capacitor across a contactor 1

[karalahana]

Hi, Can you tell me how can a parallel capacitor added to the control circuit lessen the effect of a high-residual-voltage occuring due to line-line capacitance here. I know that added capacitor means a higher total capacitor and it will create a bigger voltage and a longer time duration voltage across the coil, am I wrong?
:

http://www.franklin-electric.com/business/Watersystems/Service/AID/pdf/vol16no2.pdf

 

[karalahana]

Also it is stated here that "Franklin recommends that any time there is more than
300 feet of cable between the contactor and the control
switch, a capacitor should be added across the contactor coil"
so adding a parallel capacitor means we aggreveate the situation we can say , just looking at this sentence alone..

 

[ScottyUK]

It's not a parallel capacitor. Draw the circuit out: the cable capacitance is across the switch contacts, and the additional capacitor is across the contactor coil. With the switch open the two form a potential divider, with the additional capacitor being dominant so X_c is small. The volt-drop across the coil is therefore small and that across the switch is large. The value of the additional capacitor will be chosen to ensure that the coil reactance is over-compensated and the parallel combination is net capacitive at supply frequency to avoid any series resonant effects. Note that this technique will make the duty quite severe on the control switch - an interposing relay would be needed in many cases.

Thanks for the paper by the way - 'twas useful.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[waross]

To put it another way;
The overvoltage problem occurs when the coil inductance and the cable capacitance form a series resonant circuit.
With the added capacitor, the circuit is no longer a simple series circuit and is no longer resonant at 60 Hz.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rockman7892]

Scotty

It sounds like you are saying that even with the addition of a capacitor across a coil if not selected properly you can still create a resonance condition? So aside from selecting correct impedance to eliminate voltage across coil must you also be aware of capacitance value for resonance condition?

Why would the application of this capacitor make the duty of the control switch severe? Is this why sometimes the caps are fitted with a series resistor?

 

[ScottyUK]

It would be certainly be possible to create a resonant circuit with the series-parallel combination. The aim isn't to eliminate voltage across the coil, it's merely to swamp the inductance so the parallel LC combination is overall capacitive at 50Hz (or 60Hz) so that the cable capacitance and the additional capacitance form a capacitve voltage divider where the proportion of voltage across the coil is small.

Capacitor switching is hard on contacts because at switch 'make' the current is essentially infinite, subject to whatever source impedance is present. It is also severe at breaking because capacitors hold charge and at worst case the voltage across the opening contacts will reach a maximum of twice the peak voltage making the arc harder to extinguish.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[karalahana]

dear scottyuk, in the very essay it is stated that the capacitance pops across the lines of the control circuit isnt it?

 

[electricpete]

Interesting link - I can't say I heard of that before but it makes some sense as stated above.

Here are my thoughts on analysis of the system.

The series inductance of the cable is small enough to be negligible compared to the coil. There is remaining a single shunt (parallel) capacitance which you can add directly accross either the switch or the coil.... it's the same circuit either way with the series inductance neglected. I momentarily think aobut transmission line equivalent circuit with capacitance on each side and series inductance in the middle... then I realize that is only applicable for circuits long enough for wave effects to be relevant... not the case here and single capacitor at either end is all we need.

The period when coil/cable resonant frequency and it's relationship to power frequency is of interest is prior to switch opening. The conditions just prior to switch opening (coil inductance and capacitor voltage) create the initial conditions for the post-interruption period where the coil/cable resonant circuit will decay away while oscillating at it's own resonant frequency (regardless of whether that resonant frequency is near or far from power frequency). The purpose of separating the coil/cable resonant frequency from power system frequency then would only to decrease the magnitude of those intitial conditions (if you have a circuit with resonant frequency close to resonance than the total initial energy stored in coil inductance and coil capacitance will be large and it will take a long time to dissipate in system resistance.... although will certainly disappear sooner or later). For that matter if you have a circuit with resonant frequency near power frequency there is another bad effect - I'd think the high initial currents could cause coil overheating.... and worse yet there is not circuit protection involved because that circulating current does not flow through supply system fuses/breakers.

=====================================
(2B)+(2B)' ?

 

[electricpete]

Also of course there are at least 2 different inductances involved - the coil inductance will likely change when it drops out.

=====================================
(2B)+(2B)' ?

 

[electricpete]

Although if it drops out, we are not particularly concerned anymore so I guess that part is not particularly important. The condition of concern would be pulled-in coil inductance and cable capacitance resonant with power system frequency to create high initial conditions for decay period which might prevent coil from dropping out for awhile until energy is dissipated in resistance.

=====================================
(2B)+(2B)' ?

 

[ScottyUK]

karalahana,

"dear scottyuk, in the very essay it is stated that the capacitance pops across the lines of the control circuit isnt it?"

I'm not sure what you mean. The capacitor is clearly shown connected across the contactor coil in figure 2, which is the condition I and others have been discussing.

epete,

"There is remaining a single shunt (parallel) capacitance which you can add directly across either the switch or the coil.... it's the same circuit either way with the series inductance neglected."

