Mounting of IP20/IP67 Remote IO Inside of Tools

cjosey

Member
Location
Fremont, CA, USA
My company designs capital equipment mostly for the semiconductor industry. For our control systems in the past, we have typically wired the PLCs in a ?homerun? configuration (where all cabling goes back to the PLC in the main electrical enclosure).

We are currently moving to a fieldbus configuration with low-voltage distributed IO nodes located throughout the tool. Most control manufacturers sell both the IP20 DIN-rail mounted versions of IO (with ?slices?) and the IP67 wash-down rated versions that require M8 or M12 connectors.

We are having an internal debate about which version of IO to use. All of the nodes will be mounted on the tool framing and located behind removable (but not locked) panels. There is also no risk of wash-down for our tools. Our goal is to select one version for all tools and be able to pass any third party/SEMI/CE safety inspection.

I believe that we can use the IP67 versions without comment by any safety inspector, but they are typically more expensive, larger and less flexible than the IP20 slice IO. The concern about the IP20 IO is that NFPA 79-2012, paragraph 11.2.2.2 states: ?Pipelines, tubing, or devices (e.g., solenoid valves) for handling air, gases, or liquids shall not be located in enclosures or compartments containing electrical control equipment.?

So, is the area behind the removable panels considered an enclosure/compartment or not? Are there any other issues I am not addressing? What do you think?
 

cjosey

Member
Location
Fremont, CA, USA
Smart $,
Thanks for the quick response. The removable panels are for accessing the interior of the tool for maintenance and troubleshooting, and they are usually removed by quarter-turn fasteners. There are things like framing members/aluminum strut (like 80-20 or Bosch), servo motors, linear actuators, pneumatic controls (i.e., DIN rail mounted solenoid manifolds and pressure sensors mounted on flat aluminum plates that are bolted to the frame), robot controllers, 19-inch racks, wiring channels (like Panduit wireways), etc inside.

Carl Josey
 

Smart $

Esteemed Member
Location
Ohio
Sounds like there is more than electrical and electronics equipment behind the covers. It's really quite simple. No other enclosed systems are permitted within the electrical/electronics enclosure.
 

GoldDigger

Moderator
Staff member
Sounds like there is more than electrical and electronics equipment behind the covers. It's really quite simple. No other enclosed systems are permitted within the electrical/electronics enclosure.
So that would mean that an enclosure which surrounds a whole house generator, and therefore inescapably encloses the electronic portions, cannot also contain the fuel system of the generator?
It seems to me that a critical question is whether the enclosure of the electronic components is required by some code or is simply a "convenience" or cosmetic enclosure of parts that otherwise would be allowed to be exposed.
 

Smart $

Esteemed Member
Location
Ohio
So that would mean that an enclosure which surrounds a whole house generator, and therefore inescapably encloses the electronic portions, cannot also contain the fuel system of the generator?
It seems to me that a critical question is whether the enclosure of the electronic components is required by some code or is simply a "convenience" or cosmetic enclosure of parts that otherwise would be allowed to be exposed.
Good questions. The typical answer is, nothing prevents an enclosure within a larger enclosure. Whether cosmetic or purposeful can be debated with no resolve short of changing the standard(s?) to which we are fabricating.
 

gadfly56

Senior Member
Location
New Jersey
Occupation
Professional Engineer, Fire & Life Safety
I used to work for EMCORE Corp, so I know exactly what you're talking about. I'm going to guess you're with Lam Research?

In any event, you raise an interesting question. First, I believe that once you wrap the tool in a cabinet, you have an enclosure.

I have only the 2007 edition, but exception 2 says "Pipelines, tubings, or devices that are part of listed equipment and are separated by suitable barriers". If this is the same in the 2012 version this could be your out, depending on what "barrier" means. Lets face it, if the silane line develops a leak to the exterior at a solenoid valve, you'll have much more exiting things to worry about no matter where the IO is mounted.

With the devices mounted on the frame, are they scattered in onesie, twosies, or clusters of serveral? If you have clusters of 5 or so and drop them in a small box inside the tool footprint that might get you your "suitable barrier". Otherwise, the IP20 might be a tough sell if you're trying to get the largest market penetration. Considering what these tools go for, I doubt material cost is really an issue for the IO.
 

cjosey

Member
Location
Fremont, CA, USA
Good discussion. This parallels our internal dialog, so it is good to see that the answer is not so clear. (BTW, I am not from Lam Research, but I am nearby).

GoldDigger makes an interesting point about the enclosure possibly being just cosmetic. The removable panels/skins really are just aesthetic, since without them the tool would look odd/ugly. Here is a poser: If we removed all of the skins, would the tool no longer have an enclosure? Would the problem just disappear, and we could use IP20 IO with no problem?

Smart $, when you say that no other enclosures are permitted within the electrical/electronics enclosure, do you mean the IP67 IO boxes as well? They have a wash-down ?enclosure,? so do you recommend that we not use them either?

Gadfly56, exception 2 is in the 2012 version of NFPA 79 as well. None of our tools are listed, so I don?t think we can use this exception (which is too bad). We have looked at small plastic enclosures with pre-mounted DIN-rail and many strain-reliefs installed (Murrplastik link: http://www.murrplastik.de/Default.aspx?mid=mp_product_kdh3&cid=en&id=56&action=group). However, this leads to higher complexity, material and man-hour costs.

