Roof Anchorage to Historic Building

[phamENG]

Working on a historic structure, triple wythe brick walls. Brick is in fair shape, will be repaired where it isn't.

Roof at the back of the building is shot. Rafters and ceiling joists rotten at bearing. So we'll be doing repairs. 2018 Virginia Existing Building Code governs, and says the following:

...Repairs to historic buildings need only comply with Chapter 9.

903.1General.

Repairs to any portion of a historic building or structure shall be permitted with original or like materials and original methods of construction, subject to the provisions of this chapter. Hazardous materials, such as asbestos and lead-based paint, shall not be used where the code for new construction would not permit their use in buildings of similar occupancy, purpose and location.
903.2Moved buildings.

Foundations of moved historic buildings and structures shall comply with the VCC. Moved historic buildings shall otherwise be considered historic buildings for the purposes of this code. Moved historic buildings and structures shall be sited so that exterior wall and opening requirements comply with the VCC or with the compliance alternatives of this code.
903.3Replacement.

Replacement of existing or missing features using original materials shall be permitted. Partial replacement for repairs that match the original in configuration, height, and size shall be permitted. Replacement glazing in hazardous locations shall comply with the safety glazing requirements of Chapter 24 of the VCC.

Exception: Glass block walls, louvered windows, and jalousies repaired with like materials.

Built in the 1800s, this building had zero anchorage for the roof. The rafters just sit up there. They were nailed to wood beams elsewhere, but they just sit on the brick wall with nothing but gravity and perhaps some continuity of the lapped framing to hold them down. By my reading of this code, I don't have to do anything. Naturally, that doesn't sit right with me.

So if I'm going to do something, I want to do it right...well I have about a 60plf uplift load at the wall (0.6W+0.6D). Not a lot, but when you consider we can't put unreinforced masonry into direct tension, and the interior and exterior faces of the wall cannot be touched as the historic character can't be damaged (tax credit job with State review)...things get difficult. Based on window spacings, I have to space to my anchors out about 6 or 7 feet (single course of brick over the windows). That gives me a concentrated load of about 335#. To engage a wedge of brick, I'd have to go down over 2ft. Having somebody drill a 3/4" hole over 2ft into the top of a wall 60ft above the ground to put a plate on the bottom in a small cavity that will be refilled sounds absurd from a constructability point of view. And to say "the code technically says you don't have to bother with this, but here's this challenging and potentially very dangerous detail to build, do it anyway" isn't going to go over well.

Anyone have any better ideas for concealed anchorage to historic brick?

 

[SWComposites]

can you face bolt/anchor some sort of straps to the inner surface of the wall that extend up to attach to the rafters?

 

[MotorCity]

I understand the 6' to 7' anchor spacing, but I assume that your rafters are spaced closer than that. How do you plan to transfer the uplift to the 6' or 7' intervals (or maybe thats part of your question)? Not knowing the specifics of the brick, I'd be hesitant to rely on the uplift capacity of it unless I was able to engage a whole bunch of it (i.e. a big wedge) which I agree is not practical. Can you add 60 plf of new construction along the wall to counteract the uplift? I'm thinking a wide flange beam around 60 plf along the wall. That would tie all the joists together to act in unison against uplift. Connecting the beam to the joists could be interesting, but doable.

 

[phamENG]

SW - interior and exterior of the brick are both exposed. Tax credit job means that the historic character has to be preserved. Steel plates with bolts weren't there 100 years ago, so they can't be there now. (I realize it's impractical, but that's not my call or even the owner's).

MotorCity - an interesting idea. The load is small enough that I have a wood beam running along the top of the wall that will be anchored and spliced with the rafters fastening to that. Using a big steel tube might get me there.

What about thoughts on the code as written? A couple tons of steel and a very large crane that the code doesn't mandate will also be a tough pill to force feed them, too.

 

[EZBuilding]

Feels like you could argue that this scope of work is similar to "Voluntary Lateral Upgrades" identified in an alteration level 2. Which I think generally conveys the idea that you can't get punished for improving the situation of a building. I would recommend discussing this in general terms with the owner - just to make sure a shark contractor doesn't try and undercut you later by saying none of this needs to be done.

Could you put a continuous 2x8 below the rafters which gets anchored into the brick wall? The rafters could have hurricane clips down into the continuous 2x8. I feel like this benefits you in being able to "increase" your wind effective area across a larger span of the roof. If you're committed to getting this to pencil out - you may want to consider utilizing the full dead load in determining what is your uplift reaction.