Surely that shunt capacitance - the inter-core capacitance of the cable to the remote switch contact - can only appear across the switch?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[electricpete]

There is remaining a single shunt (parallel) capacitance which you can add directly accross either the switch or the coil

Not quite right since there is never any capacitance considered accross switch contacts. What I meant is that you can add the capacitor representing cable equivalent capacitance between the two cable conductors at either end of the cable (coil end or switch end). If we cut both conductors just downstream of the switch we have remaining 3 parallel components: coil inductance, cable capacitance, added-correcting capacitance. The latter 2 of course can be combined into a single effective capacitance.

Sorry I didn't mean to make circuit description a big deal. I imagine it's the same as what everyone else was picturing, but takes a lot more words to describe it than it would with paper and pencil.

=====================================
(2B)+(2B)' ?

 

[electricpete]

I was writing before I saw this:

Surely that shunt capacitance - the inter-core capacitance of the cable to the remote switch contact - can only appear across the switch?

The capacitance I was imagining was between conductors and if there is ground capacitance involved, then the conductor-to-ground capacitance of each conductor contribute to the conductor-to-conductor capacitance. As far as resonant circulating currents which might make the coil remain energized, ground capacitance is only important to the extent it contributes to conductor-to-conductor capacitance. That's the way I see it. Anyone looking at it differently?

=====================================
(2B)+(2B)' ?

 

[electricpete]

I was thinking the voltage source and switch were both at the remote end from the coil. Now I see the slightly different. You may be right... let me think about that.

=====================================
(2B)+(2B)' ?

 

[electricpete]

So, our equivalent capacitance goes wherever the long run of cable is.

The circuit is a tad bit ambigous because both the switch and the power supply are included within the label "300 feet"... the only thing outside that label is the coil.

If we ignore the label "more than 300 feet" and assume the switch is far away and power supply is close, you are right. That may be what the relative proportions of the circuit (without any label) are suggesting.

My initial assumption was that we have one long run from coil to power supply and switch and no big distance between the switch. Might not be exactly what they intended but then again the 300 foot label doesn't make it clear.

In real life there's yet another possibility - 2 long runs.

In summary, I will have to retract my statement that you can put the equivalent circuit either place - it applies applies only to my initial assumption, but not too all possible scenario's which might be assumed from this ambiguous diagram. Thanks for correcting that Scotty.



=====================================
(2B)+(2B)' ?

 

[electricpete]

Sorry everyone for making a mess of this thread. Please ignore everything I said about location of equivalent capacitance of the cable. Put it accross the coil as Scotty said. I'm done now.


=====================================
(2B)+(2B)' ?

 

[karalahana]

Dear electricpete what does this mean? I think this sentence is not clear muchif you have a circuit with resonant frequency close to resonance than the total initial energy stored in coil inductance and coil capacitance will be large and it will take a long time to dissipate in system resistance.... although will certainly disappear sooner or later) I think it has to do with the relationship between supply frequency and resonance frequency in way of proximity .I guess what you said is a very important and practical knowledge and I wish you explained it further and in much more detail...

Dear scottyuk , I had tried to say that the cable capacitance is not across the switch but is across the supply side, but When I thought I got that what you assumed is true and to think the way you think leads me to the solution, yes in reality cable capacitance is across the switch , thanks to both of you ...

 

[electricpete]

Dear electricpete what does this mean? I think this sentence is not clear muchif you have a circuit with resonant frequency close to resonance than the total initial energy stored in coil inductance and coil capacitance will be large and it will take a long time to dissipate in system resistance.... although will certainly disappear sooner or later) I think it has to do with the relationship between supply frequency and resonance frequency in way of proximity .I guess what you said is a very important and practical knowledge and I wish you explained it further and in much more detail...

The scenario I was considering when I wrote that would be long run of cable between the coil and the VoltageSource/Switch. The voltage source would have no influence after circuit interruption and would only play a role in establishing initial conditions for the resonant decay of LC circuit after interrutioion. The natueral (unforced) response of a given RLC circuit last longer when more energy is stored at the moment of the initial conditions. If circuit was resonant prior to interruption then that would provide more current in the coil and provide for more energy stored in circuit elements at the moment of interruption.

It's possible a completely different scenario is envisioned where the voltage source is close to the coil and there is a long run to the switch. In that case the capacitance between cables in the long run continues to complete the circuit even after the switch is open. The resulting loop still includes the voltage source and would have low impedance on if the the coil L and cable C (bypassing the switch) are resonant at power frequency.


=====================================
(2B)+(2B)' ?

 

[ScottyUK]

Pete's Scenario 2 is the one I was addressing, and which I am pretty sure Franklin are addressing in their paper. Thanks for the discussion fellas - just shows there are many ways to look at a problem!


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[electricpete]

I think you're right Scotty.

It's interesting that early in the article they said "While there is no known “one size fits all” method of
curing conditions of high control cable capacitance, it can
usually be solved by connecting either a resistor or capacitor across the contactor coil."
A resistor would help the circuit bleed down faster for scenario 1. But I don't think a resistor would not help for scenario 2.... Then later in the article they settle on using a capacitor.

But they also talk about the call remaining energized after switch opens as if it persists. Scenario 1 would only keep the coil energized until the energy decays away. So I think you're right they're talking about scenario 2.

Once again, sorry to OP and others for dragging this thread way off track. But an interesting discussion.

=====================================
(2B)+(2B)' ?