Cable Running from Class I Div 2 all the way to a Nonhazardous Location

[EricPeterson]

OK, looks like I might have found somewhere with some resident experts!

At work we're conducting an ongoing project to transition all of our oilfield equipment wiring products to CSA certification. These are engine management, compressor, control system, and sensor harnesses mounted to everything from small pumps to heavy lift palletized gas compressor arrays. We've been successful clearing certification for harnesses residing entirely within the Class 1 Div 2 volume but are now faced with a new challenge. One of our cables originating within the hazardous zone passes completely outside of the area and needs to break out to several different arrays of sensors, none of which are designed for hazardous environments themselves and all of which are fitted with automotive style pigtails with automotive style plastic single-snap connectors.

The origination inside the hazardous zone is a modified MS 5015 style threaded, hermetically sealed connector/backshell assembly. The cable itself is TC-ER with two extra layers of protection, the first an additional layer of PVC insulation, and the second a braided abrasion resistant oversheath. This style of assembly satisfies all the conductor separation and insulation requirements for certification and is working well through CSA and for our customers when located completely inside the hazardous area.

Some, but not all, of the circuits in this new single TC-ER cable carry incendive signals.

Now that we need to connect to sensor arrays outside the hazardous area, what do we do? How do we transition from the cable for the hazardous area to the pigtails to the automotive style connectors?

Opinions have included:

(1) Not possible. Once incendive protection required, always incendive protection required, regardless of the environment. The sensors outside the zone need to be upgraded to Class 1 Div II units, and if they're not available, existing units need to be modified, rewired, and separately qualifed/certified.

(2) The style and construction of the sensor cable needs to remain fully certifiable, using appropriate materials and connectors, and at the other end, the sensors can be modified to wire into mating connectors.

(3) The style and construction of the sensor cable needs to remain fully certifiable, using appropriate materials and connectors, and at the other end, needs to plug into a sealed junction box entirely outside the hazardous area. The junction box presents breakouts for cables running to the sensors, which remain stock.

(4) The TC-ER cables need to run all the way to the automotive style connector mates maintaining all standards. Some sort of effort needs to be made to seal the back end of the automotive style connectors to the TC-ER cables. Some sort of "needs a tool" protective device needs to be applied to each connector assembly at each sensor pigtail. The sensor pigtails don't need to be modified.

(5) Full protection of the TC-ER cable and oversheaths need only run some prudent distance outside the Class I Div 2 zone. Once safely outside the hazardous environment, the outer layer of the TC-ER cable insulation can be removed and the individual circuits and twisted sets can be broken out and routed to mating connectors for the sensor pigtails with industrial-automotive level wear protection. No "needs a tool" locking device is required for the connector pairs, which will unsnap from one another with a thumb press on a plastic tab.

I haven't identified a section that specifically addresses this case. Where do I look or what's correct?

 

[rbalex]

Are you using the NEC or the Canadian Electrical Code (CEC)? The installation answer is the same, but the code references are different.

 

[EricPeterson]

NEC, thanks.

 

[rbalex]

Sealing Type TC-ER (or any Type TC) is covered by Section 501.15(E)(3) and its cross-reference to Section 501.15(E)(1).

 

[EricPeterson]

Thank you. Yes, I'm reading and re-reading and re-reading 501.15 (B) through (E). I just don't see a clear answer, and neither does my industry nor my customers. And apparently CSA isn't totally clear on it, either.

What I'm talking about is the specific case where a cable originates in a classified zone outdoors and runs a substantial distance to an unclassified zone still outdoors. There are no walls, or boundaries, or cabinets between. No seals. No pass-throughs. The industry feels that the vicinity of the engine should be a classified volume because of the fuel, ignition sources, and fossil fuel product. The customer demands the CSA cert for the components inside the zone. But now we've got this long cable running from the zone -- compliant connector, extra sheathing, hermetic seal, TC-ER core cable -- all the way to an area the customer doesn't believe is or even might possibly be classified. We can't get a straight answer from them as to their reasoning, and we can't get a straight answer from CSA, who says simply "send us a sample and the specs and we'll think about it." The area into which the cable runs is far enough from the well or pipeline pumping station that the concentration of gas flammables is insignificant.

We don't want to trail and error it. I want to do it right. And I want to be able to point a spec that describes terminations of cables outside the classified zone. Or I want an approach that simply can't be challenged.

