Repair of Corroded Beam Web 2

[AK4S]

For an existing Highway Bridge we are reviewing, there is significant loss of section in the beam webs at the bearing location. I have attached some sample photos (these locations are not too bad, some others are worse).

The current proposal is to reinforce the beam web by bolting on new plates on either side of the existing web and extending the plate (away from the bearing) till the full web thickness is available in the original beam.

The idea is to provide new plate thickness to match the original web thickness of the beam(ignoring the deteriorated web thickness)and hence maintain the shear capacity of the section.

For example:
Existing web thickness =0.625"
reduction due to chamfer =1.0625"
New Plate thickness required = (0.625"/2) + 1.0625" = 1.375". Hence provide 1.5" thick plate on either side of existing web.
[ul]
[li]I found similar details (Link)shared earlier in this forum and so I guess the philosophy makes sense. Also See attached file with details from one of the responses.[/li]

[li]How should I go about evaluating/preparing calcs for the proposed repair? evaluate New plate thickness (match or thicker than existing web?), Connection bolts (check for shear only?), extent/length of plate required? Does AASHTO have any guidance?[/li]
[/ul]
[li]What can I use to Seal the edges of the new plate to prevent moisture from getting between the new plates and the existing web[/li]

Would appreciate any guidance and feedback.
Thanks

 

[bridgebuster]

AASHTO doesn't have any guidance but attached is an FHWA publication that should help; they have design examples. Typically we design for shear only, matching the web thickness. We typically use sandwich plates; it just looks cleaner. Sometimes, it's a single side because there's too many things in the way. I've used beveled plates, like you sketched, and sometimes a fill plate, to get past to get past the fillet, with an unbeveled repair plate. Clients have different preferences. You can seal the edges with silicone or Splash Zone (a Carboline product). To me silicone is easier.

 

[AK4S]

@bridgebuster
Thank you for the feedback and sharing the FHWA document. I will look though it and reach out in case I have further questions.

 

[bridgebuster]

@AK4S - no problem

 

[r13]

Before going too far, I suggest to thoroughly clean up the girder and have a non-destructive testing (NDT) to fully uncover the extent of damages, and maybe find the root causes. Sometimes the rust sections/areas can look real bad, but the lose of material is really minimal. Also need to think ahead on how to prevent the corrosion from reoccurring and progressing further. No body can see the defects once it is covered up.

 

[AK4S]

@retired13: the root cause is leaking joint in the deck slab above the pier.

 

[r13]

Glad that you found it. The leak should be stopped before the girder works.

 

[bridgebuster]

@AK4S - it almost looks like we were at the same bridge recently.

This bridge opened in 1963; it was rehabbed in the early 90's - steel and concrete repairs, new deck joints, etc. A number of stringer ends were repaired; I only found a handful of stringers that need to be repaired. However, 13 out of 27 piers have to be replaced because of the joint leakage over the past ~25 years. They're too far gone to repair.

 

[r13]

Was that your initial on the web? It looks like someone wrote BB using yellow chalk

 

[bridgebuster]

It sort of does, you made me look twice. It say 66-2; not always easy to write on rusty steel.

 

[AK4S]

@bridgebuster:
looks like something similar, even around the same time period. But the one I looked at is quite smaller. 3 spans, 2 piers.

 

[AK4S]

@bridgebuster
From the NYCT repair details you shared on the other thread, I want to focus on two different conditions for web strengthening:
1) New plates are added on either side of the deteriorated web (existing section remains as is):

2) Deteriorated web is cut and removed and new plates added as replacement:

What determines if the existing deteriorated web needs to be removed or not? It is my understanding that for typical section loss, the corroded area can be cleaned and prepped to bright metal and the new plates added on either sides. If the section loss is quite significant (paper thin remaining) or there are holes in the web, then it is required to cut and remove the existing web to create a defined cut edge and avoid stress concentrations in the existing web.
Are there any guidelines?

 

[bridgebuster]

@AK4S - steel repairs are combination of art and science, if that makes sense, especially with bearing area repairs. In theory, unless the dead load is jacked out the repair will only be for live load or if the deck is removed the dead load stress can be reduced, but that's a "whole 'nother" discussion. As I said earlier, different agencies have different ideas, even within an organization different people have different ideas. I posted those transit details because they covered a wide variety of situations. Personally, I don't like "amputating" part of a stringer or girder and not putting something back, especially below a leaking deck joint. I have no doubt about the effectiveness of NYCT amputation detail; they're a conservative group. Typically, what I've seen and done is leave what's there and clean it up before adding the repair plate. Sometimes, if the web is really pitted badly, I'll call for a filler material, like Splash Zone, to get better contact between the plates.

