Need NFPA Help

[CEDEng]

This is an awesome forum for many things, but doesn't really have the "chapter" I'm looking for - in fact, can't find it anywhere!

I need help/direction/clarification/interpretation with NFPA 79, Electrical Standard for Industrial Machinery.

If you know this standard very well, or if you know of a forum or FAQ place, or you know who to ask or call, please share it with me! While many of the "shalls" are crystal clear, there is no direction
on how to implement them - and while that's common in these documents some of the "shalls" are downright impossible without the magical mystery hardware I have not yet invented.

For instance, if the standard said, All Machines Shall Hover, that's a nice touch, but how to do this is unexplained. (OK, that's a BIT of a stretch, but I'm trying to make a point..."

Thanks, all, for any input.

 

[Smart $]

While the forum(s) here are not particularly geared for NFPA 79, there are a few members familiar with it. Why don't you post the specifics of your plight and see what happens... :happyyes:

Aside from that, perhaps https://community.nfpa.org/community/explore/content

 

[petersonra]

the NEC does not have all that much on implementation either.

if you post your questions, you will no doubt get some pretty good answers.

I have always found most of it pretty straightforward.

 

[CEDEng]

Thanks, Smart $

Yeah, I know this isn't really an NFPA thing - but it does have an NEC thread, which is as close as I've found.

Thanks for the link, though, excellent so far.

Brief background - we've been building machines here since the dark ages, before standards were invented - and obviously we'd like to move towards satisfying them before someone catches on that we do NOT follow them.

That sounds devious, but it's really not - most custom one-off machine makers have trouble following these standards - there's nothing "standard" about custom work. But increasing scrutiny and regulation, as well as emphasis on safety both here and worldwide, is driving this shift towards "following the standards."

Thing is, some changes are difficult or extremely expensive.

Here's an example of one that is easy to understand and impossible to implement: No greater than 50% "fill" of wire in duct work. Our ductwork is, conservatively, 110% full. Boxes are a finite size, customers want a machine to fit in a tiny space, there are a thousand parts in there - short of tripling the enclosure size, this requirement is impossible to meet. But we will try...

To the point of one that's NOT very clear: 6.2.4, enclosures must be interlocked. To what? How? What does "power removed" mean? What if the switch is IN the machine, and so power will obviously be IN the box TO the switch - does that count? Every thing we do here - including the enclosures - is custom. An "off the shelf" box with a interlock will not work for us. In addition, those off the shelf boxes have one or sometimes two doors. I sometimes have 10. How am I to interlock all those doors?

So on, so forth, I have dozens more questions - but I can't even find a training class on this material. So this forum is one of my last resorts!

Thanks again.

 

[dkidd]

I found this article with a list of various engineering forums. Some of them look possible.

http://machinedesign.com/technologies/best-web-best-engineering-forums

 

[Smart $]

...
To the point of one that's NOT very clear: 6.2.4, enclosures must be interlocked. To what? How? What does "power removed" mean? What if the switch is IN the machine, and so power will obviously be IN the box TO the switch - does that count? Every thing we do here - including the enclosures - is custom. An "off the shelf" box with a interlock will not work for us. In addition, those off the shelf boxes have one or sometimes two doors. I sometimes have 10. How am I to interlock all those doors?
...

Shunt trip main power breaker with external operator mechanically interlocked to door and with defeat mechanism. The shunt trip is for interlocking to other enclosure doors electrically. The defeat mechanism is to open enclosure door for energized testing purposes only.

 

[just the cowboy]

Sorry to say this.

Sorry to say this.

The suggestions are for a good reason. Your example of wire fill is a very good reason on why you should be implementing the code. For many of us that repair machines on a daily basis, that is one of the things that are a pain, too much in a small area or too many wires in a wireway that they rub and short out and are hard to find because so much is in there.

Go to customers plants and look at other manfactures equipment for ideas.

 

[CEDEng]

Good ideas, all, thanks.

