Design Projects

The rest of this post will reference engineering school projects.  However, I believe that most of this is also valid to open ended real world projects.

Many engineering schools require semester design projects.  The design project’s goal is to demonstrate to your professor that you learned the key opints of the class.  Typically, demonstrating these core competencies requires being more explicit and thorough in your analyses than you are naturally inclined to be.

The biggest mistake most people make is to assume the “real” work comes late in the process.  It doesn’t the real work in any successful project is done up front.  Serious effort must be given to the choice of projects.  Even more thought and effort must be put into determining the essence and purpose of the project.  The rest of this post will explore this further.

Design Project – the purpose…

The purpose of any design project is the design of a product which can manufactuered.  But what is the real purpose of the end product?

One common mistake is to assume that your end product must be everything to everyone.  For example, in my senior design project (years and years ago) we designed hardware to help movie studios move their sets around the studio more safely and with fewer people.  The first inclination for the group was to design the hardware be able to move any set the studio might build.

Ultimately this turned out to be ridiculous.  We couldn’t levy requirements back on the studio so we couldn’t keep the studio from attaching delicate items to the set right where we needed our clamps to grab the set.  The end result being that we had to scale back our ambitions such that we were capable of picking up the majority of the sets.

A more subtle mistake, related to this one, is that if 90% of the sets were less than 400 lbs but the remaining 10% could be as heavy as 1000 lbs do you let the 1000 lb sets drive the design?  The answer most of the time is no.

This is why determining the essence of a project is so important.  In industry this is why is takes months to write requirements.  The cost of a design can be driven up and up by any single requirement.  Obviously you’d like to avoid having the cost of the design be driven by a requirement that doesn’t need to be as demanding as it is.  In other words, do you really design hardware to meet the 1000 lb sets if hardware to move 400 lb sets is half as expensive?  or requires half as much analysis?

The lesson…

The lesson in all of this is to boil down the project to its essence.  The previous examples essence was the ability to move most sets more safely with only a couple of people instead of the previous ten.

Since the essence is most sets not all sets we were able to keep the design simple with C-clampes, square tube frame aluminum, and a manually operated jack for lifting the set.  Had our design required the ability to lift 1000 lb sets then we would have need to design the hardware with an electric jack instead of a manual jack.  We also would have had to add twice as many C-clamps meaning twice as many attachment points.

If the design had been for all sets such that we had to avoid delicate parts of the set then we also would have needed to design the hardware with a telescopic arm for the C-clamps.  This would have required a wholesale change to the design for allowing the telescopic positioning and then fastening down those telescopic attachment points.

The argument among space enthusiasts…

Space exploration funds are very limited and have been for decades.  I haven’t worked onsite at NASA for years but the last number I remember is something like $15B for NASA’s annual budget.  This sounds like a lot but with the better part of $1B devoted to each Shuttle launch $15B doesn’t last long.

So the classic argument among space enthusiasts is should we bother with human space flight?  It’s far more expensive than robotic space exploration.  Can humans bring enough extra to the mission that it is worth the extra expense and risk?

The risks of human space flight

The most obvious risk to manned space flight is a fiery and most spectacular death.  It is most likely to happen on launch or re-entry.  If we start landing on the surface of other planets then that landing will be risky as will any surface exploration.

When we lose a space vehicle meant for manned space flight we lose people and an extremely expensive vehicle.  The people, at least to date, are very well trained, very intelligent people.  Often they have PhDs in engineering and years if not decades of training.  The vehicle is extraordinarily expensive because of its ability to support human life in space for any significant time duration.  Additionally, there are redundant systems and user interfaces.

Robotics, the cheap and (largely) risk free way to explore space

Robots are cheap.  Robots that are lost in exploration don’t have crying widows; there is no lost human potential.  And robots like Spirit and Opportunity offer glimpses into greatness.  Robots designed to survive months have run around the planet for years despite “injuries” and degraded solar panel performance.

I don’t think proper space exploration can be done without robots.  They allow us to explore without risk to human life.  They provide us with information beyond our own senses.  We spend billions of dollars to send people without spending an extra couple of hundred million on support robots?

Robots are essential but exploration with robots alone misses the point

If I had my druthers I’d work on rovers like Spirit and Opportunity.  I’d work on adding lots and lots of artificial intelligence to them.  How much more area would Spirit and Opportunity cover with smart fault detection and robust automatic use of remaining resources?  How much more area if we could say go here and leave the obstacle avoidance to the rover rather than having to move it a little at a time and wait for the time lag of signals coming back from Mars?

