Substation bus structure using STAAD 1

[oengineer]

I am working on a single-phase bus support for a substation. I am trying to model the insulator in STAAD. The insulator is made of porcelain material and rest on top of a HSS column. What is the best the best way to model the insulator in STAAD so that the loads for the bus items sitting on top of the insulator are transferred to the HSS column. What type of member should be used for the insulator? Should the insulator be treated as a dummy member? Any suggestions/comments are appreciated.

 

[transmissiontowers]

I'm not real familiar with Staad but there is probably a place to specify the material properties for E for the insulator. I would put in a solid round member with the correct E for the insulator which can be obtained from the manufacturer. If you are modeling the aluminum bus pipe, you can use the properties of aluminum and the cross section of Sch 40 pipe. Many utilities put a 795 MCM AAC conductor inside the pipe for either dampening or to carry more current (I'm a structural so don't quote me on the reasons). Take a look at ASCE 113 and IEEE 605 for some guidelines for the loads to check for. If you are worried about the deflection of the Bus during a wind event, you probably need to model a length of Bus on top of the insulator to see what the deflection is. If you have a short circuit force on the bus, you will need to see if the owner wants a load combination with SC plus wind. There are equations and load combinations in ASCE 113.

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

 

[oengineer]

I am also working on base plate design for bus support structures. I am designing moment resisting base plates for HSS square columns. I am having some issues. I am aware of using 0.95 times the depth and width to determine m and n. The issue that I am having is with designing the base plate without it being supported by grout/concrete. The base plate is on leveling nuts. For this task I have been instructed to use AISC design guides. Based on everything I find in AISC design guide examples, all of the base plates are resting on concrete and all of the formulas are based on the base plate resisting on concrete to determine bearing pressure.

Would you happen to know how to design the base plate without it being supported on concrete? I have designed base plates in the past for other structures in another industry, but all of them rested on concrete/grout. ASCE 113 mentions the design on base plates and the formulas given do not mention concrete strengths.The main issue I am having with the ASCE 113 formulas is the calculation of beff value. My base plate is a square with 4 anchor bolts. The center to center bolt distances are equal. Does AISC have any provisions for base plate that are not supported by concrete? What is the substation industry standard method for designing base plates? See link

http://files.engineering.com/getfil...b16-4d16-8db4-d8695e79c25b&file=Document1.pdf

for example sketch. Would you happen to have an example of a moment resisting base plate design without grout? Any suggestions/comments are appreciated.

 

[transmissiontowers]

It depends on how conservative you want to be. Read over the base plate section in 113 to see the equations for finding the bolt loads. For 4 bolt patterns, I usually divide the moment by d1 from your sketch and by 2 to get the load due to the moment then add 1/4 of the axial load. Then you consider the bolt loads the plate and bends the base plate along some assumed bend line. If your HSS is welded on top of the base plate the effective bend width Beff will be the plate width L or W but if the plate is very wide, we limit the bend line to 12 times thickness. OTOH, if the base plate has a hole in it the size of your HSS (for galvanizing drain) you can't use the whole plate width. You will subtract the width of the HSS from W and that is Beff. You can consider 2 bolts loading the bend line and have it parallel to the HSS. Or you find the max bolt load for biaxial bending and consider a bend line on a diagonal through the corner of the HSS.

I don't do the AISC method so I am not familiar how they do it. Now, since the plate is on nuts and there is a gap between TOC and bottom of baseplate, you might consider bending in the bolts. Traditionally our industry has neglected bolt bending if the gap is smaller than 2 bolt diameters. When you consider the nut is about 1 bolt diameter and the shear acts at the bottom of the leveling nut, the moment arm is one bolt diameter (or less) and we consider that to me mostly shear loading. If you want to be very conservative, you can always consider bolt bending. There are equations in ASCE 113 and ASCE 48 for the interaction of stress in the anchor bolts.

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

 

[oengineer]

@transmissiontowers

Based on you post above, are you using the diameter of the anchor bolt for the c value shown in the ASCE 113 base plate thickness formula (on page 103-104)? Here is the ASCE 113 reference:

http://files.engineering.com/getfil...file=Microsoft_Word_-_ASCE_113_base_plate.pdf

 

[transmissiontowers]

No, the small c refers to the distance from the center of the bolt to the assumed bend line so you can sum the moments about that bending plane. Equation 6-5 solves for the base plate thickness and you have to do this for each assumed bend line and use the largest t.

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

 

[oengineer]

@transmissiontowers

Thank you for your response. I ask because I have reviewed some calculations with very low c values and it appeared that they were half of the anchor bolt diameter. I have another question for you. Would it be possible to use the formula shown in AISC Design guide 10 to determine the required thickness for non-grouted base plates?

Another question for you is have you ever used 2 inch base plate thickness for high bus, low diagonal bus, and High Switch support structures? That seems to thick to me. Also is it odd to use 50 ksi grade for base plates in a substation?

