ASCE 136 - Turbine Foundation design 2

[bkal]

I am going through this manual(ASCE 136) to get some understanding of the turbine foundation design. One of the recommendations (3.3.1.6) for an elevated space-frame pedestal foundation is that the total weight of concrete is 3.5 times the machine weight. I am guessing that this is to do with its response to harmonic loads from the turbine/generator; is that correct? This seems to apply to a monolithic tabletop-columns-basemat. Is there a similar rule if the tabletop is isolated from the columns by vibration isolating devices on top of the columns? Or is it not required as the tabletop is isolated from the foundation.

 

[JoshPlumSE]

I am guessing that this is to do with its response to harmonic loads from the turbine/generator;

That's my impression as well. However, some of it is standard of practice / rule of thumb. You want to throw enough mass in there to make it more difficult to excite. And, I believe 3 to 5 times the machine weight has been a rule of thumb for a long time.

Is there a similar rule if the tabletop is isolated from the columns by vibration isolating devices on top of the columns?

We're talking about large steam turbines for electricity production, right? I've only dabbled in these projects, but I don't think I've ever seen one where the table top was isolated from the columns. I've seen that with relatively small machines, but never with the mega turbines that I thought this ASCE manual of practice was supposed to cover.

 

[JedClampett]

In my past career the company I worked for designed power plants. The turbine foundation was, of course, a major item. They were massive, with 10 ft. square legs tapering up to 10 ft. by 20 ft. at the slab level. The "table" was proportionally thick as well, like 10 ft. They were either low tuned, with the first mode natural frequency about 1200 rpm and the second mode about 3000 rpm and high tuned, where the first mode was about 2400 rpm. Of course, the high tuned ones were much bigger.
For power plants, a dynamic analysis was required. We couldn't trust a rule of thumb. The turbine suppliers were very quick to blame the design firm for any vibration issues and walk away.

 

[bkal]

Thanks very much for your prompt response Josh. Yes, I am talking large steam turbines. I looked into GERB web site and they have provided isolation devices for sets up to 1000MW and more. It seems that out of a very large number of installation only two are for the stations in the USA (Vogtle and Virgil C. Summer.

 

[bkal]

Thanks Jed. I imagine that a full blown dynamic analysis would be required.

On a slightly different note, an ASCE publication from 1987 (Design of Large Steam Turbine-Generator Foundations) touches upon steel foundations (tabletops) and states that they are not as widely used as concrete. On the other hand, it seems that their construction duration/costs are much lower. Searching the internet has not shown given any examples of steel turbine foundations.

 

[EdStainless]

I have seen steel used with smaller (CC) turbines but never large ones.
I would guess that damping is the issue.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[DayRooster]

Jed is correct. Also he is describing a concrete table top not steel (based on his sizes). In addition, he is correct that a full scale dynamic analysis needs to be performed. For reference, the initial sizing of the mass is like step 1 out of 12. So much detailed analysis is required to get it right. I have seen multimillion dollar lawsuits when it was done incorrectly.

If you are out of your comfort zone then I would recommend contacting an expert. Below would be my recommendation if you do decide to obtain assistance (which I strongly encourage based on the questions you are asking).

https://www.eng.uwo.ca/civil/faculty/el_naggar_h/index.html

 

[bkal]

Thanks everyone. I will definitely seek expert's advice in case I get involved in the design. At the moment I am just trying to get my head around key issues involved.

 

[WARose]

Agree with Jed. A full blown dynamic analysis is required. And try to minimize amplitude at the source. With that frequency, it would be tough to fix away from it.

 

[SlideRuleEra]

1)I looked into GERB web site and they have provided isolation devices for sets up to 1000MW and more. It seems that out of a very large number of installation only two are for the stations in the USA (Vogtle and Virgil C. Summer).

2) At the moment I am just trying to get my head around key issues involved.

1) Both Vogtle (in Georgia) and V.C. Summer (in South Carolina) are deeply troubled recent projects designed by a company whose engineering capability is questionable. To put it mildly, I suggest not using either project as an example of "success".

2) A tradition turbine-generator pedestal, as Jed described, looks like the illustration on the cover of ASCE 136:

This type image gives the illusion of a solid "table top", which is misleading. Because of sizeable penetrations for equipment, the pedestal is shaped more like "Swiss cheese":

The dynamic analysis should include the legs, the foundation, and probably any piling. A high-tuned pedestal is superior to low-tuned.

Take DayRooster's advice... this is a job for skilled specialists.

 

[Enable]

I'll second DayRooster's recommendation. I studied at UWO and did my thesis on stone column foundations for wind turbines under Newson, but had access to Naggar and others. Him and the entire geotech bunch at UWO were absolutely fabulous. I also miss the facilities very much (we had some really cool stuff!)

 

[EdStainless]

The Nuc plants are also different animals because of steam conditions.
There is no superheat so there is no HP as such.
And sometimes the steam flows are so large that you IP, 2xLP1, 2xLP2.
Three zone condensers are not uncommon in Nuc, while multiple zones are rare in other plants.
The dimensions for the same power rating are much large.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[JedClampett]

Tip (from my old company):
Embed survey points in the vicinity of each foundation column. Survey the points using very accurate levels before the turbine goes into production. That way you have a record of the plumbness of the foundation. As I said before, if the turbine generator vibrates, the manufacturer will immediately say (without any backup or evidence) the foundation settled, not our problem. So we go out there, show there is no settlement, fix your system. Without the survey points and initial survey, you'd have to prove the foundation didn't settle without backup and meanwhile, your owner has a very expensive turbine they can't run.

 

[WARose]

Embed survey points in the vicinity of each foundation column. Survey the points using very accurate levels before the turbine goes into production. That way you have a record of the plumbness of the foundation. As I said before, if the turbine generator vibrates, the manufacturer will immediately say (without any backup or evidence) the foundation settled, not our problem. So we go out there, show there is no settlement, fix your system. Without the survey points and initial survey, you'd have to prove the foundation didn't settle without backup and meanwhile, your owner has a very expensive turbine they can't run.

That's a good point about putting in permanent control points/monuments.

 

[bkal]

Thanks everyone.