Pipe Racks vs. Atmospheric Ice

[CrabbyT]

I'm in Midwest USA, this project is about 50 miles away from the leeward side of a great lake, and I'm analyzing some existing pipe racks (a.k.a. box trusses). The old drawings show that it was designed per the UBC, so ice wasn't a consideration.

Since these were engineered back in the day, they designed the trusses on the sides to carry the vertical loads, and they used the X-braces as tension only members to take up the wind loads. They didn't consider interaction between wind and gravity, so that's already kind of an issue.

In the image below, the chords are WT 10x15.3's. I'm having a hard time wrapping my head around the idea that this big truss would be able to accumulate as much ice as what ASCE 7-10 prescribes. If I were directly on the big lake, or if the members weren't so large, I'd have an easier time accepting it. But I've never personally witnessed anything like this in my area.

Would y'all consider this to be an ice-sensitive structure?? I understand that it's a judgement call and that I take responsibility for the analysis, I'm just curious what your opinions are.

 

[WinelandV]

I would, yes. If you were starting from scratch, why wouldn't you think that ice would be an applicable load?

 

[phamENG]

But I've never personally witnessed anything like this in my area.

I'm not in a cold climate, but I am on the edge of the "Hurricane Prone Region" area of the wind maps. Grew up on the Gulf Coast of Florida and moved to the Mid-Atlantic when I left home. Never have I ever experienced a Hurricane. The edge of a tropical storm brushed by us once, but the winds where I was never got up to tropical levels. But that doesn't mean I don't design every building for the 120mph gusts that I'm required to. (Or the 160 or 170 or so if I were to do one in my home town.)

I'm not familiar with the MRI for for atmospheric icing, but my guess is it's a good deal longer than your memory for seeing (and understanding the importance of) icing on an open, trussed structure. (Not trying to discredit you...just don't overestimate the importance of your personal observations over the collective knowledge and judgement of the larger engineering community.)

 

[CrabbyT]

@winelandv

I'm questioning it because the definition of "ice sensitive structure" is pretty vague, IMO. The code also lists several examples, but most of the provided examples would consist of small-ish members or cables (which are relatively small and seem more prone to ice accumulation). They mention cable stayed bridges and light suspension bridges, but those typically exist over bodies of water where ice accumulation is more likely.

Also, if considering this structure to be ice-sensitive is overly conservative, I'm a little hesitant to tell the client that they need to budget $50 - $100M to reinforce their pipe racks.

ICE-SENSITIVE STRUCTURE.
A structure for which the effect of an atmospheric ice load governs the design of a structure or portion thereof. This includes, but is not limited to, lattice structures, guyed masts, overhead lines, light suspension and cable-stayed bridges, aerial cable systems (e.g., for ski lifts or logging operations), amusement rides, open catwalks and platforms, flagpoles and signs.

 

[straub46]

That pipe rack looks a lot like a lattice structure to me! I would worry about icing. I'm working on designing exterior trusses for a pipe arbor, and we're asking about the ice loads for those for the same reason.

Go Bucks!

 

[CrabbyT]

@straub46

I agree that it looks like a lattice! And I definitely have included ice loads in my model. I agree that the right thing to do is to include ice.

Philosophically though, from a code standpoint, it's tricky. This is an existing structure and I know the original design loads. Adding a single 2" pipe wouldn't increase the stress beyond 5% for what the truss was originally designed for, and the total dead load on the truss is less than what it was originally designed for. However, if I take ice into account, it's likely that the truss will exceed its original design load. At the same time, ice is not a dead load, and the rehab code allows you to consider the original design loads if the stress doesn't increase by 5%.

I think the proper solution is to implore the ICC to do away with the 5% rule and design it for ice. Also, I don't think it's valid to overlook ice just because they didn't consider it back in the day. Just wanted to see what others thought.

 

[phamENG]

The wording is a bit circular. But it reads to me like it says "if these loads create a worst case load effect, use them in the design of the structure." Using ASD as an example, if D+L+Di is greater than D+L, D+0.7Di+0.7Wi+S is greater than D+(Lr or S or R) or 0.6D+0.7Wi+0.7Di is greater than 0.6D+0.6W, then it's ice sensitive.

This does strike me as something that would be ice sensitive. "Open catwalks and platforms" would likely have some sizeable members- deep channels or wide flange members in the 8 to 24 inch range depending on span. I think your structure falls into a similar category.

