Understanding ANSI Electrical Device Numbers Without Losing Your Mind

When I first started working on substation control schematics, the whole numbering system felt like someone was trying to gatekeep information. Turns out it's just been around since the 1960s and hasn't changed much because it actually works once you stop fighting it. The ANSI C37.2 standard assigns a two-digit number to every type of relay, contact, or device function in an electrical system. So 87 is differential protection, 27 is undervoltage, 59 is overvoltage, and so on. Most of it is standardized by IEEE/ANSI and widely adopted from North America through the rest of the world, even if local utilities tweak things a bit. The trick is not memorizing every number. Nobody does that anymore. What matters is knowing how to navigate the list efficiently and recognizing the patterns. The numbers cluster loosely by function — the 20s are mostly voltage-related, the 50s are protective relaying, the 70s involve switches and breakers, the 90s cover generators and turbines. If you're looking at a schematic and see F62, you can guess it has something to do with a failure indication before you even open the manual.

What Are Ansi Electrical Device Numbers and Why Do They Matter

At their core, device numbers are a shorthand language for engineers who need to communicate across projects without writing paragraphs of description. When I'm reading a relay setting sheet from a contractor and it says "49HT trips on 79-3," I instantly know what's happening: the transformer hot spot temperature is triggering a trip based on recloser failure. That's three pieces of information in six characters. Writing that out in English would take twenty seconds and take up three lines on a page. In a 400-page manual, that adds up fast. The standard device numbers come from the IEEE C37.2 standard, originally ANSI C42.16. It's maintained by the IEEE Standards Association now but the numbers haven't moved much since the late 1990s revisions. You'll find them in IEC contexts too, though IEC 61346 uses a different layering approach that some people mix together carelessly.

How to Actually Use This in Your Work

I keep a laminated quick-reference card at my desk and a full table in the shared drive, but the real workflow is about reading schematics backwards from the tag. You see a symbol on the drawing with the designation "62ATS" next to it. The base number 62 means time-delay relay. The supplementary letter S means start and T means trip or transfer. So 62ATS is a time-delay relay used for starting and transferring. The symbol might look different depending on the drafting standard your firm uses, but the alphanumeric tag stays consistent across drawings, setting sheets, and logic diagrams. Here's where things get tricky and where most juniors get burned: modifier letters and subscripts stack on top of the base number. Two letters modify the base function. A subscript in parentheses tells you which operating coil or contact is involved. So 25-1 means synchronizing check relay number one, and 25-2 is the second one. You'll see this especially on bus arrangements with multiple sources where you need more than one synchronizing check relay. I've seen people read 25-1 as "25 dash one" and treat it as a completely different category. It's not. It's the same device, just numbered differently for identification. The practical side of this is learning to read the tagging convention rather than the drawing style. Drawing styles change from company to company. My last two employers used different symbols for the same 86 lockout relay. The tag was identical in both cases. That's the invariant you should be using as your anchor point when cross-referencing documentation.

Get the Full Details

ANSI Device Numbers List | Electrical Engineering Reference
ANSI Device Numbers List | Electrical Engineering Reference

A Real Problem I Ran Into and How I Fixed It

Last year I was reviewing relay logic for a 138kV breaker replacement and the original contractor had labeled a backup overcurrent element as "51X-1" on the logic diagram but "51G-1" on the actual relay setting sheet. The relay manufacturer used X as a generic time-overcurrent placeholder while the engineering firm had defaulted to G for ground. Both are valid per the standard — 51X is a generic inverse time overcurrent and 51G is specifically ground overcurrent — but the inconsistency meant nobody noticed the mismatch during the review cycle. The backup element was configured for phase current instead of ground current, which would have been a real coordination problem if a ground fault had occurred on the feeder. My workaround was straightforward but tedious: I ran a script across all the PDF documentation that pulled every instance of "51" followed by any letter and flagged where the supplementary designator didn't match between the logic diagram and the setting sheet. It took about forty minutes to set up and caught eleven more mismatches I hadn't looked for. Since then I require that anyone submitting relay documentation include a tag consistency check as part of their deliverables. It usually adds about ten minutes to their process and catches errors that otherwise show up during field troubleshooting, which is exponentially more expensive.

