When designing a complex system that incorporates different grounding types (TN/TT and IT), we must consider the safety characteristics of each. How can we cleverly utilize earth leakage relays and insulation monitoring devices to improve safety?
When designing a complex system that incorporates different grounding types (TN/TT and IT), we must consider the safety characteristics of each. How can we cleverly utilize earth leakage relays and insulation monitoring devices to improve safety?
How can we cleverly utilize earth leakage relays and insulation monitoring devices to improve safety?
Please could you be more specific? What safety issues did you have in mind?
Where ADS is used in TN, TT and IT systems, IEC 60364-4-41, 411.3.1.1, requires that simultaneously-accessible exposed-conductive-parts are already connected to the same earthing system, which helps control touch-voltage.
There are, additionally, provisions in 411 of IEC 60364-4-41 to ensure safety.
Although casualy it may seem that the 2 devices perform a similar fuction - to warn and then remove the supply if there is a leakage path from a live conductor to earth, there are in effect solutions for 2 very different situations here.
Normally insulation monitoring is associated with floating (IT) systems where there is almost no CPC current by design, so leakages to earth of a scale that would be early warning of insulation failure can be detected. Typically thresholds corresponding to insulation resistances of hundred plus k ohms, and only a small circuit and a few dedicated devices would be protected by a single detector. Also it can be arranged, as initial leakage is not dangerous, to alert but not to disconnect, so the current process may be safely concluded before shutting down for repair. This is an expensive approach, but may be justified in some settings, perhaps, say in a surgical theatre.
Earth fault relays are in effect RCDs with more options, and are looking at the imbalance between live and neutral conductors - the assumtion being that whatever goes out but does not come back, has probably done something bad, and worthy of operating the Automatic Disconnection. This is normally done with one relay covering a TN-x system or reasonable extent, so there will be CPC currents as a matter of course, due to wiring capacitance and the nature of the likley loads, that amount to a few tens of mA. The most sensitive RCDs in common use have a threshold of 10mA, and 30mA devices are the CENELEC standard for safety of life.
However, a 10mA CPC current let alone a 30mA one, is a not really an early warning of insulation failure, rather at 25k ohms for 10mA and ~ 7k for 30mA, it is only detecting pretty much total failure sometime after the fact. If that level of fault occurs at a single point, (in a volume of a few cubic centimetres perhaps ) it is likely to be noticeably cooking, and there is a risk of ignition. As such the normal sensible option is immediate ADS rather than pre-warning.
I'm not seeing any benefit from combining the two on the same circuit as one more or less excludes the other.
What one might imagine, at least for single phase circuits is a device that looks at the phase of the CPC current relative to the phase-neutral voltage, so that resistive leakage - dissipation, shocks and possible ignition can be separate from capacitive - low loss cable capacitance charging and discharging, and capacitance in EMC filters. Given the sort of accuracy with which phase angles can be reliably estimated without building a lab grade vector volt meter, one might imagine that one could see a few mA of resistive leakage under a few tens of mA of capacitive.
Mike.
I would add that, whilst it is permitted by the standards to use IMD and RCD, along with other protective devices, there are some important notes, particularly around the use of IMDs and RCMs:
1. An IMD is not a "protective device". It can be used to monitor for a "first fault" in IT systems, and initiate disconnection (or action by someone to remove power or provide ADS (for second fault in IT system, or faults to Earth in TN and TT systems).
2. An RCM, similarly, is not a "protective device. It also can be used to monitor for a "first fault" in IT systems, and initiate disconnection (or action by someone to remove power or provide ADS (for second fault in IT system, or faults to Earth in TN and TT systems).
3. An RCD or OCPD can be used as a protective device to achieve ADS for a second fault in IT system, or for faults to Earth in TN and TT systems.
The reason for statements in 1. and 2., is that there are no standards to ensure disconnection for relays, contactors or other disconnectors actuated by these monitoring devices ... they are, as stated, 'monitoring' devices.
In TN and TT systems, faults to Earth lead to an immediate shock risk, so use of RCM or IMD without guaranteed disconnection time is not really a safety improvement.
In IT systems, yes there is an improvement in safety to provide some action, including causing supply disconnection for the affected circuit, after first fault (for which there need not be a danger ... BS 7671 and IEC 60364 don't really account for multiple faults).
Finally, as to whether RCMs can be used with IMDs, that might depend how they work, i.e. whether they are compatible.
IT systems with insulation monitoring devices are ubiquitous on ships and very common offshore. However, there is conflicting guidance from HSE and IMD manufacturers about the applicability of earth leakage protective devices in IT systems.
Guidance from HSE basically recommends use of RCDs on final circuits and implies that these may indeed tip on the 1st fault:
"The IT electrical distribution system insulation impedance to earth will often be at a value low enough to allow an RCD to function when an earth fault occurs. This is particularly so with electrical distribution systems that incorporate harmonic line filters."
It also makes the point that the fault impedance of any 1st fault may be high enough the limit the fault current of any 2nd fault such that the protective device doesn't operate correctly.
https://www.hse.gov.uk/offshore/infosheets/is2-2011.htm
However, a pamphlet from the IMD manufacturer Bender argues that RCDs should only be used in IT systems in exceptional cases only. This is because depending on the position of the RCD in the circuit, any leakage current via parasitic capacitance in the supply cable, providing the return path, may appear behind (load side of) the RCD and prevent it from "seeing" any differential current. For this reason the RCD, if installed, is recommended to be located as close to the load as possible. In situations where you have parallel RCDs it difficult to guarantee that the device actually associated with the circuit under fault will be the one that trips.
If IT systems are incorporated with IMDs it's imperative that operators are trained to find and rectify 1st fault situations without delay.
For systems where IT is selected for fault tolerance of critical loads, one option is to have TN systems with 30mA RCDs, downstream of the IT network, on parts of the system that are less critical and more likely to have earth faults e.g. socket outlets, laundry and galley equipment etc
We're about to take you to the IET registration website. Don't worry though, you'll be sent straight back to the community after completing the registration.
Continue to the IET registration site
