Transformer Seismic Qualification 3

[shepherd]

We have a potential client that wants assistance in providing seismic calculations to qualify their transformer in a nuclear plant application. I am a little unclear (as I think they are) of what the scope of work actually is. My understanding is they are referencing IEEE standard 344 "Recommended standard practice for seismic qualification of class 1E equipment for Nuclear Power Generating Stations", this code is referenced by ASCE 7 section 9.6.1.1. In this section, ASCE 7 prescribes a static force procedure for analysis of the structure. To anyone who has performed this type of work, has your scope been limited to analyzing the transformer housing structure and providing anchorage forces, and have you used this section of ASCE 7 for analysis?

Thanks for any help you may be able to provide

 

[Bagman2524]

There are 2 types of seismic qualification in the nuclear world. One is called a "2 over 1" where a non-safety related component must be ensured to stay in place and not fall onto a safety related component during/after the seismic event. It is not necessary for the component to remain functional. For such a case typically only the anchorage has to be verified to be strong enough to hold the component in place. The other type of seismic qualification is where not only must the component remain in place, but it must continue to function after the seismic event in order to ensure safe operation/shutdown of the plant. Seismic qualification is typically either by calculation or seismic "shake table" testing. Wyle Labs in Huntsville Alabama does this sort of thing and there are one or two other companies as well (Fallwell & Hendricks?). Where I came from we had what where called Required Response Spectra (RRS) curves (of period vs acceleration)covering the different buildings and elevations in the plant. These curves will vary based on where in the country the plant is located. For the shake table test these curves are provided to the tester and they shake a prototype of the component to determine its' response during the seismic event. To qualify the transformer by calculation you would have to determine the natural frequency of the transformer, use the RRS curves to determine what "g" force to apply to the transformer. I've never seen electrical components qualified by calculation, always by testing. Valves & structures are typically qualified by calculation.

 

[JAE]

Nice post Bagman

 

[shepherd]

Thanks Bagman - From the brief read I've done of the IEEE standard they do allow Certification by analysis, and that is what these guys are looking for. They have performed shake table test in the past on similar components, so If I was required to perform a dynamic analysis hopefully I could get a feel for the damping ratio to use from those past tests. Do the nuclear facilities typically have their own response spectra curves to work with, or has your experience been they adopt NEHRP maps to develop the design response spectrum? Sounds like I may need to pass this on to someone more familiar with the procedures and stick with the typical commercial work!

 

[mrMikee]

Several years ago I worked on a proposal that required compliance with the ASCE 4-98 Nuclear Design spec and with a special response spectra provided by the contractor (a large national consulting firm) for the actual location of the equipment. This spec may or may not be what you need but I would carefully research who is the governing authority and specifically what is required.

-Mike

 

[Bagman2524]

If you haven't done any nuclear seismic qualification in the past, I would suggest turning it over to someone who does this sort of thing on a regular basis. We had some unexperienced people try to do this sort of thing for us and it just bounced back and forth several times before they ended up going to someone familar with the procedure. If you just need a 2 over 1 qualification you might be able to do it yourself by calculation. I would find out from someone in the Civil Engineering department at the specific nuke in question about what seismic RRS curves they use. I think most plants have their own RRS curves. Thanks JAE, I done this more than I would have cared to. Just another note. Most Nukes are stuck in a time warp. They use the specs including the year that the plant was licensed to. When a spec such as AISC gets revised, they do not switch to the new revision.

 

[oldrunner]

We've done both the anchorage anlysis and the internal components analysis for transformers per the IEEE code. It was a static analysis for a UBC Zone 3, which is basically .25g horizontal in each direction and .20g vertical, simultanuously.

In one case, the transformer was already on line and they had to shut it down, drain the oil and were able to change out some of the bracing and bolts inside the can. It gets to be fun adding in the sloshing forces.

We used Risa3d.

 

[Bagman2524]

The Nuclear Industry doesn't use any of the standard building codes like ASCE-4, UBC or IBC for seismic design. I think the IEEE-344 standard mentioned above is what applies. The sesmic forces required are typically higher for than UBC,IBC etc. The anaysis cited by oldrunner sounds ok, it would just have to be per IEEE-344 seismic forces not UBC.

oldrunner, was this only a structural analysis of the transformer structure and anchorage? Does this assume if the structure remains intact then the transformer continues to operate normally?

 

[oldrunner]

Bagman2524:

Actually the required code was IEEE 693, "Recommended Seismic Design for Substations". You are right, zones are not really identified, but IEEE 693 has only two force levels for this type of project and California has only two different seismic zones. The project is located in Zone 3 so the lower force levels were required.

Yes, this is only for the transformer. It is interesting that California does NOT require, at least for these new plants which are using jet turbines, that the plants to be designed as "essential" facilities with the 1.5 increase.

We worked on a plant at Klamath Falls, Oregon, and Oregon DOES consider the plants as essential facilities.

Just yesterday I attended a technical meeting put on by another engineer and he brought up that point concerning California not requiring the essential facility designation for these plants.

 

[kucukerdem]

We are preparing for a transformer tender (25-500 kVA rating 11kv/0,433kV ONAN distribution transformers) and it is written (in tech. requirements) that "transformers will stand earthquakes 0,5g in horizontal and 0,25g vertical limit. manufacturer will supply needed equipments and calculations." Anyone knows the procedure, calculations etc. can help me please? Thank you

With Respect...

 

[Bagman2524]

I assume you mean 0.5g and 0.25g not 5g and 2.5g.
0.5g and 0.25g are not extremely high seismic accelerations; my guess is your transformer should pass fairly easily.

The original post indicated this was a nuclear application, but subsequent posts seem to indicate its' a gas turbine power plant.

You should get either a seismic "shake table" test or a set of calculations by the engineering firm that use a lateral load of 0.5 x W and a vertical load of 1.25 x W to demonstrate the transformer works during/after the earthquake.

 

[kucukerdem]

Thank you Bagman2524. Your assumption is correct I also considered that it is not a high acceleration value and it is negligible for these transformers. But you know it is the client and wants calculations. Have you ever experienced any kind of calculations or know any literature about the topic?

With Respect...

 

[oldrunner]

kuckerdem:

At least two or more of the preceeding posts have pointed you towards your answer.

More specific, review the Annex R Analysis report template in the IEE Std 693-1997, which will provide you with almost all the questions that you are asking.

An internal analysis of core coils and other components as well as the structural stability members may suprise you as to their adequacy. Some of the structural components may not work. The two different companys that I have provided analysis for only set the internals on the bottom of the tank; they are held in place laterally by studs. To resist the overturning (usually the short direction), there is some sort of bracketing to the side walls. All of this has to be analyzed and they might not be adequate as one of my customers found out. (These were using the lower seismic values!)

Until someone in your firm has actually performed at least the static analysis, you comments concerning "negligible" is possibly expressing an engineerng opinion outside of your expertise.