 

[phamENG]

Oh - I understand what you're doing now. Yeah. I'd say you're okay to ignore the ice...but make sure you disclose this to your client in writing. Let them know the justification under the building code for ignoring it and give them a picture of what it would like you if they had to upgrade. Let them choose the IEBC exception with your guidance.

 

[straub46]

Agree with pham. I would still consider ice on the new members and make sure the stress ratio doesn't increase 5% with it considered.

Go Bucks!

 

[WinelandV]

Crabby,

I used to work for a company that fabricated conveyor catwalks, pipe racks, and bucket elevator support towers (like what Warrior and Brownie do). All of the designs I did included ice as laid out in the LRFD 7. To my knowledge, everyone in that industry included ice loads in their catwalk design. So that's what drove my answer of yes, this is ice sensitive.

Regarding your situation, I feel for you. The pipe rack is there and obviously wasn't designed for ice. I don't utilize the IEBC very often (probably only once or twice in my career), but hopefully there's something in there that can help you out.

 

[dhengr]

CrabbyT:
It’s not just fresh water spray that causes ice build-up. Don’t you have sleet storms in MI? Those particular structures are ice sensitive structures, in part, because a significant ice build-up on them can amount to a significant percentage of their weight (DL), maybe more; and double/triple the wind exposure area. That doesn’t mean that there can’t/won’t be ice build-up on other structural members too. The older versions of the codes assumed that the design engineer had some experience and imagination as to ALL the various loads that his/her structure might be subjected to. We now know that is not a safe/wise assumption. I don’t think you will find anything in the UBC which says you shouldn’t include ice loading where appropriate. They just didn’t have 37 page iterative formulas and calc. processes to determine how much bird poop, or some such, that there could be on a sq.ft. of structure, and the geometry of the pile. Although, there have been some corrosion problems caused by this, and nobody paid any attention to that potential until the structure started failed. How old are these trusses? They have at least withstood that test-of-time, what additional pipes and loads have been added over that lifetime, above the original design loading? Look at the weather data over that time for max. wind and ice storms, etc. What were the actual design criteria under that older code? You might find some allowable stress increases for combined, short term loading, like wind and gravity. And, ask your local AHJ what their wind and ice loading criteria, was then, and is now, and how they want you to determine/justify grandfathering of such structures, within reason. Have you inspected the structures to determine their current condition, and how that my down-rate them?

 

[CrabbyT]

wine,

Neat coincidence! I worked at a similar place where we designed/fabricated conveyors, grain bins, and farm structures. We included ice on everything we designed. Loved that job, but they shut down the plant... I'm pretty proud of the one shown below. It survived some pretty strong winds back in 2018 when Hurricane Michael rolled through (although I think it was a tropical storm by the time it hit the thing below).

dhengr,

We definitely have sleet storms in Michigan. But in my 30+ years here, aside from the piers on Lake Michigan and aside from trees that accumulate ice, I've never seen anything remotely close to what you get if you do a Google image search for "atmospheric ice loads". That's not to say it can't happen (I mean, there's a lot of things I've never seen... doesn't mean they didn't happen). I'm just curious about the body of knowledge around ice loads. I'm curious if the amount of ice prescribed by ASCE 7-10 actually applies to large diameter pipes and wide flange beams, or if it's just theoretical and potentially overly conservative.

The thing about the UBC is that they make zero mention of ice loads. In my mind, that's problematic. For the engineer who originally designed this truss, ice was probably out of sight / out of mind. For the box truss in my original post, the ice loads are about equal to the dead weight of the steel and pipes. That essentially cuts the allowable load carrying capacity of the structure in half.

Also, the ASCE 7-10 commentary states the following (and this is kind of why I was questioning the validity of whether or not open structures made of relatively large steel sections can be considered ice sensitive). Based on the quote below, the ice loading criteria in the code is based on transmission wires. I'm curious how well that extrapolates to things with larger cross sections. I was hoping someone here might point out some wind tunnel tests involving ski hill snow cannons that experimentally evaluated ice accumulation on things. Kind of makes me want to contact my local ski hill...

Very few sources of direct information or observations of naturally occurring ice accretions (of any type) are available. Bennett (1959) presents the geographical distribution of the occurrence of ice on utility wires from data compiled by various railroad, electric power, and telephone associations in the 9-yr period from the winter of 1928–1929 to the winter of 1936–1937.​

Can confirm that I've looked at the structures several times. They look pretty good aside from the fiberglass bar grating... that stuff freaks me out. Cracking noises at 30' in the air are unsettling.