Technical Talk

Even complex analysis methods can be made simple if the tools are available. In this section several examples are shown of the finite element method (FEA) being used in industry to solve problems quickly.

The simplifying techniques used may not appeal to the engineer whose primary responsibility is for new designs or providing 3D models to customers.

They are useful to the engineer whose job it is to understand what has caused a failure and needs to explain it to management and get it fixed quickly. In this respect, it is like the simplified analytical analysis described in the articles and text shown in the Technical Papers section.

Technical Talk 1

This is a CFD (Computational Fluid Dynamics) analysis of a reactor with a restrictor head plate. This 2D simplified analysis shows the high velocity (red) in the 1 inch gap region. It also shows the low flow in the inner tube. Increasing the gap (c) reduced the high pressure, which would have overloaded the propeller type pump, and also provided better distribution to the tubes. Notice the simplification here. If you sweep the tubes you have channels not individual tubes, so the results are only approximate, but did solve the problem. Sometimes the engineer has to be creative in order to make timely decisions. Case 61.

Technical Talk 2

This is a CFD of a pick-up truck at 60 MPH. It’s a 2D model as if the truck had been sliced down the middle. The velocity vectors down the center are shown. This helps to explain the recirculation you see in a pick-up bed as it travels down the road. The leaves stay circulating in the bed for a while. Also shown is the low pressure area at the rear of the truck and why “drafting” occurs in racing, meaning following close behind to save fuel. While this is for a truck, the reader should be able to visualize other places this type analysis would be useful. One example would be flow in ductwork or piping systems.

Technical Talk 3

This represents a ¼ axis symmetric coupled thermal - stress analysis of a screw compressor rotor. The rotor was over heated for a short period of time and it was requested that an FEA be done to see if rotor tip contact was possible due to rotor growth. It was. Case 46.

Technical Talk 4

This is a cycling valve spool piece that developed cracking over several years. An axis symmetrical thermal-stress analysis verified that it was due to the thermal cycling stress. Notice how simple the models is and that the steam jacket didn’t have to be included, only the boundary temperature of the steam. Notice how cracking only developed in the high temperature zone where the temperature cycled from -40° to +250°F. Case 44.

Technical Talk 5

A 50 ton extruder gearbox was moving on its base, even after careful alignment and bolt tightening. The simple “block” thermal-stress FEA was performed. It showed that the fouled cooling passages in the barrel caused barrel distortion. This resulted in a thermal moment, which would move the gearbox. The analysis was verified with laser measurements and thermal imaging. A better water passage cleaning schedule and temperature instrumentation solved the problem. The picture shown is not the actual extruder analyzed but is similar. Case 24.

Technical Talk 6

This transient thermal-stress FEA model was used to determine if the changes in temperature at each stage were enough to have caused the rotor to rub against the diaphragms as was witnessed. It was determined that the cold assembly clearances had been adequate and the rubs were due to another cause. In this example the model was axis symmetric but was swept through 180° for the presentation to management. Case 19.

Technical Talk 7

This is a complex heated mixer that had developed cracking on its diaphragms. While a 3D FEA could have been built, it would have taken too long to implement fixes on larger new units, which were about to be installed. The simple (full length not shown) axis symmetric FEA model with all the loads and temperatures included provided answers on how to modify the mixers with a minimum amount of modifications. The 2D model was easy to explain to management and changes could be evaluated quickly. Case 45.

Technical Talk 8

 

 

 

 

 


This was a study which reviewed the use of a thicker one piece polymer windscreen on an aircraft. Bird strikes can cause damage such as shown, so the FEA stress analysis looked at the stresses produced using different thickness windscreens.

The analysis required the impact loads to be calculated analytically and applied to the 3D FEA model. With large bird impacts it was impossible to keep the windscreen from failing no matter what reasonable thickness was used. The solution is to stay away from birds. Book Information, p.40



Technical Talk 9

Case 65: Taking Risks and Making High Level Presentations

As engineers we like to limit our risks. As a general aviation pilot and a mechanical engineer this has served me well over the years. I didn't do things that were too risky and always had a couple of alternate plans in case something went wrong. My requests for a design modification were always supported with adequate calculations. I have noticed that when someone has performed a reasonable analysis their arguments usually carry more weight than those who are speculating on the cause with no supporting data.

There can be problems with this approach that I learned early in my career and had to adjust for.

The first is that there is always some risk involved in every engineering decision, and you cannot progress far in your career if you are unwilling to take calculated risks.

Consider a large steam turbine vibrating slightly above normal levels with blade fouling thought to be the problem. Management wants to know if they can run one week until a planned outage with that level of vibration. Your career will not be enhanced if you say it has to be shut down immediately, with no supporting data. Likewise this is not the time to try your first attempt of on-line washing of a steam turbine while it is in operation. This is a risky business if you have no experience and operating guidelines to follow for this procedure. This would be a good time to monitor the vibration level, talk with the manufacturer and others with similar machines and determine the risk in just monitoring the vibration levels. Defining at what level it will have to be shut down will still require some risk, but now others are involved. Obviously there is much more but this illustrates the need for some risk. You can expect to make some judgment errors but they should not occur early in your career or consecutively.

Correctly communicating to management what has occurred and what needs to be done is so important to an engineer. It would be wonderful if engineers had the verbal ability of attorneys in presenting data to management. An attorneys job is to make juries feel comfortable with what they are telling them, the decision that has to be made with the evidence and data they have. Unfortunately many of us don't have this type training. Fortunately it can be learned by experience and watching other successful engineers. Your company's senior technical personnel didn't get where they are via a lack of communication skills or poor judgment.

There are three things I have found important when discussing work with senior management. The first two items are self-explanatory but the last will require an example.

1. Management does not like to hear bad news, so present positive plans.

2. Management does not want to hear a wish list of solutions so present only your best and most cost effective choice.

3. Management is not impressed with complex analysis or technical terms so the engineer must simplify the cause, solution and implementation so it can be understood and acted on.

The management you are presenting to may not be familiar with mechanical engineering and may have their expertise elsewhere. It is useful to adjust what you are presenting to suit your audience.

Suppose you are discussing the resonance of a structure and its failure. Now as engineers we know resonance can be a highly damaging vibration caused by exciting a structures natural frequency. Resonance could be clearly demonstrated by bringing a tuning fork to the meeting, striking it and showing the resonance of the tuning fork. With no continued exciting input, meaning striking it, the vibration dies out. However with a continued input and without material damping the tuning fork would fail in fatigue. The fatigue failure part could be demonstrated by bending a paper clip back and forth to demonstrate and explain fatigue. At any point you can stop bending but some of its life has been used up.

When reviewing a technical presentation, I always make a list of all the questions I think may be asked during the meeting and research them thoroughly. This allows the best possible answers for the audience I�m presenting to.

This short section is certainly far from being complete but it may help in assessing risks and your presentations to management.