• 4 Posts
  • 36 Comments
Joined 26 days ago
cake
Cake day: June 16th, 2026

help-circle





  • No problem! This is actually more in the factory level of power than it is grid. 480v is the common voltage inside plants.

    But sometimes you’ll see different voltages (like 380 and 720). 600v and below is considered low voltage, but sometimes 720 is considered low too if its feeding machinery. Everything above that is usually considered high voltage. But the mysterious medium voltage exists too, I’d call that somewhere up to like 20kV.

    The 13kV in this picture is controlled by the plant. Some equipment uses 13kV, but usually plants just get 13kV (or other distribution voltages) and distribute it themselves. It’s really fun equipment to operate, big switches that you have to put your body into, and it makes big mechanical thuds and you’ll hear the electric make and break connections before the switch dies. It sounds like bzzzTHUDmmmmmmm








  • It sounds like you all don’t have task specific job safety plans?

    Depends where I’ve worked, what the task is, and how it’s structured.

    Data center? Yes, we had quadruple redundancy. Boring job.

    Auto plant? PMs, yes. Reactive? Sometimes, but if the line was down, we weren’t getting them printed out. The maintenance training program was incredible though, trained the troubleshooting process, even the dumbest could have written a work plan for any piece of equipment. All equipment had PPE requirements and showed LO locations and procedures. For reactionary work, work plan was quick chat with everyone involved, then follow SOP for de-energization, verify de-energization. The work was done up front similar to how you’ve described, but on every team. Nothing got missed.

    100 year old steel mill? PMs and only sometimes. Fun place to work, but that job was almost 100% fire fighting. Safety culture didn’t exist, especially when private equity took over lol. Early on had an untrained maintenance guy in street clothes operate a tripped breaker rated over 90 cal, he did not look to see what caused trip. That was when I was able to create a glove and training program lol. But that plant is the biggest reason why I’m against double labels. That plant hasn’t filled my position, and when I left, very little of it got delegated. The arc flash stickers are probably the last line of defense, I don’t think the drawings have been looked at since I left. That’s common in a lot of plants I visit at my new job, unfortunately.

    I’ve had arguments with arc flash study providers over it - unfortunately this isn’t necessarily a dumb new engineer - these are well seasoned vets. And double labels in a book, or on a drawing makes sense. But when you look at plants that do less than bare minimum safety, it highlights how important those stickers are. The well trained facilities with good safety culture will have drawings to get information that’s missed on the removed label (primary side incident energy). But safety culture can go to shit overnight, turnover, etc. Sticking with 1 label per enclosure ensures that the safety you provide on those stickers will outlast your program.

    Labeling requirements per NFPA70E are nominal voltage, arc flash boundary, incident energy or PPE category*, and minimum PPE requirements. I believe that minimum PPE requirements should be carried through entire enclosure, and adding that to the book would close the transformer debate. But I think its been unaddressed for too long that they should be adding a note specifically calling out transformers and DPs.

    *this requirement is why you’ll see a PPE level listed instead of a PPE category if the incident energy is precisely calculated


  • That makes sense. I shouldn’t have bashed IR windows. The plants that mandate rigid conduit everywhere, IR windows make sense and are good. For the limited budget plants, I think IR windows on transformers are low priority. But infinite budget, IR windows are great.

    IR windows have a lot more use cases outside of transformers where I think they are more important. And everything has to do with frequency and down time availability.

    The last job I worked was a steel mill. I was in charge of all electrical distribution from utility to disconnect prior to production equipment. Because of the metal dust, we needed to vacuum out distribution equipment yearly at a minimum and NFPA70B recommended some tasks yearly for us. I left when my 2 day outage on a 3 day weekend was canceled. I consolidated the outage plan to a single day, and they still wouldn’t let me have it. Once we got a year behind, I left. I’ve been gone a year, and they still haven’t done it.

    I had the main transformers on a 5 year EOL plan because of oil samples and age. The lead time was 18 months for each, and they needed 3, lol. The EOL plan was also neglected.

    I would’ve felt more comfy with IR windows lmao but I wasn’t going to stick around to be the scape goat when the plant goes down.


