MSE wall connection design

[MSEMan]

Here's a really strange one.

I was reviewing a MSE design the other day, and the (well known and very well respected) supplier assumes ZERO corrosion at the connection between the tie strip and the reinforcing strip. After some inquiry, I learned that the theory is that the tie strip acts as a gasket for the reinforcing strip, so no corrosion occurs on the reinforcing strip at the bolt hole. Also, only the exposed face of the tie strip suffers corrosion.

Well, I've been on sites and seen these things cast in strange angles and bent all sorts of ways so I just can't buy that no water gets in there, but they say all State DOT's accept it. By my numbers the design life is half of what it should be.

AASHTO says all connection elements need to have corrosion allowances, and to my thinking a hole in a strip is a critical part of the connection. I'm thinking that too many DOT's may let geotechs run with approving the design without getting structural involved. Who knows what the panel design would throw up?

Any thoughts?

 

[jdonville]

MSEMan,

Probably the influence of corrosion, if accounted for, would require a resizing of the metal strips and/or connections at the facing panel.

Naturally, this would mean that 1) somebody missed it or made the assumption during the initial system design years ago, 2) the MSE supplier would have to update their standard drawings to account for the change and likely incur a review with all DOTs that have approved the system, 3) this may make certain standard parts obsolete (and therefore unsaleable).

Since this would probably end up costing the MSE suppliers money that they wouldn't otherwise want to spend, the obvious solution is to pretend that it's not a problem.



Jeff

 

[Panars]

I pulled out some calculations I have from Reinforced Earth (circa 1992) that show how they calculate the maximum strength of the strip. They are assuming corrosion of the exposed surfaces at the connection between the strap and the tie strip. The exposed surfaces in this case are one side each of the two tie strips. They are assuming no corrosion of the strap at the connection because it is protected by the tie strips. I believe this is the part you are talking about MSEMan, correct?

AASHTO 17th edition, Seciton 5.8.6.1.1 states that "The sacrificial thickness is computed for each exposed surface ..." Based on this, I would say ReCo is in compliance with AASHTO.

You may be right that water will get into some of the connections. But it won't be all the connections, and the rate of corrosion within the connection would be less than the exposed portions of the strap. I'm not too concerned with this issue.

Since Jeff is quite a humorous guy (albeit a dry sort of humor) I will take his comment in the spirit in which it was intended (i.e. only semi-serious).

 

[MSEMan]

Panars,

I'm curious about your assertion that the design method is AASHTO compliant.

The next AASHTO article, 5.8.7.1 relates to connections and states "... connection elements shall be designed to accomodate losses due to corrosion ..." Well I'd say that a bolt is a connection element and so is a tie strip - now if I'm taking corrosion allowances off of those how can I justify not taking it at the strip/bolt hole section since this is the section that encases the bolt? Here's the rub: I'm supposed to beleive that at the beginning of the design life the strip at the bolt hole is the weak tensile section, but at the end of the design life the strip body governs? Well, maybe, but I've got a drainage layer behind the facing and even if I didn't that's where all the water is going to migrate. When I have a look at a site installation, far from seeing that "not all" connections aren't water tight, I can't see a single one that would meet this standard. So what is an "exposed surface?" To my mind it is any surface that may become exposed to corrosive influences such as soil, water, or humidity.

I did a little research and found the UK BBA certificate for the RECO system (similar to a US HITEC) online. In this certificate, which is apparently used throughout Europe, corrosion is taken at the cross section with the bolt hole, and thus this reduced section governs tensile capacity throughout the design life. If RECO take these corrosion allowances in Europe why not in the USA?

Adding all this together, I'm not feeling comfortable with the assumption that the tie strip provides effective corrosion protection at the bolt hole.

I'm interested in your considered response.

 

[Panars]

You have some good points. Is the UK BBA certificate available online? I would like to see how they are calculating the strength at the connection. Also, what design life and corrosion rates do they use?

I'll see if I can go out and see some RECO walls being built in my state.

 

[jdonville]

Panars,

Sorry - I've been too busy recently to respond properly to your half-correct identification of "dry humor".

In point of fact, I would think that you are probably better positioned than many to check into the design assumptions, research and realities with respect to connection corrosion in MSE wall systems.

A literary character (probably Hugh McWearie in Robertson Davies' excellent "Murther and Walking Spirits") once said something like "Always ascribe the worst motives to people - always hoping that you're wrong, of course." This is similar to Scott Adams' question, "What's more likely?", from his excellent "The Dilbert Principle".

Jeff

 

[MSEMan]

Here's the link:

http://www.bbacerts.co.uk/certs/roads/R106i2_web.pdf

What you'll see is a special strip with an 8mm thickened section used for connection when tensile strength governs.

 

[Panars]

Looking at the BBA certification, it is interesting that they end up using a design strength of 25.9 kN (5.8 kips) for the 50mm x 4mm strip. This is very close to the 5.5 kip strength I am used to seeing for AASHTO design with a 100 year design life. Even though the BBA and AASHTO have different assumptions about the corrosion, they end up with very similar design strengths.

The galvanization thickness for the BBA is a little thicker, but this would be offset by a longer design life (120 years). The BBA must use lower corrosion rates than AASHTO. I know there has been a lot of discussion lately about the AASHTO corrosion rates.

We are planning on demolishing a RECO MSE wall that was built in 1989 (have to make way for a wider and bigger bridge). I plan on studying the corrosion of the soil reinforcement, and will try to remember to also look at the connections.

 

[MSEMan]

Panars,

Is the currently accepted design practice in your State not to take any sacrificial thickness at all for corrosion at the bolt hole section? If so, is it stated in the special provisions?

I'm looking for any published reference at all that suggests that this is an OK practice. The closest I've been able to come is a comment in BS8006 about taking a reduced allowance for bolts enitrely encased in the connection. Since the edges of the strip aren't encased I'm even having trouble using that.

 

[Panars]

It is the current practice to assume no corrosion of the reinforcing strip at the connection because it is protected by the tie strips. This is not stated in the specification. It was the design method used by RECO when they submitted their wall system to Ohio DOT for approval.

 

[MSEMan]

Panars,

Thanks for your clarification.

If the zero corrosion assumption was accepted then fair enough - an owner/approval authority can specify whatever design methodology they deem most appropriate and live with the consequences.

In my State, the DOT was actually surprized to learn about this issue, it simply hadn't been considered in any formal way before. Given what I've found out, I'm going to stick to my interpretation of AASHTO, which is that the bolt hole is a critical connection element and that AASHTO requires corrosion to be taken at that critical section.

 

[MSEMan]

Panars,

I looked up some docs. Here's one from your State:

http://www.dot.state.oh.us/construction/oca/Manuals/2006_MOP/Web_Manual/SS840.htm

Check out fig. 840.06.H8. I'm not convinced that's effective corrosion protection.