought up the question of lifting the electrode.So i want to know within the short time is it possible to lift it or we should leave it to pre-set tuning.

I wrote in post #4 some ideas for doing an investigation. Your post #5 confuses me, you wrote "we have a thyristor controlled furnace also showing same behaviour." Does this mean thyristor furnace also trips frequently??!! You write "question of lifting the electrode" but it is my understanding that, after lowering the electrode to start the arc, it is essential to raise the electrode to reduce the arc current to running values. It seems, in essence, that the electrode is not rising enough before the limit set by overcurrent x time is exceeded. You have not made it clear if there are three controllers [amplidyne], one for each electrode, or a single controller for all. You mentioned the Y phase, but this may be just that the Y phase overcurrent relay [of X,Y,Z ??] is the most sensitive and initiated the trip. You mention that operation is normal with 6 inch gap - do your words "lifting the electrode" mean increasing the gap to say 7 inch?? Not knowing the system I can guess that the amplidyne generator (plus a motor it drives?) is a control system that regulates 1) the electrode gap or 2) the electrode current or 3) furnace temperature or all of them. There are probably ancillary limits like speed of electrode movement and certainly limit micro-switches at top and bottom of electrode travel. If you are hitting bottom of travel, it might be that the limit switch is, in some way, failing to arm circuits to raise the electrodes. You mentioned unease about the "metal-diode bridge" which provides field supply for amplidyne - if I am guessing correctly, these are selenium or copper-oxide plate stacks. If so, they do wear-out with increasing ON-resistance. While the absolute value of voltage may not have to be exact, if positive and negative supplies are used, an unbalance (to common) between them (or high AC ripple on one, due to high resistance of one leg) could change set-points for electrode lift etc. It would be worth checking these supplies for DC voltage and AC ripple ( usually DVM type meters have a DC blocking capacitor on AC measurement) - so if you switch meter from DC to AC, the AC reading gives the ripple -make sure AC reading is same when you swap meter leads, so negative lead is on positive. Are these DC supplies 3 phase or single phase rectified - it makes a lot of difference to the possible ripple?? Are there any DC smoothing capacitors which could have deteriorated?? Another item to check is any position measuring device which gives electrode height to the control loop, if it is no longer reading true then settings will be altered.

In reply to #9

I have answered according to your questions.
The 3 electrodes are under controlled operation when we put 3 Nob in automatic mode.There are three controllers [amplidyne], one for each electrode (amplidyne motor is one)
The 6 inch I mentioned is just for imagination its purely controlling the electrode current(V/I=R,resistance control).
Speed is constant with winch motor with limit switches("2") as you mentioned.when the electrode hit down the furnace we say it as slack.

Supply is 3ø supply and single phase rectified for each electrode,no DC smoothing capacitor.

Control panel

In reply to your post #11, giving info about control panel. Now I understand it is a current control loop, I can have a guess at "what the system is like" and "what is wrong"! I have drawn a sketch of the system, details may not be right - but good enough, I hope. I have left out items like resistors and switching which give manual control by connection direct to F1, F2 also limit switches. Description cannot be short, alas, to get an understanding. Sorry about underlined text at end but I cannot turn it off!

In image you have :-

Power circuits

1. Amplidyne, which is a direct current generator, driven continuously at constant speed by motor M2 (probably same shaft as generator, one housing). Its field F1, F2 will cause output (to reversible hoist motor M2) to be one polarity if F2 is positive to F1 and the opposite if F1 is positive to F2. Output voltage to M2 will be proportional to amplidyne field voltage.
2. Hoist motor M2 (probably with continuously energised shunt field), moving electrode E up/down by mechanism I have called RACK, say a bevel gear drive, probably with counterbalance C so motor does not have permanent current while stationary.

3. Arc from electrode E to steel S, the arc current being monitored by current transfo CT.

Control Circuit

1. On right of control circuit - Transfo T, With diode D2 giving a DC voltage, from which an adjustable proportion is got by resistor R3 and hand variable resistor VR1.
2. On left, CT secondary, with ammeter and over-current relay, and load resistor R1 (to give sufficient voltage to supply field).
3. Between CT and F1, D1 gives a DC voltage to balance that from VR1 when current is at controlled value. R2 gives an essential path for field current. Consider when there is no current in CT, a positive voltage is at F2 BUT D1 is reverse biased - without R2 there can be no current through field, M2 voltage zero, no motor M2 movement.