Re: other machines - done this many times - in fact, I've worked at at least four different places, all of them do it the same - jam it full. Sure, some folks don't - these are the machines that occupy several hundred square feet, so two or three or 23 panels is no big deal. When you compress the machine to Mini-Van size - it's difficult. I'm not saying I disagree with the standard in that regard, or that it's not a maintenance nightmare (I DO maintenance), I'm just saying that's something that was easy to write, and impossible to easily implement.

Re: the Shunt Trip - been there. The standard says: shall be interlocked such that none of the doors open unless power is disconnected. Now, clever me interprets that to mean that the very act of opening the door can CAUSE the power to be removed, and, well, there ya go - open door, power is removed, problem solved.

The more stringent (and practically universal) translation of that paragraph is that the power Must Be Removed BEFORE you are ABLE to open the door. See the distinction?

With the shunt trip case, a single malfunction (namely, the solenoid) means the door can open with power applied. The standard is intended to force you to power down BEFORE you even TRY to open the door.

Now, if you're going to argue interpretation with me - I mean that kindly - that's exactly why I am needing help with this standard. Just because "I" think it means one thing does not make it so. I need solid concrete answers. Otherwise, I'm just doing what seems right - which is what I've always done, before The Standards Police came along.

 

[Smart $]

Good ideas, all, thanks.

Re: other machines - done this many times - in fact, I've worked at at least four different places, all of them do it the same - jam it full. Sure, some folks don't - these are the machines that occupy several hundred square feet, so two or three or 23 panels is no big deal. When you compress the machine to Mini-Van size - it's difficult. I'm not saying I disagree with the standard in that regard, or that it's not a maintenance nightmare (I DO maintenance), I'm just saying that's something that was easy to write, and impossible to easily implement.

Re: the Shunt Trip - been there. The standard says: shall be interlocked such that none of the doors open unless power is disconnected. Now, clever me interprets that to mean that the very act of opening the door can CAUSE the power to be removed, and, well, there ya go - open door, power is removed, problem solved.

The more stringent (and practically universal) translation of that paragraph is that the power Must Be Removed BEFORE you are ABLE to open the door. See the distinction?

With the shunt trip case, a single malfunction (namely, the solenoid) means the door can open with power applied. The standard is intended to force you to power down BEFORE you even TRY to open the door.

Now, if you're going to argue interpretation with me - I mean that kindly - that's exactly why I am needing help with this standard. Just because "I" think it means one thing does not make it so. I need solid concrete answers. Otherwise, I'm just doing what seems right - which is what I've always done, before The Standards Police came along.

Whether the shunt trip opens the main before you open the door is dependent on mechanically linking the switch with the door lock mechanism. With switch operator keyed to the door lock mechanism, the shunt trip will engage during the act of unlocking the door, which is before you can open said door.

 

[CEDEng]

Hmmm...I like your thinking on this one!

The next obstacle I run into with that set-up is the shunt trip itself. It requires energy to operate, if a wire falls off it doesn't work, etc., etc, - now, this is not an ESTOP device, it's instead just to satisfy the door-interlocking paragraph - so does that mean it doesn't need to be control reliable? Redundant? Positive action? I feel like I am now crossing not one but two lines of interpretation...

First...I've allowed that the door opening will CAUSE the power to disappear (as opposed to the existence of the power prevents the door from opening), and then I've went ahead and assumed that since this is not a Safety Function and it's not an ESTOP (what is it, exactly?), then now I can execute said technology with whatever means I have handy - meaning, a door switch that closes (instead of opens) a circuit, which supplies power to a part somewhere else that hopefully does its task and turns off the power.

I'm not saying it's wrong. I like it. I've used it. I'm saying I don't know - for sure - that this satisfies the standard. Here's a scenario I have also seen:

Maintenance man turns door lock, with key. Pulls door open. Shunt trip, being ancient, does not work. Now the door is open, and power is on. If he were to touch that panel - will I be able to say at the deposition that I followed the standard?

