Cable residual voltage time constant 6

[njengr1]

An electrician got a minor shock from a 4 kV EPR insulated cable after the cable was deenergized for 45 minutes. While the fellow wasn't hurt the safety folks are interested in establishing a procedure that will prevent this.

Initially grounding the cable appears to have been done by momentary connection. However, the trapped charge in the insulation will drive a relaxation voltage to build up again on the core of the cable.

The timing of the voltage buildup and the maximum level is of interest. Can anyone provide any leads to a source that has information on this??

 

[davidbeach]

If it isn't visibly grounded it's live. That he would touch a live 4kV circuit means he's really lucky to still be alive.

 

[Marmite]

I echo davidbeach's comment. Most countries have stringent safety rules governing setting people to work on HV systems (meaning > 1kv in Europe). The fact that the "safety folks are interested in establishing a procedure that will prevent this" speaks volumes.There are lots of folks out there who are ignorant as to the specific hazards of working on HV systems. It sounds like you have fundamental policy and procedural issues to address rather than seeking a quick answer to this specific highlighted incident.
Regards
Marmite

 

[njengr1]

Thank you for your opinions. I was really looking for some helpful experience. Note that OSHA and most industrial safety procedures don't require grounding LOAD circuits whenr the entire switchgear lineup has been deenergized, cleared, tagged, and locked out.
If you have done some HV DC testing you may know that cables will buildup voltage after thay have been grounded post test, and the grounds have been removed. This residual voltage is due to charge stored in the polymer that gradually polarizes the shield and conductor; somtimes termed relaxation voltage. To avoid this, the grounds are normally left on 4 times as long as the DC test voltage was applied (rule of thumb).

This brings up the case in hand. How long will cable insulation retain a trapped charge when operating in an AC circuit (as opposed to a DC test)? How long need a ground be held to release the trapped charge?

Note that where grounds are required they are used, but a ground inadvertantly not removed can be disastrous. When a drawout contactor on a RADIAL circuit is being service the line side is locked-out and grouunded, but the load side is not. It was the load side cable that developed the residual charge (less that 200V and micro energy)that was felt when the technician's knuckle hit the load terminal while replacing the drawout contactor into a dead enclosure.

 

[burnt2x]

njengr1,

"Note that OSHA and most industrial safety procedures don't require grounding LOAD circuits whenr the entire switchgear lineup has been deenergized, cleared, tagged, and locked out."

That's a totally false statement! Try 29CFR 1910.269(N) and 29CFR 1910.269(W).

 

[njengr1]

burnt2x
Your getting close to the point. from1910.296
"General." For the employee to work lines or equipment as deenergized, the lines or equipment shall be deenergized under the provisions of paragraph (m) of this section and shall be grounded as specified in paragraphs (3) through (9) of this section. However, if the employer can demonstrate that installation of a ground is impracticable or that the conditions resulting from the installation of a ground would present greater hazards than working without grounds, the lines and equipment may be treated as deenergized provided all of the following conditions are met:
1910.269(2)(i)
The lines and equipment have been deenergized under the provisions of paragraph (m) of this section.
1910.269(2)(ii)
There is no possibility of contact with another energized source.
1910.269(2)(iii)
The hazard of induced voltage is not present.
.....

It is precisely the last point of this excerpt that I'm after. It's easy to see where the electric field and magnetic fields of adjacent circuits can produce an induced voltage. It's not clear if a trapped charge in a cables dielectric will provide a measurable relaxation voltage after being deenergized from A.C.operation.

The 1910.269 section OSHA leaves open the decision to apply grounds where they can be do more over harm than good. This is the point of the section.

The OSHA section re caps clearly recognizes the trapped charge physics. That is certainly not in question.

 

[Marmite]

I think that you are missing the point njengr1. You are never going to be able to perform a calculation that reliably comes up with the answer that after x mins that conductor is safe to touch. Who would be daft enough to trust the calculation if their life depended on it?
You need to devise test/work procedures, including the use of PPE, which reduce the risk of contact with a charged conductor to an acceptable level.
Regards
Marmite

 

[burnt2x]

njengr1,
Please read my reply fully! It's not just 29CFR 1910.269 that I wanted you to refer to. It would be helpful if you treat your charged cable as a "capacitor". Section 1910.269(w)(1) specified a wait time of 5 minutes(?) together with other procedures that you should do to prevent electrocutions. Charge decay depends on the circuit parameters and you start on that idea.
I also believed marmite is correct that nobody will trust on any estimates you can come up with regarding wait times before conductor is safe to touch. There are many ways to minimize the risks, and marmite just nailed it!
The fact that your electrician got jolted is an indication that need to ground your cable (discharge whatever charge it has) before tinkering with the cable.

 

[ausphil]

there seems to be 2 arguments going on here.

njengr1 seems to be arguing the recovery voltage phenomenon (or relaxation or polarisation), whilst the rest of the posters are discussing procedural methods to stop workers getting injured when working on high voltage systems.

njengr1, essentially you have a capacitor that needs to be shorted to ground to dissipate the stored energy in the insulation. If you were to go down the track of calculating the time to discharge, you would need to know:
- the point at which the cable was isolated (ie at what point on the voltage waveform the cable was de-energised, as this directly determines the charge at the start of the discharge process),
- the insulation resistance of the cable, as this will determine the rate of discharge (and polymeric insulation is almost too good, as it has very high insulation resistance),
- if there is any other equipment in the system that may speed up the discharge rate (ie magnetic VTs, as capacitiv VTs won't help you),
- and most importanly of all, you must determine at what voltage level, you ae going to deem it safe for the staff to touch the cable (knowing that the stored energy is proportional to the square of the voltage across a capacitor).

