I'm writing an article regarding Article 645 EPO's. I would like to include a picture of a poorly installed EPO, or an EPO that is blocked by something, or some other orientation showing that EPO's are not always accessible. Any help would be great.
Bob, I often have questions about EPO's. Why does the client want them!?
Many don't take advantage of the leniencies available by complying with 645. If they knew that they don't need the terrible liabilities associated with that optional code article, would they really NEED the leniencies, probably not! Save the headache, don't do it. That is a summary of the article.
Re: EPO (Emergency Power Off)#57121 10/05/0503:59 PM10/05/0503:59 PM
The building I referenced above is a great example of what you describe.
The rooms do not even remotely comply with 645 but when they installed large APC UPS in each of these rooms someone sold them on 'Genuine' APC EPO buttons, they look real expensive for what could be done (if needed) with a single SP switch.
This place has about 1,500 people answering phones, if a broom stick falls and hits the EPO many of these folks will be on an extended break.
Bob Badger Construction & Maintenance Electrician Massachusetts
Re: EPO (Emergency Power Off)#57122 10/05/0509:41 PM10/05/0509:41 PM
IBM pretty much invented the computer EPO. It was originally intended to drop the power if someone was being eaten by a card reader or check sorter. As systems became humming boxes that you don't open and voltages were 3vdc or less the EPO stopped getting connected. Finally in the late 80s they dissapeared from consoles. DP managers still liked the idea so they started being a shunt trip for the whole room. One way to losr the allure is to actually have someone trip the EPO. Then they end up being in a "break glass" box, next to the Halon trip. NOBODY really wants that extended break. It might easily cost a company $100,000 or more to power off their servers, even if it only takes 15 or 20 minutes to Re-IPL the systems. If you corrupt some data it really starts getting expensive.
You certainly want that switch to be unique and hard to get to by accident. I saw a snap switch, used as an EPO. One guy mistaking it for a light switch fixed that. The customer had us disable all the system EPOs and he removed the shunt trip system.
Re: EPO (Emergency Power Off)#57123 10/05/0510:05 PM10/05/0510:05 PM
The EPO's I have done are more for Fire and Smoke rather than just the UPS. Often are tied into seperate smoke dampers and duct detectors and seperate priority on the building fire system, so a fire drill wont drop the system. I take a circuit from the UPS to power a relay that when dropped out will open the EPO contacts for the UPS, and close the dampers, the a trouble signal to the fire system. Heat det.s DD's and smokes are seperate in the room for a full alarm. For the switch I use a 22mm guarded at the door, with a momentary key reset for the relay. (Much more expensive than a single ploe switch!)
Anyway, the way I understand it.... It's all completely unnessesary unless there is air/smoke interconnection to the rest of the building. The room is supposed to be fire-rate either way, and have smoke detection in the very least. But if the batteries are vented, or if the room ties to any outside HVAC, then the EPO for the batteries, and HVAC is required by the Fire Inspector more than the Electrical Inspector. And, here at least, it is the Fire Inspector that will ask for the operation of the EPO, I guess the had a battery fire a while back, and this is now a pet peev.
Anyway, here some commentary, (Not code) on it...
In 645.10, two separate disconnecting means are required, but a single control, such as one pushbutton, is permitted to electrically operate both disconnecting means. The disconnecting means is required to disconnect the conductors of each circuit from their supply source and close all required fire/smoke dampers. (See the definition of disconnecting means in Article 100.) The disconnecting means is permitted to be remote-controlled switching devices, such as relays, with pushbutton stations at the principal exit doors. The 2002 Code specifies that the actuation of the emergency pushbutton(s) be accomplished by pushing the button in, rather than pulling it out. The requirement recognizes that in an emergency situation the intuitive reaction to operating the control is to push, not pull, the button. The requirements of 645.10 and those of 645.7 for sealing penetrations are intended to minimize the passage of smoke or fire to other parts of the building.
Mark Heller "Well - I oughta....." -Jackie Gleason
Re: EPO (Emergency Power Off)#57124 10/06/0505:04 AM10/06/0505:04 AM
NFPA 75 (If it applies in your area) notes exactly the same thing in section 10.4.7. In annex A it gives a bit more info in A.8.1.2 about method of shutting down power, and in A.11.1 it states a written emergency plan be prepared and posted.
If you guys are not pushing the installation of EPO's I'd like to suggest you reconsider, if not for simple code compliance then for basic life safety.
UPS fed equipment can make your life very uncomfortable in an emergency, let's give 'em a way to cut it down on the way out the door, most people will do it.
Re: EPO (Emergency Power Off)#57125 10/06/0510:20 AM10/06/0510:20 AM
Because, Article 645 in the NEC is opional. There is no requirement that it be used for an "Information Technology Equipment Room". The rules are permissive, not required.
645.2 This article shall apply, provided all the following conditions are met: ...
If you use it, there are some things that can be done in a manner that would otherwise be a code violation, but there is no requirement that the article be used. I don't have a copy of NFPA 75 so I can't comment on rules in that document. Don
Re: EPO (Emergency Power Off)#57126 10/06/0512:33 PM10/06/0512:33 PM