Temporary Works in Construction Discussion Group

Flush end-plate connections have historically been simplified as pinned connections.  However, it is known that these connections will attract some moment.  While it may be conservative to consider that this connection is pinned for analysis of the member, this assumption unconservatively ignores any tension that is developed in the bolts from the restraint.  With many types of temporary works (for example building steel erection), I would recommend analyzing an envelope (pinned for global stability and member loads, partially restrained to check possible connection demands).  However, with support of excavation, in my experience, these types of connections are typically just analyzed as pinned and no further envelope analysis is performed.

During a recent design review for a waler/support of excavation, the EOR considered the connection to be pinned and selected A325N bolts, resulting in 99% shear utilization.  The design reviewer then ran a failure analysis (removed all safety factors and determined how much moment the connection could take, up to the combined shear/tensile strength of the bolts). The design reviewer found that the connection could safely handle <5% of the moment based on ultimate rupture strength of the bolts.  Therefore, the design reviewer recommended several changes to the design 1) change the bolts to A490X to provide increased capacity 2) install the bolts hand tight/use loc-tite (do not torque or overtighten them), allowing the connection to naturally open a bit and rotate to limit the tension that would be induced on the bolts.  These recommendations were relatively easy for the EOR to incorporate (caused no changes to fabrication and no delay to the work).  However, the question remains whether these changes were actually necessary and/or if treating these connections as partially fixed should be the new norm.

Survey to the discussion group:
1) Do you treat flush end-plate connections as pinned, partially restrained, or both?
2) For support of excavation, do you believe it is generally safe and appropriate to just treat these connections as pinned, since this has been standard industry practice and the loads for this type of structure are not necessarily uniform or occur at full magnitude?  Or, do you think it's important that engineers consider the moment that can be attracted by the connection and ensure that the connection is designed to safely handle that?

Thanks in advance for your input!


------------------------------
Angela Hunter P.E., C.Eng, M.ASCE
Teaneck NJ
------------------------------

Angie,

Typically we've designed flush end-plate connections as pinned in the ERS realm (with no issues in performance), but we have run into similar partial restraint comments.  There have been a few instances where we have designed using 50% moment releases based on comments with minimal revisions to the connection details.  Every situation should be reviewed individually, but typically we believe it is appropriate to design as pinned.  However, knowing the potential for some tension to develop from partial restraint, good design practice would be to make the connection more robust to leave some reserve capacity (i.e. not 99% utilization).

For the particular instance you mentioned - leaving the connection finger tight or loose may lead to other unintended issues (ERS rotation/movement).  It may be more prudent to make the connection more robust.

Jason

------------------------------
Jason Schneider
Engineer
Collins Engineers
Chicago IL
------------------------------

Original Message:
Sent: 01-30-2023 11:12 AM
From: Angela Hunter
Subject: Pinned vs Partially Restrained Flush End-Plate Connections

Flush end-plate connections have historically been simplified as pinned connections.  However, it is known that these connections will attract some moment.  While it may be conservative to consider that this connection is pinned for analysis of the member, this assumption unconservatively ignores any tension that is developed in the bolts from the restraint.  With many types of temporary works (for example building steel erection), I would recommend analyzing an envelope (pinned for global stability and member loads, partially restrained to check possible connection demands).  However, with support of excavation, in my experience, these types of connections are typically just analyzed as pinned and no further envelope analysis is performed.

During a recent design review for a waler/support of excavation, the EOR considered the connection to be pinned and selected A325N bolts, resulting in 99% shear utilization.  The design reviewer then ran a failure analysis (removed all safety factors and determined how much moment the connection could take, up to the combined shear/tensile strength of the bolts). The design reviewer found that the connection could safely handle <5% of the moment based on ultimate rupture strength of the bolts.  Therefore, the design reviewer recommended several changes to the design 1) change the bolts to A490X to provide increased capacity 2) install the bolts hand tight/use loc-tite (do not torque or overtighten them), allowing the connection to naturally open a bit and rotate to limit the tension that would be induced on the bolts.  These recommendations were relatively easy for the EOR to incorporate (caused no changes to fabrication and no delay to the work).  However, the question remains whether these changes were actually necessary and/or if treating these connections as partially fixed should be the new norm.

Survey to the discussion group:
1) Do you treat flush end-plate connections as pinned, partially restrained, or both?
2) For support of excavation, do you believe it is generally safe and appropriate to just treat these connections as pinned, since this has been standard industry practice and the loads for this type of structure are not necessarily uniform or occur at full magnitude?  Or, do you think it's important that engineers consider the moment that can be attracted by the connection and ensure that the connection is designed to safely handle that?

