Design of Hydraulic Steel Gates. 1

[21121956]

Hello everybody:

In a general manner, a normal procedure for the design of hydraulic steel gates is by performing, as a first stage, a hand computational design and then, a software based design, by building up a 3-D model for the gate.

In one word: Design by hand and check with software based model simulation.

With the aforementioned procedure and if there is no any objection, the gate is ready for fabrication.

But, inevitably, one question comes to one's mind: when will be necessary to fabricate a model (a gate at reduced scale) and perform on it laboratory tests, in order to simulate a wider field of operation of the gate under a variety of conditions that that gate should afford under normal operation.

So, for the manufacture can be considered sufficient the CFD simulations or, under what circumstances it is required to ask to the manufacturer to conduct laboratory tests on the model?

Thanks in advance for your inputs and comments.


El que no puede andar, se sienta.

 

[MintJulep]

Phrased more broadly: When is it necessary to validate calculations and analysis by physical tests?

Generally, the answer is related to how complicated is the design, is the design unique, how confident are you that the model and analysis is correct, and what are the risks associated with the analysis being wrong?

Things to consider:

Is the design completely new? An adaptation of an existing and demonstrated successful design, a scale-up or down of an existing design, etc.

How complex is the design and what simplifications were made in the FEA model? Are the simplifications reasonable? Are the simplifications in high-stress areas, or in fatigue sensitive details?

Are the modelling techniques and simplifications the same as used on a past successful design, or on a design that has had its model validated by test?

How accurately are the in-service loads known, or alternatively how conservative is the design?

 

[dhengr]

21121956:
I agree with MintJ’s comments above, I wouldn’t expect that most run-of-the-mill gates would require small scale modeling and lab testing, the Structural Engineering should be fairly straight forward in most areas. Although, I would want an experienced designer and Structural Engineer involved in the design of the gates. Certainly, in today’s world one would expect that the final design would be carried out with computer analysis. However, I would also expect my long hand analysis and design to be pretty close to the final computer design. I would use the computer and FEA for the details and the fine points. If I have changed support points, or support and movement (lifting, swinging) mechanisms, I would want to study how they integrate with the overall structure, and I would as likely as not do that by hand too. I would probably do the welding design by hand too. I tend to use the computer to fine tune my design and to look at particularly complex conditions, not to show me how the basic structure actually works, I pretty much know that from experience.

 

[3DDave]

Funny story - Before awarding a contract to Perkin-Elmer, NASA requested they produce a demo mirror, a scale model. They did this with exemplary results. When the large mirror contract came in, they did not use the same team, turning it over to an entirely different group.

It's a factor I've seen before. The prototype works great because first string guys are involved and then the production is turned over to significantly lower skilled or more poorly equipped facilities which produce poor results.

 

[21121956]

Hello everybody:

I hear you loud and clear. First of all, thanks for your comments.

The type of steel gates we are talking about, are installed in hydropower plants around the world and their design engineering is a mature one.

The fear I feel is that, as the sole support for trusting the manufacturer is only his reputation for having built steel gates for other projects and a list of those projects.

But, how about the situation pointed out by 3DDave?

Apart from its name and reputation, what more one can ask the manufacturer in order to rely on his supply?

Once again, thanks for your inputs.


El que no puede andar, se sienta.

 

[MintJulep]

You ask the manufacturer for the locations of installations and a person to contact. The person needs to be a useful person. The head of maintenance or operations for example. Not a buyer.

Follow up. Call the person. Ask if the gates are good. Ask if service is good. Ask if there is anything that should be changed or made better. Ask if they have modified anything.

Ask the manufacturer for the names of the individuals who will be designing your gates. Ask for their resumes. If at least some of them have not worked on past gates insist that you have someone with actual and relevant experience in change of the design of your gates.