The NEC articles are always explicit that seals and components installed or used within the Class I Div 2 locations shall be this and shall be that. We get that. Gas tight. No spaces between conductors. continuous outer sheathing. No issue. My problem is that one end of the cable won't be in the Class I Div 2 classified zone, and we're getting strong vibes from the certifying agency that they don't care and want it all compliant, and strong vibes from the industry that they don't want to pay for that. 501.15(B) through (D) discuss boundaries. (B)(2) addresses an exception for conduits passing through, in to, or out of a classified zone and says neither end need be sealed if both are outside the classified location and the conduit is unbroken, and provides for other exceptions, including outdoor locations. (C) addresses "seals" for Class I Div 2 but doesn't seem to be discussing cable connector seals. (D) discusses cable seals but for Class I Div 1. (E)(1) discusses cable transitions between zones in Class I Div 2, but generally in terms of definite boundaries and enclosures. "Shall be sealed at the point of entrance" means... what? If there is no clear point of entrance, does it mean the end deep inside has to be sealed, or the end far outside does too, or that the continuous and unbroken gas-tight sheath passing from deep within to some point outside, where there is no seal into the interior of the cable, satisfies?

The larger issue here is intent of the specifications and design goals directed towards hazard mitigation. It's clear reading in detail through 501.15 that the intent of the specs centered around cables and the sheathing and fittings is twofold: (a) first, physical protection of the incendive conductors and their physical isolation from the flammable gas environment and (b) gas management along and through the cable, so as to avoid transmission of gasses along the cable into an otherwise unclassified zone.

Let me ask this: Let's say that I build a completely certifiable armored straight-through cable with compliant connectors at both ends that routes from within the Class I Div 2 classified zone to well outside the zone. That cable operates within the classified zone and will bear a CSA cert label and the design will have a file number. It provides both the physical protection required and the gas management.

Now let's say I have a non-certified cable that connects to the second end of that certified cable and provides the breakout to the automotive sensors distributed over some equipment, all entirely outside the classified zone.

Does that system present a compliant solution?

 

[bwat]

I'm a little lost. Doesn't rbalex's post in #4 answer what you are to do? Sealing per Section 501.15(E)(1) applies to you, correct? It doesn't matter that the other end of the cable is in unclassified location.

Multiconductor or optical multifiber cables
with a gas/vaportight continuous sheath capable of transmitting
gases or vapors through the cable core that are installed in
a Division 2 location shall be sealed with a listed fitting after the
jacket and any other coverings have been removed, so that the
sealing compound can surround each individual insulated
conductor or optical fiber tube in such a manner as to minimize
the passage of gases and vapors.

 

[EricPeterson]

All the paragraphs that present details like that are always subordinate to a clause that contains "In Class I Div 2 locations..." Thus, it's not expressly clear that those terms apply to a cable end outside the zone.

That having been said, I tend to agree. It's a good, safe interpretation irrespective of the grammatical legalistic interpretations possible. I can build a cable that's hermetically sealed with qualified connectors at both ends, hand it to CSA, and get a file number. No problem.

So the question becomes what I posed at the end of my long post: if I connect that cert cable to a non-cert cable at the end totally outside the classified zone, is the system still qualifiable?

 

[bwat]

All the paragraphs that present details like that are always subordinate to a clause that contains "In Class I Div 2 locations..." Thus, it's not expressly clear that those terms apply to a cable end outside the zone.

I interpret it to be clear, but I see it different than what it seems like you are saying. It only talks about what you need to do in the classified end because you don't have to seal it at the unclassified end.

If you think about the reasoning behind why you need such a seal, then it might make more sense. I'll try to fumble my way through explaining why I think it says what I'm seeing:

• The whole purpose is keeping flammable/ignitable/whatever gases from causing a problem.
• One of the areas that could be a big problem is if the gases from the classified area could enter into the cabinet in the unclassified area, where the cabinet is likely non-XP and has unclassified spark-inducing components. This transmission of gases could happen through the cable.
• If gases could enter the cable anywhere along the length of the cable, then the cable needs to be sealed at the boundary from classified to unclassified, because gases could enter into the cable anywhere along the cable before it gets to the boundary. See Section 501.15(E)(4).
• If the gases could only get into the cable at the ends, but then travel through the cable (your situation) to unclassified location, then you only need to seal at the classified location termination, because once you leave that termination, there's no way for gases to get into the cable. See 501.15(E)(3) which takes you back to 501.15(E)(1)

One other thing to keep in mind. The NEC is a prohibitive code. Meaning, unless it tells you otherwise, you are allowed to do it. Unless you see something that says you must have a seal at the unclassified location, then you don't have to have a seal at the unclassified location.

 

[rbalex]

If the cable is listed as a Type TC, it doesn't have to be sealed in Class I, Division 2 unless required in Section 501.15(E)(1). If it is required by Section 501.15(E)(1), it still isn't required to be sealed at the boundary or the unclassified terminus. (There must be a boundary somewhere if the run begins in Division 2 and ends in an unclassified location). If the method of connection/termination in the unclassified location were otherwise "legal," it would still be, without regard to its Division 2 origin.