Thanks in advance for your input!


------------------------------
Angela Hunter P.E., C.Eng, M.ASCE
Teaneck NJ
------------------------------

Excellent topic.  It makes a difference if the cofferdam is in earth or in water.  Earth loads are often over-estimated, while water loads are 100% of estimated. Load probability plays heavily on design attention. The welding of the end plate is a higher concern than the bolts, as poor welding details can lead to premature failure.  That said, I use 80% of friction bolt capacity, as a guide.  

When I started designing cofferdams, the field practice was to use shiplap corners, cutting one half of each flange off each piece and lapping the webs to each other. The webs were then bolted together, loading the bolts in shear only.  This morphed into end plate connections, similar to how the strut ends were fabricated.  We tried double clip angle connections, but stopped as you have to believe they act as true pins, and the field guys didn't like the look of them.

If you do pay the price for all or partial moment capacity, remember to take advantage of the positive moment reduction in the adjoining members, or just keep it in your back pocket for when they come and tell you the cofferdam has to go deeper or will have a higher surcharge load than they told you originally.  

Anybody want to talk about inny or outty sheet pile corners of cofferdams?

------------------------------
Alan D Fisher, PE
Construction Structures Engineering Consultant
Cianbro Corporation
Portland, Maine
207-553-2703
------------------------------

Original Message:
Sent: 01-31-2023 04:28 PM
From: Jason Schneider
Subject: Pinned vs Partially Restrained Flush End-Plate Connections

Angie,

Typically we've designed flush end-plate connections as pinned in the ERS realm (with no issues in performance), but we have run into similar partial restraint comments.  There have been a few instances where we have designed using 50% moment releases based on comments with minimal revisions to the connection details.  Every situation should be reviewed individually, but typically we believe it is appropriate to design as pinned.  However, knowing the potential for some tension to develop from partial restraint, good design practice would be to make the connection more robust to leave some reserve capacity (i.e. not 99% utilization).

For the particular instance you mentioned - leaving the connection finger tight or loose may lead to other unintended issues (ERS rotation/movement).  It may be more prudent to make the connection more robust.

Jason

------------------------------
Jason Schneider
Engineer
Collins Engineers
Chicago IL

Original Message:
Sent: 01-30-2023 11:12 AM
From: Angela Hunter
Subject: Pinned vs Partially Restrained Flush End-Plate Connections

Flush end-plate connections have historically been simplified as pinned connections.  However, it is known that these connections will attract some moment.  While it may be conservative to consider that this connection is pinned for analysis of the member, this assumption unconservatively ignores any tension that is developed in the bolts from the restraint.  With many types of temporary works (for example building steel erection), I would recommend analyzing an envelope (pinned for global stability and member loads, partially restrained to check possible connection demands).  However, with support of excavation, in my experience, these types of connections are typically just analyzed as pinned and no further envelope analysis is performed.

During a recent design review for a waler/support of excavation, the EOR considered the connection to be pinned and selected A325N bolts, resulting in 99% shear utilization.  The design reviewer then ran a failure analysis (removed all safety factors and determined how much moment the connection could take, up to the combined shear/tensile strength of the bolts). The design reviewer found that the connection could safely handle <5% of the moment based on ultimate rupture strength of the bolts.  Therefore, the design reviewer recommended several changes to the design 1) change the bolts to A490X to provide increased capacity 2) install the bolts hand tight/use loc-tite (do not torque or overtighten them), allowing the connection to naturally open a bit and rotate to limit the tension that would be induced on the bolts.  These recommendations were relatively easy for the EOR to incorporate (caused no changes to fabrication and no delay to the work).  However, the question remains whether these changes were actually necessary and/or if treating these connections as partially fixed should be the new norm.

Survey to the discussion group:
1) Do you treat flush end-plate connections as pinned, partially restrained, or both?
2) For support of excavation, do you believe it is generally safe and appropriate to just treat these connections as pinned, since this has been standard industry practice and the loads for this type of structure are not necessarily uniform or occur at full magnitude?  Or, do you think it's important that engineers consider the moment that can be attracted by the connection and ensure that the connection is designed to safely handle that?

Thanks in advance for your input!


------------------------------
Angela Hunter P.E., C.Eng, M.ASCE
Teaneck NJ
------------------------------

I would definitely classify a flush end plate as some degree of moment connection by default.