I know we're beating this one up pretty good - but that's OK - it's just an example - a figurehead - for all the rest, so I'm interested in everyone's input.

Thanks again.

 

[petersonra]

Here's an example of one that is easy to understand and impossible to implement: No greater than 50% "fill" of wire in duct work. Our ductwork is, conservatively, 110% full. Boxes are a finite size, customers want a machine to fit in a tiny space, there are a thousand parts in there - short of tripling the enclosure size, this requirement is impossible to meet. But we will try...

It is not impossible. I do it all the time. it sometimes requires some rethinking of how things are laid out and designed.

In any case, you would be surprised how full the wire ducts look when they are 50% full. Do the calculations.

I have run into a fair number of machine builders that are using #12 wire for a lot of things that could be run in smaller sizes. That can make a big difference.

Just moving things around a little bit can reduce the amount of wire runs.

Take a look at the panelmax wireduct from panduit. you can install it on the panel and put the din rail on top of the duct.

 

[Smart $]

...
Maintenance man turns door lock, with key. Pulls door open. Shunt trip, being ancient, does not work. Now the door is open, and power is on. If he were to touch that panel - will I be able to say at the deposition that I followed the standard?
...

First I'm going to say this is far from a field I'm experienced... well take that back and let's just say I'm not well versed any more. So take what I say with a grain of salt... but if you are this concerned, come up with a solenoid-operated door lock. As long as there is power, the lock prevents anyone from opening the door. However, a lot of this type equipment needs to provide energized testing means... so now you have to build in a means to either defeat the solenoid, or a means to re-energize the cabinet after its door is opened. That may disturb the scenario under which testing is required. You can go back and forth on this issues with no comprehensive resolution.

Ultimately you realize, comply with the Standard as a minimum. Signage can go a long way. Anyone entering a electrical enclosure is supposed to verify it is deenergized... so it will not be your fault if you complied with the Standard when you built it and it failed at some future point in time.

 

[CEDEng]

Good ideas, thanks.

I like the solenoid door lock - I actually was already kicking this around - but I'm not sure how to make it UNeasily defeatable. Seems to me you just open the door...then turn the machine back on.

Re: 50% wire...so it's 50% COPPER, with regards to cross section, vs. duct, and the insulation takes up a fair amount...is the math and the visual something like that?

Yes, on the rest, would require a better design/layout strategy - which is where the cost comes in. As it is now, the panel size is the panel size, and circuit is the circuit, and the two don't meet until they're at the panel shop.

Bad plan, yes, but it's the way it is. So changing it will not be easy, old dogs, new tricks, etc., etc.

 

[Smart $]

...
I like the solenoid door lock - I actually was already kicking this around - but I'm not sure how to make it UNeasily defeatable. Seems to me you just open the door...then turn the machine back on.
...

That's one way... but as I said, you may "lose" the troubleshooting condition.

Re: 50% wire...so it's 50% COPPER, with regards to cross section, vs. duct, and the insulation takes up a fair amount...is the math and the visual something like that?

When it comes to fill, you have to include insulation. Visually, 50% is pretty full because seldom are conductors installed truly parallel is cabinet duct.

If it's installed looking like a rat's nest, I have to wonder if that even exceeds 50%. :blink::blink::blink:

 

[CEDEng]

Well, yep, it's always installed looking like a rat's nest...is there any other way?



Thanks again, all, for the great advice.

 

[Ingenieur]

Use a uv breaker
any wire breaks it drops out

I don't think codes are intended to be absolutely fail safe
just a basic set of engineering controls
you can have signs
procedures, ie locked enclosures that require 2 prople to open, doubling your chance of disconnecting power
training

build a better mouse trap and along comes a smarter (dumber?) mouse

if others build code compliant panels so can you
I can't believe your clients and insurers don't demand it

 

[kwired]

NEC requires 40% fill on raceways with over 2 conductors, some runs that do meet 40% seem like they are 125% filled