  • Arc flash incident energy IS the explosion. That’s the size of it. Anything over 40cal is considered dangerous (no PPE exists)

    The distances are based on the covers being off as opposed and someone working on it. NOT during normal operation. It is generally safe to be near electrical equipment if it is not being worked on/opened up.


  • The primary and secondary can be in different enclosures.

    Thus is true. In which case the labels should be applied to the separate enclosures.

    I agree the sticker config does not make sense for something with two enclosures.

    Glad we agree.

    Shock hazard and arcflash hazard analysis is independent, as are the PPE requirements. You need to account for both when planning work. The Shock hazard PPE only looks at the voltage of the system (see table 130.7©(7)(a)) whereas arcflash PPE needs to be suitable for the incident energy at the distance they will be performing their task.

    I agree with all of this. The nuance I’d add is that PPE needs to be suitable for the shock and incident energy they could receive while performing their task. The common field example would be the controls guy working on their low voltage DC inside a box with exposed 480v. Even if they 480v is further than 42" away from the control part they’re working on, they should still be wearing gloves. Similar with arc flash boundaries. For my guys, I consider the arc flash danger to be a plane parallel to the enclosure door as opposed to a single part, which goes above and beyond NFPA70E, but is becoming more and more industry standard/consensus. Basically, if you enter an enclosure you need protection for all exposed parts.

    The secondary sticker is basically saying there is no safe way to work live, regardless of shock protection, because the arcflash incident energy is too high.

    I don’t think I disagreed with anything prior to this statement. But this is where your argument starts to fall apart. The arc flash boundary for the secondary side is 613 inches, or 51 feet. Since the labels are both placed on the outside of a single enclosure, and moreover, they are placed together (as opposed to one clearly on the primary side and one clearly on the secondary side). Because we don’t live in a fantasy land, we can assume that the transformer is not 51’ in any direction.

    This means that there is no working condition where you could be safe while working on the primary side and NOT exposed to the secondary side.

    PPE is the last line of defense. Administrative, engineering, etc. all come before PPE. And one of the biggest causes of error is human error. Meaning designs should be designed to limit human error.

    While drawings are important to look at, the reality is that not every maintenance worker will look at drawings. And especially during an outage or failure, steps are more likely to be skipped. To reduce errors, arc flash and shock hazard stickers should not have different information they should only list the highest required PPE or highest determination of PPE (as in highest voltage, and highest arc flash potential) of all internal components. It’s the same reason why MCCs and Distribution panels should only have one label - despite having breakers that technically have different incident energies. Oddly enough, the debate of placing different stickers only seems to exist around transformers (with some exceptions where distribution panels have cabinets that could isolate arc flash potential).

    The reality is that the stickers are technically a higher level of protection than simply PPE. They should be clear, and they should give clear direction. I cannot enter the primary side of this transformer without being exposed to the arc flash potential of the secondary side, so there is no reason to list the arc flash boundary of the primary side. The arc flash boundary of the transformer is the higher of the 2 boundaries, and it should be labelled as such. Similarly, you need class 2 gloves to protect you from the guts of the transformer. A transformer is one single device and should be treated as such. Control boxes have 480v and low vDC, we don’t go into the cabinet with no PPE just because we’re working on vDC.

    I do applaud the training of your plant though, it sounds like your guys have more awareness than many plants I’ve seen. But even with that training, you are introducing human error factors when you label like this.



  • I mean, all that thermal imaging is really doing is checking for hot spots. Just follow NFPA70B to find when you should be performing PMs on this transformer. But really the thermal imaging would tell you to tighten some bolts. Why not just clean the contacts and tighten the bolts? That’s a shit consultant - recommending people try to kill them self to see if a bolt is loose is dumb. Turn the thing off and tighten the bolt.

    An IR window won’t tell you if the windings are going bad, and that’s much more critical than the terminations. With the transformer de-energized, do a winding ratio test at the frequency recommended by NFPA70B, while you’re tightening bolts.