In the right circumstances (thin plate, large bolts, stiff beam), it may be appropriate to analyze as pinned.
But if that was the intent, why not just detail an end plate shorter than the beam depth to get a prequalified shear end plate connection?

------------------------------
Ian Ebersole P.E., S.E., M.ASCE
Technical Lead
Foothills Bridge Co
Boulder CO
------------------------------

Original Message:
Sent: 02-01-2023 10:33 AM
From: Alan Fisher
Subject: Pinned vs Partially Restrained Flush End-Plate Connections

Excellent topic.  It makes a difference if the cofferdam is in earth or in water.  Earth loads are often over-estimated, while water loads are 100% of estimated. Load probability plays heavily on design attention. The welding of the end plate is a higher concern than the bolts, as poor welding details can lead to premature failure.  That said, I use 80% of friction bolt capacity, as a guide.

When I started designing cofferdams, the field practice was to use shiplap corners, cutting one half of each flange off each piece and lapping the webs to each other. The webs were then bolted together, loading the bolts in shear only.  This morphed into end plate connections, similar to how the strut ends were fabricated.  We tried double clip angle connections, but stopped as you have to believe they act as true pins, and the field guys didn't like the look of them.

If you do pay the price for all or partial moment capacity, remember to take advantage of the positive moment reduction in the adjoining members, or just keep it in your back pocket for when they come and tell you the cofferdam has to go deeper or will have a higher surcharge load than they told you originally.

Anybody want to talk about inny or outty sheet pile corners of cofferdams?

------------------------------
Alan D Fisher, PE
Construction Structures Engineering Consultant
Cianbro Corporation
Portland, Maine
207-553-2703

Original Message:
Sent: 01-31-2023 04:28 PM
From: Jason Schneider
Subject: Pinned vs Partially Restrained Flush End-Plate Connections

Angie,

Typically we've designed flush end-plate connections as pinned in the ERS realm (with no issues in performance), but we have run into similar partial restraint comments.  There have been a few instances where we have designed using 50% moment releases based on comments with minimal revisions to the connection details.  Every situation should be reviewed individually, but typically we believe it is appropriate to design as pinned.  However, knowing the potential for some tension to develop from partial restraint, good design practice would be to make the connection more robust to leave some reserve capacity (i.e. not 99% utilization).

For the particular instance you mentioned - leaving the connection finger tight or loose may lead to other unintended issues (ERS rotation/movement).  It may be more prudent to make the connection more robust.

Jason

------------------------------
Jason Schneider
Engineer
Collins Engineers
Chicago IL

Original Message:
Sent: 01-30-2023 11:12 AM
From: Angela Hunter
Subject: Pinned vs Partially Restrained Flush End-Plate Connections

Flush end-plate connections have historically been simplified as pinned connections.  However, it is known that these connections will attract some moment.  While it may be conservative to consider that this connection is pinned for analysis of the member, this assumption unconservatively ignores any tension that is developed in the bolts from the restraint.  With many types of temporary works (for example building steel erection), I would recommend analyzing an envelope (pinned for global stability and member loads, partially restrained to check possible connection demands).  However, with support of excavation, in my experience, these types of connections are typically just analyzed as pinned and no further envelope analysis is performed.

During a recent design review for a waler/support of excavation, the EOR considered the connection to be pinned and selected A325N bolts, resulting in 99% shear utilization.  The design reviewer then ran a failure analysis (removed all safety factors and determined how much moment the connection could take, up to the combined shear/tensile strength of the bolts). The design reviewer found that the connection could safely handle <5% of the moment based on ultimate rupture strength of the bolts.  Therefore, the design reviewer recommended several changes to the design 1) change the bolts to A490X to provide increased capacity 2) install the bolts hand tight/use loc-tite (do not torque or overtighten them), allowing the connection to naturally open a bit and rotate to limit the tension that would be induced on the bolts.  These recommendations were relatively easy for the EOR to incorporate (caused no changes to fabrication and no delay to the work).  However, the question remains whether these changes were actually necessary and/or if treating these connections as partially fixed should be the new norm.

Survey to the discussion group:
1) Do you treat flush end-plate connections as pinned, partially restrained, or both?
2) For support of excavation, do you believe it is generally safe and appropriate to just treat these connections as pinned, since this has been standard industry practice and the loads for this type of structure are not necessarily uniform or occur at full magnitude?  Or, do you think it's important that engineers consider the moment that can be attracted by the connection and ensure that the connection is designed to safely handle that?

Thanks in advance for your input!


------------------------------
Angela Hunter P.E., C.Eng, M.ASCE
Teaneck NJ
------------------------------