 

[EricPeterson]

You're not fumbling. You raised the same issue of intent I did -- keep the gasses from moving along the cable. I see both Thou Shalts and Thou Shalt Nots, but I understand your point -- as long as you comply with the directives, an unprohibited solution that satisfies the letter of the regs is allowed.

The thing is we're seeing some relatively strict interpretations, and the process is so costly for a small business that I won't want to try to afford multiple runs through certification.

It turns out I have an added layer of complexity. My customer is going to take the entire pallet to CSA for a system-level cert. Thus, if there's any room for interpretation in the subassemblies, I'd better not exercise that latitude as I might see it, so at this point it's pretty clear that any cable with either end in the classified zone had better darn well be fully compliant. When I hand the sample over to CSA, they're really not going to care which end goes into the classified zone and which end doesn't.

Thanks!

 

[EricPeterson]

If the cable is listed as a Type TC, it doesn't have to be sealed in Class I, Division 2 unless required in Section 501.15(E)(1). If it is required by Section 501.15(E)(1), it still isn't required to be sealed at the boundary or the unclassified terminus. (There must be a boundary somewhere if the run begins in Division 2 and ends in an unclassified location). If the method of connection/termination in the unclassified location were otherwise "legal," it would still be, without regard to its Division 2 origin.

All the cables I'm using are listed type TC with file numbers. Because the cable type TC is capable of transmitting gases within the continuous vapor tight sheathing, all connectors within the Class I Div 2 locations must be sealed to the cables so as to minimize transport of gas per (E)(1). This we've been told pointedly and it's not open for discussion. Unfortunately we have no clearly defined boundary between classified and unclassified zones. These pallets can be 60' long and all the fuel and gas is at one end. The other end has insignificant detectable flammable gas. There are no cabinets or panels per se, just components mounted to the engine, compressors, and sensors.

I think the only way subassemblies are going to fly is if anything that touches inside the classified zone, such as it is, is fully qualified. Looks like feedback the customer is getting is directing us towards my original option (3).

Thanks!

 

[rbalex]

...Because the cable type TC is capable of transmitting gases within the continuous vapor tight sheathing, all connectors within the Class I Div 2 locations must be sealed to the cables so as to minimize transport of gas per (E)(1)...

Is the enclosure required to be explosion-proof or is the cable end subject to a pressure in excess of 1500 Pascals? (That hasn't been clear to me) If not, the Type TC isn't required to be sealed anywhere. If it is, the Division 2 cable end is the only location that is required to be sealed.

... Unfortunately we have no clearly defined boundary between classified and unclassified zones. ...

I understand that may be the nature of skid or pallet installations. But there will be a boundary somewhere - there must be AND it must be documented (Section 500.4)

 

[EricPeterson]

Is the enclosure required to be explosion-proof or is the cable end subject to a pressure in excess of 1500 Pascals? (That hasn't been clear to me) If not, the Type TC isn't required to be sealed anywhere. If it is, the Division 2 cable end is the only location that is required to be sealed.
I understand that may be the nature of skid or pallet installations. But there will be a boundary somewhere - there must be AND it must be documented (Section 500.4)

Not normally. However, I was giving that some thought, and I believe it's possible that branches of the harness routed to sealed gas regulator units -- basically, the carburetors -- or to engine sensors, or to compressor sensors, could experience pressurization on failure of those components. It would only be prudent to allow for it.

Thanks for the point to 500.4. I'm going to bring that up with my counterpart at our parent firm immediately.

 

[rbalex]

Not normally. However, I was giving that some thought, and I believe it's possible that branches of the harness routed to sealed gas regulator units -- basically, the carburetors -- or to engine sensors, or to compressor sensors, could experience pressurization on failure of those components. It would only be prudent to allow for it.

Thanks for the point to 500.4. I'm going to bring that up with my counterpart at our parent firm immediately.

Now you're getting into philosophy. "Prudent" may be true, but when we are talking about Division 2 it usually isn't a concern unless "failures" are expected to be fairly common. (See Section 500.5(B)(2) - read the whole Section carefully and give attention to the Informational Notes as well. Often here the problem is "normal" is undefined.)

 

[EricPeterson]

Now you're getting into philosophy. "Prudent" may be true, but when we are talking about Division 2 it usually isn't a concern unless "failures" are expected to be fairly common. (See Section 500.5(B)(2) - read the whole Section carefully and give attention to the Informational Notes as well. Often here the problem is "normal" is undefined.)

Thank you. I will. Failures aren't common at all. However, that decision was made for me, and mine is not to question why; mine is but to do and die. Besides which, I have other battles on other fronts, and if management or a customer demands something that leans towards even marginally enhanced safety and conservatism -- and offers to pick up the tab -- then I'll prioritize my efforts elsewhere!

 

[mooreaaryan]

Lots of good information on post