Attached are some other details. These are bearing area repairs. I had to fix up the original details because the person who did them didn't account for the stringer fillet. There are 7 sheets but essentially they're all the same concept. These bridges also had low rocker bearings.

I also have some details, by others, with welded details, if you're interested.

 

[AK4S]

@bridgebuster:
I see now and agree, there is no reason "amputating" part of existing beam. I could just design the new plates to match existing web thickness. At the repaired area, the load path should be through the new web-plates bypassing the existing web. So stress concentration around any holes in the existing web should not be an issue (the deck is getting replaced, so dead load stresses will be reduced). I can apply a filler material to fill up cavities in existing web to get better contact with the new plates.

Thanks for sharing details from your work. I plan to use bolted connections for proposed repairs, so these give me a good reference.
I was wondering regarding the use of vertical angle sections in many of these repairs(see snapshot below)instead of just plates parallel to the web.
Is the idea to provide a matching stiffener to reinforce the web? i.e. to match the existing stiffener on other side connected to the diaphragm.

 

[bridgebuster]

@AK4S - you're quite welcome. I like discussing bridge rehab with others, although some people hate it with a passion.

These were fascia all stringer repairs. They didn't stiffeners on the outside face but if it were an intact stringer we wouldn't need them anyway. However, most of the beam ends were in poor condition. This wasn't a deck replacement project. We couldn't get the web repair plates where the diaphragms are on the inside face; it would have been messy. So we opted to use an angle with equivalent area to stiffen it up. Hope that helps.

 

[AK4S]

Rehab work is interesting and I find it beneficial to bounce ideas off others. There are multiple ways to fix an issue and such discussions can bring to light an easier approach or potential issue which I might not have thought about.

Thanks for your clarification on the detail. It makes sense

 

[bridgebuster]

no problem

 

[r13]

bb has contributed significantly on this thread, I am fully with him in hating to cutoff a piece of metal and thinking can restore the structure back to the original condition. Depends on the severity of the corrosion, my repair preference is usually 1) thoroughly clean the metal, apply anti-corrosion agent, and strengthening with additional steel; 2) cut off a segment of the defective beam/column, and reattach a segment of the same; 3) replace the defective beam/column. I think all of us consider these steps as well. The only thing that serving as a note here is, I don't worry much about "under strength" of the repair, but I do worry "over strength" nature of it, as the over stiffness locally sometimes will cause negative effect on the remaining structure, and tip the structural balance of the entire structure one way or the other.

 

[AK4S]

I am revisiting a detail on the above mentioned repair and wanted to see what opinion others have on the following assumption:

I proposed the below detail using 1.5” thick plates initially on each side to get around the fillet in the existing beam at the web-flange interface.
The 1.5" thick plate was beveled at the bottom end. By strength and to keep the typical bolt spacing not closer than 6", I would need only 0.75" thick plate on each side.

I am now considering to reduce the plate thickness to 0.75" (see markup in red in above sketch) and to stop the new web plates with no bearing contact with the bottom flange and rely on connection through the angle for force transfer, as shown in sketch below.
To provide proper shear strengthening, the new material needs to be connected to the web and bottom flange to provide resistance to the horizontal component of the shear.
I believe the detail shown achieves the purpose.

However, right above the bearing location, I cannot install the angle due to interference with existing anchor bolts. Can I assume that the horizontal shear transfer would occur before the bearing, so I do not need the angle here.

The existing web right above the bearing has about 1/8" section loss and 2" dia. hole. I believe the 0.75" plate provided should locally span over the deterioration. However the new web plates will not have direct bearing contact with the bottom flange at this location. Does that look ok?


I need the angle (away from bearing) at some of the repair locations since the bottom of the existing web is not suitable for force transfer to the flanges.
And I am trying to avoid using a fill plate to get past the fillet and with a 0.75" unbeveled repair plate (fully bearing on bottom flange) for the desired thickness for strength alongwith with the angle, unless that is the only option.

 

[r13]

AK4S,

First, allow me to congratulate you, as your question has gaining depth nowadays.

Second, unfortunately that both the most trustworthy bridge experts on this forum are missing, I can only try to offer my opinion the best I think it works: 1) for non-sliding bearing support, consider to bring the plates as low as possible, then provide full/partial penetration weld at plate-beam fillet; 2) for sliding support, consider use a larger/longer angle with oversized, or slotted hole, release the existing bolt and place the new angle, then retighten. Would these help?