Things the Standard Doesn't Tell You

One counter-intuitive thing about ANSI device numbers is that a single device can legitimately carry multiple designations depending on which function you're describing. A protective relay with both differential and distance elements isn't "wrong" for being tagged as both 87 and 21. On complex bus or transformer protection schemes you'll see this constantly. The convention is to tag the primary function first and secondary functions as needed. But I've also seen contractors insist on splitting everything into separate devices just to make the numbering "cleaner." That creates twice the documentation for no actual gain in clarity. The standard explicitly allows multiple numbers per physical device. Another thing that catches people off guard: the numbers aren't sequentially assigned by era of introduction. There are gaps everywhere — 48, 60 through 63, 74 through 78, and a bunch of higher numbers — because different functions were standardized at different times by different committees. Those gaps aren't reserved. They're just empty. Sometimes new functions pick up nearby numbers anyway, which is why you occasionally see someone create a device number like 115 for a function that doesn't appear in the published standard. That's technically outside the formal numbering system but it happens in practice, especially on custom or proprietary equipment. There's also the issue of what happens when your system has more instances of the same function than the standard naturally supports. I worked on a project with five separate unit transformers, each requiring its own overvoltage relay. 59-1 through 59-5, right? That part is fine. But then we needed auxiliary overvoltage relays on the same busses for battery charger supervision, and suddenly the numbering got crowded. We ended up subdividing into 59-1A through 59-1D for the transformer supervision elements and 59-2 through 59-4 for the bus-level elements. It's not elegant but it's documented and traceable, which is the whole point of the system.

Where the System Breaks Down

The biggest limitation is that ANSI C37.2 was written for electromechanical and early solid-state relays. Modern numerical relays combine dozens of functions into a single physical unit, and the standard doesn't have great answers for how to tag those cleanly. A modern SEL or GE relay might contain overcurrent, earth fault, frequency, rate of change of frequency, loss of excitation, and impedance elements all in one box. Do you tag the entire device as 21-1 and then reference 51G-1 internally? Do you create separate virtual device numbers for each function? There's no single correct answer and different engineering firms do it differently. I've seen both approaches used successfully and both used poorly. Another practical limitation is the growing disconnect between the standard numbers and how relay manufacturers actually label their software interfaces. A relay programmer might see "46-2" in the documentation but the touchscreen menu calls it "Negative Sequence Current Element 2." That's not a problem for someone who knows both sides, but it creates a translation step that slows down field technicians who are only familiar with one naming convention. Some utilities have solved this by creating internal cross-reference tables that map the ANSI designator to the manufacturer's menu path. Others just accept the friction and move on. If you're starting from scratch on a new system and the scope is large enough to justify it, I'd recommend adopting IEC 61346-2 for the device identification layer alongside ANSI C37.2 for the functional numbering. The IEC standard gives you a more rigorous hierarchy for tagging physical devices independently of their function, which resolves a lot of the ambiguity that comes up with modern multi-function relays. It's not required anywhere I know of, but it's what I've switched to on projects over about two hundred bays because the administrative overhead pays off by the time you hit commissioning.

Electrical Protection Device Numbers at Joyce Collins blog
Electrical Protection Device Numbers at Joyce Collins blog

Quick Reference for the Most Common Numbers You'll Actually See

27 — Undervoltage relay. Used for lost voltage detection, source failure indication, and motor protection lockout. 51 — AC time-overcurrent relay. With subscripts: 51G for ground, 51P or just 51 for phase, 51X for generic. 52 — Circuit breaker or disconnect switch. Not the relay, the actual switching device.

59 — Overvoltage relay. 59N is neutral overvoltage, 59PG is programmable overvoltage on newer relays. 62 — Time-delay relay. One of the most used supplementary numbers. 62ATS, 62DT, 62TR are everywhere. 63 — Pressure switch. Usually for SF6 or oil pressure monitoring on breakers and transformers.

67 — Directional overcurrent relay. 67N is directional ground, 67P is directional phase. 74 — Alarm relay. General purpose alarm output. Often confused with 86 but they're different — 74 is an alarm, 86 is a lockout. 81 — Frequency relay. 81O is overfrequency, 81U is underfrequency, 81RO is rate of change of frequency.

Ansi Device Codes List – Table of ANSI IEEE Standard Device Numbers – UODP
Ansi Device Codes List – Table of ANSI IEEE Standard Device Numbers – UODP

86 — Lockout relay. The master trip contact that shuts down a bay when something critical fails. 87 — Differential relay. One of the most important numbers on any substation schematic. 87B for bus, 87T for transformer, 87L for line, 87G for generator. 94 — Trip relay. The final element in the trip chain, usually directly energizing the breaker trip coil.

You don't need to memorize this list. You need to recognize that when you see these numbers on a drawing, they're telling you something specific about what the device does, not just what it is. That distinction matters when you're tracing a fault through a four-wire schematic at 2 AM and you need to figure out why a breaker didn't trip when it should have. The full ANSI C37.2 standard is available through the IEEE Standards Association website. There's no free version of the current edition, but older revisions sometimes surface in university libraries and on government documents from the DOE and NERC. The numbering itself hasn't changed enough between revisions that the 1996 version is still useful for reference purposes on most existing systems.