Also, you've got to be aware that the rate of discharge is not linear, and the less charge that is across the cable, the slower it will discharge, so it depending on where you decide it is "safe to touch" you may be waiting for many days.

As the others have discussed, the only way to safely access de-energised high voltage equipment is to apply an earth (or a ground, depending on where you come from).

Coming from a testing background, I would never dream of touching any piece of equipment unless it was earthed, either after it was de-energised from the system, or after applying test voltages.

You mention that a "ground inadvertantly not removed can be disastrous". Most of us would agree that they would rather run the extremely small risk of damaging equipent due to energising onto a ground, than the very high risk of injury or death that is apparent when not providing an earth after de-energising high voltage equipment.

We had gone down the track of calculating stored energy in cables after discharge, however it was to determine how long after de-energising high voltage cables (132kV) an operator could safely apply a portable earth (ground). On our paper / oil cable systems, by waiting 15 minutes, the energy had discharged enough to safely let the operator apply the earth (an arc was still drawn, but it was minimal). To get to the same charge level in our newer XLPE cables, we needed to wait for close to 2 weeks! Needless to say, we installed magnetic VTs in the circuit to ensure the discharge hapened quickly.


ausphil

 

[DanDel]

It's very simple; the basic rule is, never touch a cable without testing for live AC voltage and then grounding (from a safe distance).

 

[njengr1]

ausphil
Thanks for your input. At last someone got it. The phenomenon is the issue. I, like you, have been involved at the high end (>>138kV). Much of the SF6 filled buses and breakers and isolating switches at these levels have built in grounding and isolating switches. Sometimes with TV or visual ports to verify the grounds are in place. Overhead designs also have mechanically operated grounding switches with the correct interlocks. At lower voltages these built-in devices are to some degree absent.

The incident here involved a fellow that isolated and gounded the supply side. He then withdrew a 4kV fused switch to check the contacts and mechanical works. Then he inserted the starter in the tracks and cranked it in. Somehow he touched the load side of the open switch and got a slight shock from the potential caused by the dielectric absorption of the EP rubber cable. The voltage was at most 150 to 200 volts with apartently no damaging energy.

No one is trivializing this incident. I've spent more that 40 years in the power business in the North and South America, Africa, Europe and Asia and have not had any really bad experiences. Over the years I have found that, to keep out of trouble, it pays to fully understand the physics of the problem. Hence, the question originally asked.

Your synopsis of the problem has the parameters nailed down. The dielectric constant and bulk resistivity will generally dictate the time constant of discharge. However, due to the fact that the cables potential distibution is capacitive (non-linear) when excited by A.C., as opposed to linear with D.C. the discharge time constant will be changed. So, I was interested in hearing about some work that may have bearing on estimating the discharge time for various cable voltages and insulating materials. Much paper has been written on the line charging and operational parameters, but not much on discharging.

One might rightly ask "Who cares?". Well and engineer in a forensic investigation just might like to know how high a voltage was possible, and at what time.

Exception to the "rules" by technicians occur often in various parts of the world (including Scotland) and lead to accidents. Sometimes it is a misunderstanding of the consequences that invites using a shortcut. A good explanation of the physics involved convinces a tech not to go adrift.

 

[stevenal]

I don't understand your distinction between AC and DC. Once the potential source (AC or DC) is removed, the remaining charge is DC. This charge is then drained through the distributed resistance and capacitance of the cable. The only difference is that the DC sourced charge is predictable, while the AC sourced one will depend on the moment of interruption.

To answer the question: How long do you need to apply the ground to discharge the cable to a safe level?: you will need to model the distributed parameters of the cable. You also need to model all mutual inductances that will tend to recharge the cable once the ground is removed.

As far as the OSHA exceptions, this bird has flown. You have already proved that the hazard of not grounding exceeds that of grounding, and that induced and/or trapped potentials exist.

 

[davidbeach]

You also need to model all mutual inductances that will tend to recharge the cable once the ground is removed.

That also includes all the mutuals that you don't know about but exist to bite somebody.

Like I said, if it isn't visibly grounded it's live. Design your test and maintenance procedures in such a manner that grounds can be safely applied and removed, including an accounting for all grounds installed prior to reenergization. There are far worse things to happen though than to close into a set of grounds provided the protection is set to clear instantaneously, after all, there is no arcing associated with a bolted 3-phase fault. But if the grounds are all removed there won't even be that to worry about. During the work though, it is visibly grounded or it is live.

 

[TurbineGen]

njengr1, I will second the posts on grounding the cable. Can you accurately model all the mutual inductances and come up with an accurate time for cable discharge? Or is it easier or safer to just ground the cable?

I was in a seminar not too long ago and I remember the instructor mentioning the L/E ratio (lawyers divided by electrical people). Check your phone book and see how many pages of attorneys there are and then ask yourself if you really want to risk it. If someone gets carried by 6, you will be tried by 12.

------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.