-
Leon Zhang Sales Manager
-
Email: sale@relayprotect.com
-
Phone/WhatsApp: +86 13655813266
Field notes: relay setting and wiring errors we have caught
Three wiring and setting errors that reached commissioning on our own projects. Each is described only by what it was and how it was found; no client, plant or product model is identified. We publish them because the same three mistakes are made across the industry, and because a buyer should be able to judge whether we would catch them on their site.
1. A 100 A pickup where a secondary current belonged
A feeder overcurrent setting sheet arrived with the instantaneous element at 100 A. Relay current inputs are rated 1 A or 5 A, so 100 A is 20 to 100 times the rated input – at or beyond the top of the setting range on most elements, and a current the input never sees in service. On a 600/5 CT it corresponds to 12,000 A of primary current, well above what the feeder can deliver, so the element was installed, healthy and functionally absent. The figure was a primary current written into a field that expects a secondary one; the value that should have been typed in was 0.83 A.
How it is caught: every row of a setting sheet carries its CT ratio and its unit, and a secondary injection test at twice the expected pickup proves the value in under a minute.
2. 5 A current transformers into a 1 A differential relay
The current transformers at site were 5 A secondary. The transformer differential relay supplied was the 1 A input variant. Both datasheets were correct on their own, which is exactly why it passed review. A 5 A secondary current driven into a 1 A input reads five times high: every setting expressed in multiples of rated current lands five times away in primary terms, the input carries about 4.4 times rated current at full load, and the linear measuring range falls to one fifth. A 1 A relay on a 200/5 CT runs out of linear range at roughly 1,600 A of primary current, so an ordinary through-fault pushes the input far beyond it.
How it is caught: the CT secondary rating and the relay input rating written on the same line of the bill of materials. Where the CTs cannot be changed, an interposing CT is the engineering answer; rescaling the settings only corrects the arithmetic and leaves the measuring range and the continuous thermal rating wrong.
3. Reversed excitation transformer polarity
On a waste-heat generation project at a steel plant the excitation transformer was wired with its voltage polarity reversed. It surfaced during the trial run, when the excitation voltage did not follow what the AVR was commanding. Reversed polarity is not a magnitude error, so a voltmeter reads the same number either way and a point-to-point wiring check can be satisfied while the phase is 180 degrees out. It is found by recording the excitation build-up as a curve of excitation voltage against AVR reference, and by a DC polarity check against the winding marks before the first energisation.
How it is caught: polarity verified before the first build-up, and the build-up recorded as a curve rather than as a single reading. Zero degrees and 180 degrees read identically on any single-channel meter.
What the three have in common
None of them was a faulty device. Each was a correct component used against the wrong reference – the wrong unit, the wrong rated current, the wrong phase. That is the class of error a hardware fault log never records, because nothing is broken and every device reports healthy. It is also the class of error a factory test cannot find, because the factory never sees the plant. Every one of them was found either by a document review before energising or by a commissioned test that measures against a reference, not by replacing parts.
Questions these three cases raise
Can a wrong setting stop a relay operating without any alarm?
Yes, and all three cases above are examples. A pickup entered in the wrong unit, a current input rated differently from the current transformer, and a phase reference reversed by 180 degrees all leave the device reporting healthy – no self-test alarm, no communication failure, no error log entry. What finds them is a test measured against a reference: a secondary injection at a known value, a polarity check against the winding marks, or a build-up curve compared with what the regulator commanded.
Can a 5 A current transformer be used with a 1 A relay if the settings are rescaled?
Not properly. Rescaling corrects the arithmetic and nothing else. The current input still carries about 4.4 times its rated current at full load, which is at the continuous thermal limit before any fault occurs, and the linear measuring range still falls to one fifth – a 1 A relay on a 200/5 CT runs out of linear range at roughly 1,600 A of primary current. The engineering answers are the 1 A relay variant, an interposing transformer, or 1 A current transformers.
How is excitation transformer polarity checked before the first build-up?
With a DC injection against the winding polarity marks, or with a two-channel instrument comparing the phase of primary and secondary. A single-channel voltmeter cannot answer the question, because zero degrees and 180 degrees produce the same reading. Recording the excitation build-up as a curve of excitation voltage against the AVR reference adds a second check, and it is the last one available before a breaker closes onto the machine.
What should be checked first on a setting sheet?
Not the arithmetic. Every current row should carry its CT ratio and its unit, and the relay input rating should be stated on the same page as the CT secondary rating. Of the three cases above, two would have been caught by that single line of documentation, and neither involved a calculation error at all.
Related reading
- Setting an overcurrent relay correctly – pickup, time multiplier and the primary/secondary conversion
- Protection relay setting calculation – where the unit of every row has to be stated
- Protection relay testing and commissioning – secondary injection and what it proves
- Transformer differential protection setting – bias, pickup and CT matching
- Current transformer CT – ratios, classes and secondary ratings
- Generator synchronisation panel – what has to be proven before the first paralleling
Installations by country: project record, 93 projects in 43 countries with the voltage band and plant type for each.
Have the checks applied to your scheme
If you are commissioning a protection scheme, send the single-line diagram, the relay list, the CT ratios with their secondary ratings, and the existing setting sheet. The review costs nothing and takes a working day. Send it through the contact page, or start from the project record to see which voltage classes and plant types we have already worked in – 93 installations in 43 countries.
Where these errors show up most often is on the 87T differential scheme, the 51N zero-sequence earth fault relay and the ASL-441H feeder relay — the three settings most often copied from another project.