 

[RangeRock]

It seems to me like the Owner should understand well that difficult or onerous construction methods will almost be a guarantee with buildings like these. Props to the engineers that take these types of projects on as they often end up turning into very complicated, high-liability, and time-consuming renovations.

All that being said... from a life-safety perspective I think if you're going to stamp it - it needs some sort of uplift anchorage even if not "required" by VEBC.

Drilling cores down into the top of the existing wall seems to be your only path forward if any type of exterior alterations will de-rail the project.

 

[dhengr]

PhamENG:
I agree, to do it right, you should probably not do nothing. What you really want is 120lb. anchors (60lbs. up, by 2' o/c) under each joist. Then take full account of every oz. of brick (DL) in those upper 3 or 4 courses. Depending upon how the three wythe brick wall is bonded, you may be able to excavate the middle wythe, fairly easily, three or four courses deep and maybe the inside wythe too, to build yourself a more or less continuous bond beam which is kinda bonded to the rest of the brick wall. The historic building aspect really pertains to the exterior appearance and details, and significant interior architectural details, etc., doesn’t it? If you can hide it or you are making a significant structural improvement to the building, with today’s materials and to today’s stds. you should be able to make a fairly convincing argument for the improvement.

 

[BridgeSmith]

I agree with what dhengr wrote, but I would suggest if you're going to drill in anchors, you should go at least 15" deep, to be sure that you're engaging enough of the mass of the wall in compression, so you're not relying at all on the tension capacity of the masonry.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[RWW0002]

I do not have a great solution for you but I will double down on you for the "problems" side..

In my experience with these types of buildings, I am usually as concerned about a lateral and out of plane tie to the roof as I am about uplift. Depending on the condition (or presence) of tie rods or similar structure, there may be very little holding the wall in place and I have seen more than one multi-wythe brick wall with significant out-of-plane deformation. The only true failures of multi-wythe brick buildings that I have seen have been due to leaning or out of plumb walls that have leaned away from floors a bit more year after year until they eventually come down. Scary Stuff. I have elected to tie off multi-whthe brick to floors and roofs and justified it with the "voluntary wind or seismic upgrades" portions of relevant codes in past projects, as mentioned above, and have not had an issue with the historic preservation guys.

 

[kipfoot]

Depending on how many locations you have, your solution doesn't sound so bad. I might lean toward more, less deep, anchors (like dhengr suggested) and consider using epoxy rather than an embedded plate, if that gets you enough wall dead load engaged. (The plate can create its own problems if it's not stainless, so be mindful not to create a worse problem with the solution)

It might even be easy enough to remove bricks in a vertical slot to place your threaded rod to engage the wall. Like, if it's a question of "make a slot in these 6 locations to anchor the new joists," that doesn't sound too unreasonable as a way to keep the roof from unzipping.

Also, I've found that DHR can give a little when issues are structural and not exposed in the ground floor public space.

 

[KootK]

They were nailed to wood beams elsewhere, but they just sit on the brick wall with nothing but gravity and perhaps some continuity of the lapped framing to hold them down.

What's the parapet detail like? I'd normally expect the masonry to continue above the roof some distance to form the parapet.

Rationally, it's pretty difficult to tag something as deficient when it's been working for 100+ years unless something significant has changed recently. As you know, in Canada's code there's a statistical way to somewhat account for that. It doesn't get you past a nonexistent load path however.

I suppose that you could hang a couple of 8" diameter, water filled pipes from the ends of the joists and call that "hold down". I'd still not be inclined to call that "positive anchorage" however.

 

[BridgeSmith]

Rationally, it's pretty difficult to tag something as deficient when it's been working for 100+ years unless something significant has changed recently.

Most likely, there's never been enough windward doors and windows open during a strong wind to create the positive air pressure needed to overcome the weight of the roof.


Rod Smith, P.E., The artist formerly known as HotRod10

 

[KootK]

I also wonder if, to an extent, an uplift event might be considered a serviceability failure. From the details provided thus far, I get the impression that the masonry wall might actually be cantilevering up from the floor below rather than relying on the roof diaphragm for support.

How does the roof look next to the wall? Does it show evidence of past movement? Is that why it's rotted near there perhaps?

 

[phamENG]

dhengr - thanks. I've used that detail before, and was planning to use it on this one. But the window heads are within a couple courses of the top of the wall which compromises the continuity. Still may work, though. Just have to make it a little deeper and work out some connections at the windows.

