Structural Design of High side A-frame Deadend Str

[tildeme]

I have been working for my firm for ~2+ years and am most familiar with using GTStrudl and have been using this program for all my analysis models.

The engineer I am working under is also a Civil Engineer who has been in the industry for 30+ years and haven't touched GTstrudl in ages. Thus the coding/modeling in GTStrudl has been left up to me. With the help of NEMA SG-6 and various GTStrudl notes, I've managed a lot of the analysis of various modeling of substation towers and structure. Now I am working on my first design of an actual structure so I've more stringent in the loads I am considering for the design.

My question to structural engineers in the transmission field is how do you take into account earthquake loading in your models? Is there a design guide or standard you take into account?

Do your design take into account notional loads? 13th Edition AISC, Appendix 7, page 16.1-196

Our company usually contract out these kinds of work and I just usually check another engineer's calculations. So I do have a reference for how other engineers does their design. Except it's usually in Staad, which is slightly different than GTStrudl. I notice they don't normally take earthquake into account either.

Granted, I know my boss is going to look over my code with a fine tooth comb but I to find someone who is familiar enough with GTStrudl coding to help me out. Thank you for your time. This forum has been very useful to me ever since I started my career with my firm.

 

[stevenal]

Can't help with most of your questions, but for a seismic standard check out IEEE 693.

 

[MiketheEngineer]

Not sure if there is a specific code for this - but the work I have done on buildings usually calls for a horizontal component any where from 0% to 33%. You simply add that to your load combinations.

Good luck.

 

[tildeme]

stevenal, thank you. I will look into IEEE 693.

MiketheEngineer, this sounds very familiar and I think this might be what I will have to do. For my senior design project, we added 15% of the dead load to all the joints in a horizontal (x and z axis) to create an equivalent earthquake loading. Now that you mention it, I think I may have seen other models of substation deadends modeled like that too. Thank you, I will consult my boss to see if this will be appropriate.

 

[MiketheEngineer]

You might check the IBC2009 code to find the forces based on location.

BTW - Most of these structures are so "light" that there is little impact. But it must be checked.

 

[hyphenme]

Agreed. Most of those structures are light that they might not be controlling load case but I do still have to check it just in case. Thank you.

 

[transmissiontowers]

Look at ASCE 113 which has some seismic suggestions for the design of substation structures. There are loading combinations to consider with seismic, although I don't recall what they are right now. I helped write and edit the manual but I never have to consider seismic for my utility.

My company also uses GTStrudl for substation design. I usually get the conductor tension and wind loads from PLS-CADD. If your frame is not too complicated, you can model it in PLS-POLE.

If you mount wave traps or line switches with porcelain insulators on the dead end, you might have seismic problems.

Do you attend the annual GTStrudl User's Group meetings? I have been going for around 20 years.

_____________________________________
I have been called "A storehouse of worthless information" many times.

 

[hyphenme]

transmissiontowers, hello. I am also narfzz. I made a different username since I didn't realize that "structural" was also an option since "civil/environmental" can be confusing since I deal more with structural.

Thank you for your suggestion of the ASCE 113. I will look into it. Right now the loading combination I am using is DL + EQ in Z and X using around 15% of self weight. We don't have wave traps or porcelain insulators. As far as I know the we have a 161kv switch on top, lightning arrestors, and post insulators on the structure.

My frame is really simple. The problem with using PLS-POLE or TOWER is that it doesn't model W shapes. We considered purchasing TOWER but since it doesn't model structures with W shapes, we ended up using GTStrudl since that was what I am familiar with from college.

I have gone to the GTStrudl hands on training but not the GTStrudl User's Group meetings. I might consider going this year if it is helpful and informative, especially if it's relevant to our industry. I will suggest it to my boss especially if you've been going for 20 years! Do you highly recommend it?

 

[transmissiontowers]

I might consider the seismic to occur during a cold event to get the wire tensions up higher, but that is up to the utility. The 161kv switches, arresters and post insulators all use porcelain which does not behave well in an earthquake and IIRC, they have a large mass.

GTStrudl is the best way to model the WF legs and beams and it will handle the dynamic loading.

The UG meetings are fun and you get to meet other users and see the many uses of the program. You also get to talk to the developers and ask any question you might have. The conferences tend to highlight the building and bridge industries because there are more users in those industries. There is also nuclear power plant related presentations since GTStrudl is one of the few QA/QC programs approved by the NRC.

Take a look at the GTStrudl web page to see previous UG presentations.

http://www.gtstrudl.gatech.edu/

Take a look under presentations to see some of the topics presented for the last few years.

_____________________________________
I have been called "A storehouse of worthless information" many times.

 

[transmissiontowers]

The next UG meeting is in Seattle.

http://www.gtstrudl.gatech.edu/gtstrudl_new/GTSTRUDL_Users_Group_Meeting/Meeting_Invitation_2012.htm

_____________________________________
I have been called "A storehouse of worthless information" many times.

 

[bacon4life]

This book on the Chi-Chi, Taiwan, earthquake of September 21, 1999 might help.

http://books.google.com/books?id=uA...eta earthquake transmission line ieee&f=false

 

[JWKilgore73]

Right now the loading combination I am using is DL + EQ in Z and X using around 15% of self weight. We don't have wave traps or porcelain insulators. As far as I know the we have a 161kv switch on top, lightning arrestors, and post insulators on the structure.

Don't forget that the wire tensions are always there. And unless you know the exact location of the nearest tower/pole, you'll need to vary the angle on the conductor tension forces.

If the "design" wire tension (at whatever temperature is required by the utility) is too conservative, remember that as the structure deflects the tension in the wire goes down. Start with the design wire tensions + seismic, calculate the deflection (the full deflection, not the "R" reduced deflection) and use that to re-calculate the wire tension. Use the new wire tension to calculate a new deflection, and repeat until convergence.

As stated earlier, because of the relatively high wire tension loads and light dead loads, seismic forces rarely control the design. But if you are in a high seismic zone what "controls" isn't the problem, it's the code-required minimum designs for connections that cause problems (can be up to 110% of the attached member strength).