Over the last couple of years, I have asked this specific question to three different safety inspectors (from TUV SUD, Abstraction Engineering, and ESTEC Solutions). All three have been OK with using the IP20 IO mounted on DIN-rail throughout the tool. I have used this method on over 40 tools that I have designed, and the customers do not seem to care either. Since wash-down is not a concern, and the IO is all low voltage, this issue seems "intuitively? to be OK, but that obviously does not make it OK by code.

Smart $ states that the cosmetic/purposeful issue can be debated with no resolve without changing the standard(s?) to which we are fabricating. That seems like it would be difficult, but does anyone know how to get an official code ruling from the NFPA 79 committee concerning this issue? If I get dinged by a safety inspector or AHJ, is it possible to get a ?waiver? if the customer is OK with it or is there some other recourse? It seems like when I really examine this issue, I am heading down the rabbit hole?
 

Smart $

Esteemed Member
Location
Ohio
Smart $, when you say that no other enclosures are permitted within the electrical/electronics enclosure, do you mean the IP67 IO boxes as well? They have a wash-down ?enclosure,? so do you recommend that we not use them either?
I didn't say no other enclosures... I said no other enclosed systems, such as pneumatic, hydraulic, or other process and control lines (pipes, tubes, hoses, etc.).

[Electrical/electronic] enclosure within [tool] enclosure is what I suggested in reply to GoldDigger.

As with most electrically operated and/or controlled machinery, you typically have a termination and/or controls compartment(s) [somewhat] isolated from the rest of the machinery. For example, motors and sensors have 'peckerheads'; extruders have a power and control cabinet.
 

gadfly56

Senior Member
Location
New Jersey
Occupation
Professional Engineer, Fire & Life Safety
Good discussion. This parallels our internal dialog, so it is good to see that the answer is not so clear. (BTW, I am not from Lam Research, but I am nearby).

GoldDigger makes an interesting point about the enclosure possibly being just cosmetic. The removable panels/skins really are just aesthetic, since without them the tool would look odd/ugly. Here is a poser: If we removed all of the skins, would the tool no longer have an enclosure? Would the problem just disappear, and we could use IP20 IO with no problem?

Smart $, when you say that no other enclosures are permitted within the electrical/electronics enclosure, do you mean the IP67 IO boxes as well? They have a wash-down ?enclosure,? so do you recommend that we not use them either?

Gadfly56, exception 2 is in the 2012 version of NFPA 79 as well. None of our tools are listed, so I don?t think we can use this exception (which is too bad). We have looked at small plastic enclosures with pre-mounted DIN-rail and many strain-reliefs installed (Murrplastik link: http://www.murrplastik.de/Default.aspx?mid=mp_product_kdh3&cid=en&id=56&action=group). However, this leads to higher complexity, material and man-hour costs.

Over the last couple of years, I have asked this specific question to three different safety inspectors (from TUV SUD, Abstraction Engineering, and ESTEC Solutions). All three have been OK with using the IP20 IO mounted on DIN-rail throughout the tool. I have used this method on over 40 tools that I have designed, and the customers do not seem to care either. Since wash-down is not a concern, and the IO is all low voltage, this issue seems "intuitively? to be OK, but that obviously does not make it OK by code.

Smart $ states that the cosmetic/purposeful issue can be debated with no resolve without changing the standard(s?) to which we are fabricating. That seems like it would be difficult, but does anyone know how to get an official code ruling from the NFPA 79 committee concerning this issue? If I get dinged by a safety inspector or AHJ, is it possible to get a ?waiver? if the customer is OK with it or is there some other recourse? It seems like when I really examine this issue, I am heading down the rabbit hole?
If you are a memeber of NFPA you can submit a Technical Question and get an informal ruling from a staff member. Your best bet may be to craft a proposal for inclusion in the next edition of NFPA 79. Every tool I've ever seen has been pretty much as you describe; a bunch of wire, pipe, and tubing that would look extremely ugly without a "skin". The current edition is 2015, so you can certainly get in on the ROP's for 2018. I looked at the committee membership, and there's no one from SEMI on the committee, nor anyone individually from a semiconductor tool maufacturer that I could see. As they say "You gotta be in it, to win it."

My personal opinion, and worth every penny you're paying for it, is that the arrangement you are proposing should be allowed under these circumstances. I'd guess that the original scope of the section had in mind machinery running hydraulic fluid, natural gas, coolant lines and whatnot where flows were in gallons per minute, not cc per minute, and "enclosure" meant something like a Hoffman box stuffed with relays, and they didn't want a fluid line taking a shortcut through the box.
 

cjosey

Member
Location
Fremont, CA, USA
Smart $, I definitely misunderstood, and I understand what you are saying now.

By the way, I did not lead with the responses from the third-party safety inspectors, since I did not want to bias anyone. I wanted to get opinions on both sides of the issue, and I definitely accomplished that.

Gadfly56, thanks for the info on the informal rulings for Technical Questions. I am not a member of NFPA, but I will have to consult with my management to see if we want to join.

Thanks for everyone's input; I think I have everything I need now.
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
it is not like nfpa79 is a mandatory std anyway, like the nec is.

owners and manufacturers are free to accept it as is, reject it totally, or use some part of it.

there is no reason that you could not add some kind of addendum to whatever it is where you are claiming compliance with nfpa79 stating what you are planning to do that may not b strictly in compliance.
 
Top