The purpose of exploration is the expansion of the human spirit.  Economically we make exploit the resources of the lands we explore.  From a survival standpoint the farther we spread the less likely we are to be wiped out by war, an asteroid or plague.  Exploring with robots alone doesn’t accomplish any of this.

MEMS Gyro models

MEMS gyroscopes are becoming common in Aerospace systems.  They are small, low power sensors accurate in frequency ranges good for Aerospace applications.  Often, MEMS gyros are the only sensors commerically available that provides the necessary frequency response, mass, power and environmental.

I’ve found 2 types of MEMS gyro modeling.  Both of these modeling types are for the design of the MEMS gyro.  A MEMS gyro sensor requires design of some key parameters - resonant frequency, driving frequency, and quality factor.  These articles are not on the frequency response of the sensor.    The frequency response and noise are the primary items to model for control systems.  So these design articles are high fidelity models and information purposes.

Traditional modeling of MEMS Gyros

Traditional design modeling of MEMS gyros often starts with an FEM of the sensor.  However, the FEM is often too large for feasible modeling.  FEM modeling can be infeasible for memory reasons or simply the length of time it takes to produce results.

The next step in traditional design modeling is to create an equivalent electrical circuit for detailed analysis in various software packages.  Again producing results from these equivalent circuit models is time consuming.

Wiki article on Traditional MEMS Gyro modeling

Simplified lumped parameter model for MEMS Gyros

I found a journal article describing a lumped parameter model for MEMS gyroscope design suitable for running in Simulink.  The  benefit of the Simulink lumped parameter model technique allows for much faster MEMS gyro design results through simple gains and trnsfer function blocks.  The results present in the journal article looked encouraging.

Wiki article on Simplified lumped parameter model for MEMS Gyros

More articles coming…

Accurate sensor models are necessary for any good control loop design.  So I have a couple more sensor model/design articles coming.  After that I will start adding details of MEMS gyros as I find them on the web.

Sensor Fusion

As discussed in the previous blog entry, sensor fusion is used to create one good sensor from at least 2 sensors that are not good enough to meet specifications.  This can be done simply but when real sensors are involved it can also become a bit of a black art requiring a lot of skill and experience.

Simple Sensor Fusion Example

The wiki has an example showing the details of simple sensor fusion.  The simple example has a low frequency sensor with a bandwidth of 20 Hz.  It also included a high frequency sensor with a lower bandwidth of 1 Hz and an upper of 1 kHz.  The sensors are blended using a second order low pass and high pass filter.  Both filters have a bandwidth of 15 Hz.

I hope to eventually create another example with better filters.  I also hope to create another sensor fusion example for sensors with non-ideal transfer functions.

Sensor Fusion or Sensor Blending

Sensors are what provides feedback to a closed loop system.  Sometime you can’t get the sensor characteristics you need.  This happens a lot in the aerospace industry.

When any one sensor cannot provide the necessary feedack then it is time for sensor fusion or sensor blending.  The simplest form of sensor fusion is a matter of two or more sensors which are filtered so that their strengths (good responsivity and low noise) are used while their weaknesses are filtered out.

Often times sensor fusion is nothing more than simple second order low pass or high pass filters with their outputs added together.  This simple fusion allows for two sensors to provide the desired output.

Simple Example of Sensor Fusion

The most simple sensor fusion that I’ve come across is the combination of two angular rate gyroscopes.  The low frequency gyro was good out to a frequency of approximately 20 Hz.  The high frequency gyro was good between 1 and 1000 Hz.  Unfortunately this system was sensitive to frequencies around 5 Hz.

Normally the blending frequency of the sensor fusion would have happened between 1 Hz and 20 Hz based on an analysis of each sensor’s noise and responsivity.  This example system was sensitive to frequencies around 5 Hz which meant that we needed to avoid frequencies between 0.5 Hz and 50 Hz.

The main weakness of the high frequency sensor was phase loss below 1 Hz.  So we designed a filter to extend the low end of the high frequency sensor down to 0.5 Hz.  More difficult to implement than to conceptualize but it takes some practice to do it correctly.

Ideal Sensor vs. Real Sensor

The ideal sensor is typically modeled with a second order system that has a natural frequency equal to the spec bandwidth and a damping of 0.707 or 1.  I default to 0.707.  This leads to a nice flat, unity response for the sensor below the bandwidth.  Real sensors are non-unity below the bandwidth – i.e. the magnitude has some ripple to it.  Sensor ripple around the blending frequency can be very problematic and must be assessed based on the system needs.