Operation

1. The manual raise is used to fully raise the electrode, before electrode power is turned-on and automatic control is selected.
2. When automatic control is selected, there is no current in CT and no voltage from it to F1. Current flows from the "set-point" DC source at F2 through the field, out through F1 and through R2 back to T - the amplidyne gives a voltage to the hoist motor M2 which lowers the electrode to start the arc.
3. A high current flows through CT, giving a large DC voltage to F1 via D1. This voltage well exceeds the voltage at F2, the field is powered at maximum voltage, the amplidyne gives maximum voltage - now in a direction to raise the electrode. M2, to provide the back e.m.f. to match maximum amplidyne voltage, turns at maximum speed.
4. As current falls towards the set value, the "error voltage" applied to the field falls, amplidyne output voltage falls, M2 speed falls. Setpoint is approached slowly when error is small - avoiding "over-shoot" and "hunting".
5. When the set current is reached, field (error) voltage is zero, motor M2 stops.

6. If current changes from set value (due to melt status etc.) the error voltage at field drives M2 to correct the error.

Possible Problems

Since the electrode seems to stay low until overcurrent trip stops control, but ammeter (and overcurrent relay?) work it would appear the rectified CT voltage is not appearing at R2/F1 or the wiper W on VR1 is not making contact, so that the "target current" is too high.

Investigation
Is voltage at F1 rising in line with current? Comparing Voltage on another electrode controller (phase) which works OK is helpful here. Is D1 good? Is F2 voltage abnormal compared to "good" control on another electrode??

Good Luck

This is a simplified explanation and thanks for putting that diagram.The diagram is the exact model of EAF here.Instead of counter balance we have break arrangement.

When the power circuit is ON we get "Voltage signal" and "Current signal" in control circuit,that's how it creates both polarities in generator field of amplidyne.We have a procedure for checking the control circuit.We create an artificial Voltage and current signals with the help of an auto-transformer,rheostat.(I don't know the current and voltage values but soon i will get it )and see if the control panel ammeters(3) and voltmeters(3) are showing correct values or not.

The underlined statements are to be verified and i will inform you soon as possible.

If you have a mechanical brake, is it electrically controlled?? Could it malfunction in automatic to slow lift??

For each winch motor we have electromechanical brake.

Sorry, had not noticed your new post because date was same as my last. Then I guess that, in automatic, brake is applied when amplidyne input error is low. Either low volt relay on amplidyne output or low-current relay in its field would do, de-energising brake (all power off = brake on). Probably, contactor would disconnect winch motor from amplidyne output when brake was on. Any trouble with brake is most likely mechanical - dirt, debris, wear.

Good luck

its fine your reply and suggestions was very helpful

we will do a thorough checkup soon or later

Here is the schematic for more understanding

i don't know how much visible that is........

both are same....!!!

Drawing only 300 x 400 pixel on CR4. Good enough to identify meters, motors, lamps. Unable to read text or clearly see smaller, more complex bits. Digital camera usually does about 2000 x 1000 pixel. I have found, with line diagrams, that conversion to .png type (from scanner black & white .TIF output) in black and white usually shrinks file size the most for transmission. Diagram I posted was ~ 2400* 2000 pixel, but only 26 KB in .png

http://cr4.globalspec.com/thread/64894/EAF

Silly question, have you replaced the tripping breakers? If they are 27 years old they may just be past there expected lifetime in handling the enormous currents in starting the arc.

On general principles I don't like the idea of changing operating procedures that use to work because a protection circuit is now tripping. This implies to me that somewhere a subtle failure is happening. I'd start with the examining the breakers that trip and verifying at what currents are they tripping.

It would be ideal if you had another furnace that doesn't trip to compare results.

Of course it is possible for the furnace to start without tripping - it must have been designed to do that! Surely you know the electrode lift against time schedule for the furnace starting?? Is that time schedule being met?? Can you test the lift system without connecting the arc power?? Breaker or Transfo failure at these powers could kill someone. Three seconds is design maximum I would expect for a transfo short-circuit.

The best operation conditions and temperatures can be controlled by;

a) By varying the resistance of the arc by increasing or decreasing the distance between gaps, OR

b) By means of tap changing on the primary of the transformer.

Do check the protection relay for overshoot / undershoot (accuracy) as well as the CT's for errors.

In case yo are confident that process and mechanicals or OK, interchanging of current transformers can give a very limited desired effect on yellow phase to circumvent inadvertent tripping.

I guess the manipulation is a difficult approach for us and i read today that in thyristor controlled furnace we have this technology.

In thyristor control the firing angle is alterable. Although altering the angles from the optimum can stop the tripping but any minor deviations will play havoc with the process time / parameters, and believe me that once disturbed retracing the original scenario would be a nightmare.

If you have to experiment with the power electronics be very familiar with the process parameters.

OCB

EAF

Photo's can be unclear at times, but it sure looks like your electrodes are not in phase balance. Let alone setting in the holders square. I would suggest researching someone in your area that knows furnace control and mechanical maintenance to have a look at them. Tripping is usually a sign of poor and improper care of furnace.