BridgeSmith - agreed. Direct tension in URM is not permitted, and would be most unwise in a 100+ year old brick wall.

RWW0002 - I have dealt with the same. This one doesn't have that problem. The walls are quite thick and well tied to the floor diaphragms. Not so much at the roof, but it appears to be cantilevering up from the floor below well enough - for now. I'm still considering it in my design, but I'm not going to lose sleep over that detail on this one.

kipfoot - thanks. I always specify at least HDG for anything embedded in the wall. Lintels, bolts, etc. SS preferred, but I don't push it unless it's near the water. I have to go through an architect and a historic preservation consultant before it gets to DHR - and they both said they won't be taking anything not hidden up for approval. Cest la vie.

KootK - no parapet. This is the back edge of a free draining, monoslope roof. That's why the deck and joists are shot. The roofing and flashing failed, and water ran down, around, and back into the roof. I do love that provision in your code, and would like to see it here. But there's a difference between understanding that the materials are significantly stronger than we assume (or were, before the fungus ate them) and saying that because a 700 year storm hasn't hit in the last 120 years, we don't have to worry about it. And yeah, non-existent load path doesn't make me feel good. I disagree on the serviceability failure idea. If it does lift up, it's reasonable to assume significant failure of the roof has occurred. Even if the wall is cantilevering (which I believe it is), it's now a 10' tall cantilevered parapet with, ostensibly, a whole lot more wind load. Even if the wall worked, I think the additional load on the anchorage to the floor below (such as it is) would likely be exceeded. So, without doing any number crunching and relying on my gut, I think we'd end up in a cascade of minor failures that would culminate in a very large boom.

 

[phamENG]

Thanks, everyone. I think I'm going to go with a middle ground approach. The code says I don't have to do anything, which is a bad idea. Doing it strictly by the numbers for new construction is onerous. My goal is going to be to achieve positive anchorage of the roof to the wall, but with a willingness to shave off some of the factor of safety with an eye toward the fact that the existing non-existent load path has "worked" for more than a century. Using the bond beam with reasonable depth, I can get a 1.0D+0.6W loading condition to reach equilibrium. I don't love it, but I believe that it's a pragmatic solution that exceeds what the code calls for and provides sufficient safety for all but the worst case storms.

Anyone think this course is in any way negligent or too lax?

 

[KootK]

With the middle ground approach, I feel that the key becomes careful communication with the owner about what they are and are not getting for their money. No big deal, you just gotta sell it well and, perhaps, document it. You want to avoid the owner hearing:

[spent money = eliminated uplift risk]

...when it's really...

[spent money = uplift risk mitigated only where work is being done and not quite enough meet modern new code requirements]

The engineer in me vastly prefers middle of the road approach to nothing at all. However, I find that clients tend to think of these thing in more of a binary fashion. It's prudent to head any misunderstandings off at the pass and achieve true buy in for your plan.

 

[BridgeSmith]

It seems like a reasonable approach to me.

For additional justification, if you feel you need it, I would look at ways to reduce the wind pressure based on realistic limits of positive wind pressure within the building due to a limited inflow of wind from any one direction (the doors and windows on one side) and the presumably substantial availability of outflow, due to the typically leaky nature of older construction. IOW, it may not realistic to assume the build up the positive pressure that the code specifies for typically tighter new construction. I'm not sure what kind of guidance is available in regard to quantifying that reduction, but just intuitively it seems it could be substantial.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[phamENG]

KootK - sounds like good advice. Thank you.

BridgeSmith - I'd have to think on that a bit. I'd agree if this were an old wood building, but foot thick brick walls tend to prevent drafts pretty well, and the architect on this job is a good one who will have good details sealing overhangs, windows, etc - so as long as they are executed properly this will be a pretty 'tight' building.

 

[KootK]

I also had an idea for a repair strategy. It probably won't fly for any number of aesthetic or heritage reasons but I'll throw it out into the ether anyhow seeing as it's already occupied my thoughts for the duration of a hot shower.

1) Attach a small timber "rail" below the joists just in front of the wall.

2) Attach a similar rail below the floor joist below.

3) Behind the masonry piers, connect the two rails with a tension rod.

 

[BridgeSmith]

I didn't realize the entire building was being renovated and sealed up well. The historic buildings I've seen, even after renovation, were not very tight, mostly because the windows had to be so 'authentic' that it was impossible seal them very well.

Rod Smith, P.E., The artist formerly known as HotRod10