Introduction to Sensor Fusion on the Wiki

Here is an article on the wiki on Sensor Fusion.  It is currently a small, simple article that I hope to expand and encourage anyone interested in Sensor Fusion to help me expand.

Randomness and why the Banks are failing

This last New Years I was talking to a friend in the Banking industry - 2 am and drunk.  He was lamenting the fact that the financial institutes based all of their risk assessments of these fancy derivatives on models.  The models were traditional mathematics as well as neural network based artificial intelligence.  The point he kept coming back to was their reliance on these models and how the models didn’t make good predictions under certain circumstances.

My response was that models are as good as their inputs.  As I’ve stated in a previous post, models are designed, built, and validated to answer very specific questions under very specific circumstances.  When the question is outside the design parameters then the results are likely to be garbage and certainly untrustworthy.

An interesting article on the Edge…

On the Edge, Nassim Nicholas Taleb details the pitfalls of trusting models and an incomplete knowledge of statistics and randomness.  Taleb is the author of Black Swan and Fooled by Randomness.

The parts of the article I found most interesting were the ones about making decisions based on a rudimentary understanding of statistics.  His analysis is dead on.

On the International Space Station one of the Control Moment Gyroscopes (CMGs) failed.  After it failed NASA made the root cause analysis a high priority.  That said, engineers tried before me to explain why the CMG failed.  A bearing expert came in to tell us it could be this or it could be that or it could be something else…  The vendor responded similarly.

While I worked on figuring out the CMGs I was asked to determine what parameters in our telemetry stream could warn us of impending failure of another CMG.  Obviously we would prefer to shut it down before failure and bring it back to Earth for dissection and study.  I used every interpolation trick I knew, including Online Recursive Least Squares and Kalman filtering, to better predict what what was coming next.  No matter the technique or trick we applied to the data it didn’t work out.  “Good” techniques and models worked well on past days data but invariably they eventually fail to predict accurately some future event.

Randomness and Stochastic Control

Many real world systems are analyzed in stochastic manner.  In other words we assume that system noise and disturbances are random processes.  Frequently they are not random.  Frequently the processes are correlated.  Most of these systems can be approximated with these stochastic processes and analyses.

Systems engineering is often based on approximations of stats based models.  Obviously we have to be careful when we build models and especially careful letting Systems engineers (or ourselves) make decisions based on these models.

Source Links

Edge discussion on stats

Follow up discussion

Maybe I’m wrong but here are a few suggestions for making cars better.

Stirling vs. Alternators

In modern automotives the alternator uses energy off the engine to generate electricity to power the radio, A/C, etc.  The alternator sucks horsepower off the engine in order to create this electricity.

What about the Stirling engine?

The Stirling engine is a piston engine driven by an external heat source.  Internal combustion engines have a lot of waste heat that can be captured.  The radiator and exhaust are 2 obvious examples.  Current Stirling designs have reported efficienies of 18%.  Alternators are most likely 90% or more but they require power directly from the engine.

Stirlings have a lower efficiency but convert otherwise wasted energy.  Alternators are robust and realiable.  Stirlings are unproven but can be designed to be small.  Several, possibly more than a dozen, can be placed at convenient places around the engine.  Designed properly they could be swappable such that if one Stirling fails it can be replaced easily and without loss of electricity generation.

Airplane Nozzles vs. Electronic temperature control

The most recent trend in air conditioning and heating in luxury cars is to provide electronic control for each side of the car and in some cases it seems like control is provided for each passenger.  On airplanes each passenger has their own control through a simple nozzle.

So why do we need electronic control that can fail when a simple nozzle would suffice?

Layoffs suck…

I work in the Aerospace industry.  It used to be a 7 year cycle of ups and downs.  Thankfully the 7 year cycle doesn’t appear to be sync’ed in time between the different Aerospace companies.  Unfortunately the 7 year cycle appears to be more like 3 or 4 years now.

During the down side of the Aerospace business cycle layoffs happen (also read s**t happens).  Over this last year my company has been in a constant state of trickling layoffs.  My employer has stated that people with the correct skill set will be kept.  The reality appears to be that dumb luck defines who stays and who goes.

A state of constant layoffs drives morale into the basement.  As a result most people who can find a satisfactory replacement job have done so.  That isn’t to say the rest of us are unemployable by another company but many are looking for a job in just the right place with the right salary and several other factors.  There’s no need to take just any job so long as you still have your current job.