 

[transmissiontowers]

I glanced over the attachment. I'm not very familiar with the AISC guide since I always use the methods in ASCE 113 so I hesitate to give an opinion. In general, my concept of the AISC is they are very conservative because they usually deal with human life safety and adding a few hundred pounds of steel to a building is a minor thing. The electric rate payers are hesitant for us to gold plate our designs so we deal with factors of safety much lower than the building and bridge guys use. I did glean that they (AISC) are talking about base plates on wide flange columns because I saw a statement that the base plate thickness was related to the WF flange thickness. Back in the 1970's and 80's we used tapered WF built up sections and had to write our own Fortran programs to do the analysis of the column. We graduated to tapered 8 sided tubes and later we used standard HSS shapes for columns.

You are free to use whatever guide you are comfortable with and if the AISC gives you a warm feeling, go ahead and use that. The ASCE 113 methods have been used for many years and we just tried to document what we did for the young or inexperienced engineers that started doing designs in our industry from the building, bridge, or petro-chem spheres.

As far as your other questions, a 2" thick plate does seem a little high but I would have to run the numbers myself since it depends on your voltage and how high the switch is mounted. I've done 345 kV switch stands on 2 columns of tapered 8 sided columns that are pretty tall to get safe ground clearance and they had pretty thick plates on the order of 2 to 3 inches. In general, if the tube is big enough to get a welder inside to do a seal weld, you can have the interior hole in the base plate smaller than the tube so your bend line can be longer as compared to a base plate where the column telescopes 1/2 way through the plate. In the second case you cannot count on the plate where it intersects the flat of the column and you can get small effective bending planes and thick plates.

It also depends if you have some tight deflection limits on your structure. We used to limit the deflection to 4" at the switch jaws so it would operate in a wind storm. We have since removed the limit and cable connected the switch to the rigid bus runs. If your utility limits deflection, the stress in the column and thickness of the base plate don't matter much since deflection most likely controls the column size and they may have picked a plate thickness to make the column very stiff.

The 50 ksi steel is very common in our industry for plates. A36 was the standard for many years and it is getting harder to get because a lot of the steel is remelted scrap from who knows where. You can specify A36 and probably get 50 ksi anyway. The days of making steel from raw iron ore has been replaced with remelting old Toyotas and Hondas and mixing in the iron ore. There is a movement to require a Charpy test on plates so you know how brittle your plate is.

The "c" distance 1/2 the bolt diameter is odd. That "c" in our equation is the moment arm from the bolt to the bending plane which is usually along the column face. If "c" is too small, the nut cannot be tightened. Now if you are referring to the distance between the bottom of the base plate and the top of the concrete, then 1/2 of the diameter is reasonable so you can neglect the bolt bending.

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

 

[oengineer]

@transmissiontowers

Thank you for your help. When designing base plates is it common to use Complete Joint Penetration Welds to weld the base plate to the HSS column? What weld is typical for the base plates in a substation?

It is interesting to hear that 50 ksi is commonly used for base plates in Substations. Coming from oil & gas we just typical referred to AISC which calls for 36 ksi for most plate thickness.

 

[transmissiontowers]

Yes, we use a lot of full penetration welds in base plate connections to tapered 8 and 12 sided tubes when the tube is big enough to get a welder inside to do the seal weld. In my part of the country we galvanize everything so consideration for drainage in the zinc tank is paramount. We always specify inside seal welds to prevent pickling acid entrapment in the joint when backing bars are used. When I design a base plate for a standard HSS shape (or small 8 and 12 sided columns), I telescope the tube 1/2 way through the baseplate and use a fillet weld on the inside and outside. This allows the hot zinc to easily drain through the column. We also tend to not use grout on our base plates to allow water to drain out when it inevitably gets inside.

I went to a substation and saw a few columns with big rust holes big enough to stick my hand through just above the base plate which had grout under it. I banged on a few others and found about 5 feet of water up inside the columns. They rusted from the inside out and we replaced several columns and removed grout from many to allow the water to drain.

AFA 50 vs.36 ksi, we just gravitated to the 50 because it is about the same cost as the 36 and the base plates come out a little thinner with 50 ksi so the structure cost will be about the same. I don't usually design for deflection limits, but when you do, the dilemma is what moment do you use for the base plate. If the column is sized to limit the deflection, you could have a very low stress in the column base. Just for safety, I would probably design the base plate to handle the moment capacity of the tube. Yes, the cost will be a little higher for a deflection controlled structure as compared to a stress controlled structure, but if you try to save money and just design the base plate for the moment that actually exists, you may end up with a little too much flex in the connection. The running joke in our industry is "analysis paralysis". If you have the time and computer power, you can do a FEM of each connection plate and generate enough calculations to equal the weight of the structure (common in the Nuclear power industry) but at some point you have to step back and realize the amount of analysis work may not bring much savings to the structure.

HTH

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