    The fucked up part is that maintenance consultant probably gets pedigree by commenting on the NFPA concensus review sessions to advocate for his way. Feel free to PM me if you need help getting that fucker walked off site. I love throwing the code book at people that try to kill maintenance workers.


  • Great question, but nope, 480 is the bomb going off.

    If you’re familiar with Ohm’s law then pick a consistent number for power and calculate the current for both voltages.

    If you don’t know Ohms law, imagine if you wanted to move the same amount of water as a river moves through a hose, you’d find the water from the hose would fuck you up a lot more than the water from the river. Similar principal with current.

    The transformer shown in the picture is likely one of the most dangerous devices in the plant, because it’s taking all that river and shoving it into a hose (turning 13kV into 480v while maintaining the same power). Because of that, the incident energy (explosion size) is at it’s highest at the 480v side of the transformer (current is at its highest).

    The only protection (such as a breaker/fuse) upstream of the 480v side of the transformer is on the 13kV side. Imagine the river was shoved into a hose via a water fall. Imagine you wanted to turn the river off because someone was getting blasted with the hose, you’d have to run up a waterfall, which would lower your response time. During which the person getting blasted by the hose would continue to get blasted with the hose.

    Transformers almost always have a breaker or fuse directly after it in a circuit, so that it can regulate and respond in a faster way to things such as an arc flash. But the transformer is generally where you see the highest incident energy. That is coming directly from the 480v side. But aside from verifying that the transformer is de-energized there is literally no troubleshooting or manual task that would warrant operating on it live.

    Arc flash is not the same as shock. Arc flash is literally an explosion. I’ve been on-site for minor ones that stay contained in the box and just leave smoke and molten metal in the box. And I’ve been there for ones that literally blast the panel front off till it hits a wall - those are the scary ones. Luckily that one no one was nearby it when it blew up. It happened from vibrations moving dust between contacts. The shrapnel would hurt, but even worse is the potential to heat flash the inside of your lungs. That’s why when operating 480v equipment you should take a deep breath first to fill up your lungs, so that you don’t accidentally fill your lungs with hot as fuck air.

    However, I would much rather get SHOCKED by 480v over 13,000 volts. But I’d rather not get shocked, so I wear proper gloves for the voltage I’m working on.


  • The ELI5 of my comment:

    I would rather work on something that’s 13,000 volts as opposed to something that’s 480 volts. Because 13,000 volts is a bad shock if I fuck up. Whereas 480 volts could be a bomb going off in front of me, not because I fucked up, but because it could just spontaneously combust (not entirely true, but I’ve seen plenty of arc flash incidents where it doesn’t seem like anything happened but all of a sudden an electric panel blows up).

    If you’re morbidly curious, emphasis on morbid, feel free to look up some arc flash videos. They’re crazy big explosions when bad. Fucking scary to see the aftermath of, let alone to be apart of.

    Anyways, the NFPA is the governing body of electric. Arc flash wasn’t talked about in the electric code till like the 90s, so its a relatively new discovery, in an already relatively new industry (vs say food prep/food code). There’s a lot that needs to be learned and improved upon.

    Transformers take one voltage and make it a different voltage in the same box. The code doesn’t have a standard way of labelling the arc flash hazards, which means you get stupid things like the original picture.

    If you read the top label it tells you to wear thick (class 2) gloves and as long as you do that you’re safe. The bottom label tells you if you work on the equipment, there is no level of ppe that can protect you. Both labels are technically true for what they’re talking about. But someone without experience might stop after the first label, put on gloves and then get vaporized in a catastrophic explosion. In my comment I pretended that the explosion level was 39kcal (instead of 391) because there is a safe level of PPE that would protect someone from an explosion that big - it’s basically a bomb suit - just gloves wouldn’t keep you safe. The gloves the bottom label tells you to wear wouldn’t keep you from getting shocked though. So you would want to wear the top label’s gloves and the bottom labels bomb suit to be fully safe.

    The code books don’t standardize how to communicate the required PPE which makes people do stupid things like in this picture: show conflicting requirements for safety. Shitty labeling like this can kill someone, but its not necessarily wrong labeling because the code leaves it up to interpretation.