Morale after over a year of constant layoffs

The morale in any company after over a year in a state of constant layoffs is always terrible.  There is no other way to describe it.  After a year, every employee

1. has lost faith in their upper management to fix the situation. 
2. is sick of waiting for things to get better on their own. 
3. is sick of the revolving door of hirings, firings, adn reorganizations at the top. 
4. has had their hopes for a promotion in the near future quashed.
5. has gotten no raise or a crummy raise.
6. has watched a lot of friends leave or get the pink slip.
7. feels unappreciated.
8. is ready to move on. 

So who’s left?

 After a year of layoffs, who’s left?  You can probably answer that for yourself but it boils down to the people with no ambition and the people looking for another job.  The people looking won’t just stop looking when things get better.

Do layoffs payoff?

Layoffs happen.  After a couple of years of good profits most companies have some deadwood.  Short targeted layoffs are probably worth doing every so often.  Long, drawn out layoffs just drive everyone to look for alternative employment.

In engineering, there is a “coming up speed” time that every new employee goes through.  New employees have to learn the ins and outs of the company’s processes.  New employees have to learn the people involved.  New employees have to learn the already developed tools and the tools that still need development.  Replacing an engineer is not cheap but the true cost doesn’t appear to be recognized by the guys at the top.

Control System Modeling: Purpose

I’m going to use model and simulation as synonyms in this post.

The purpose of modeling in any discipline, including control systems, is to answer a question; often a very specific question is answered.  There are several reasons for why any given model only answers a small set of questions.  Budget and Schedule.

Modeling Complexity

Budget and schedule force engineers to model only those aspects deemed necessary to answer the question posed.

Modeling the universe in detail – even the very localized universe around a small object – takes a lot of work and time.  Budget and schedule concerns always force engineers to start with first principles and then model progressive deeper levels of details and fidelity.  The deeper layers are only modeled if the desired level of result accuracy requires this extra fidelity.

There are several reasons for keeping a model as simple as possible:

1. Initial time to development goes up with complexity
2. Time required for maintanence goes up with complexity
3. Odds of a mistake go up with complexity
4. Time between simulation start and delivery of results goes up with complexity

My observation is that items #1 through #3 increase roughly exponentially with complexity.  Turn around time (#4) increases but the amount of increase is highly dependent on the slowest part of the model as it exists prior to the increase in fidelity.

Control System Modeling: Pitfalls

Expanded Purpose

Engineers and other professionals who do not create or run simulations on a regular basis often forget about the narrow focus of a good model.  As a result these people often ask for results the model is not designed to produce.  Obviously the engineer being asked for the results needs to consider the request very carefully.  There may be an assumption built into the model which invalidates its use for this expanded purpose.

Juggling Programs

Each day that I work on a model I go through a process of “loading my RAM” or short term memory.  In order to work on the model and produce meaningful results a certain number of details and parameters must be loaded up into short term memory.  I find this process takes no more than 30 minutes and rarely takes more than 45 minutes.

The pitfall is in assuming you can juggle certain types of work.  Last summer I was asked to juggle modeling work and hardware maintanence work.  The hardware work needed me for 30 minutes at a time about 4 or 5 times a day.  As a result the hardware work repeatedly interrupted my efforts on the modeling work.  The interruptions came about every hour and a half.  So I used half of my time in between “loading my RAM”.

After about 2 or 3 weeks of trying to juggle the hardware and the modeling work I realized I was never gonig to get anything done on the model if I didn’t set some limits.  I asked the two programs how they wanted me to handle the problem.  The basic response was just deal with it.  So I decided to tell the hardware guys that 2 days a week they couldn’t bother me, except for emergencies.  No one was happy but it was the best I could do.

Clean Energy: The Sterling Engine

A while back I came across the Sterling engine.  The Sterling engine takes waste heat and turns it into piston motion.  Obviously if it moves we can generate electricity from the motion.

The main problem with Sterlings is the efficiency of the power extraction.  Using waste heat the available energy for extraction is significantly less than the original fuel.

Wind Turbines and Controls

Wind turbines faced a similar energy extraction problem.  The folks at NREL chose some advanced controls to extract the maximum amount of energy from the available wind.  With wind there isn’t a consistent energy source and maximizing its extraction is key to making the technology economically viable.

Controls…

The available energy for extraction is small.  Like the wind turbines, Sterlings need control to maximize the energy extraction.

Thoughts?