DailyEE

Your daily buzz of Electrical Engineering.

Recent Comments

Search DailyEE

To Program or Not Program, That is The Question

Posted January 9, 2009

This is a guest post by Sheehan Alam of The IT ReportMr. Alam is an accomplished software engineer working at an international, multi-billion dollar firm.


Every engineer should know how to write a program. Every engineer should be able to write an Excel macro at the LEAST. Automation of your tasks is one of the most important things you can do for yourself because it frees up time, and as the age-old saying goes, "time is money". I'm not saying you need to build a full blown web 2.0 application when all you're doing is generating some reports, but if you can save time by eliminating a manual process, you are working smarter and saving dollars.

Delegation is the best way to do something you don't want to. Managers do it all the time, and so can you. Delegate your manual calculations and reports to the computer. Programs are simply instructions meant to be followed. By boasting your productivity, over time, you will get a raise or a promotion and spend it however you please on geeky things.

For those of you who say "programming isn't for me" you are wrong. As an engineer you are both analytical and methodical. Writing a program is a matter of thinking out loud and defining every step. Programming languages themselves are not complicated. You are simply learning the semantics of a language. Not only that, there is something called the application programming interface (API) which is the documentation on exactly how to use every aspect of the language. If English had an API I would be the modern-day Shakespeare. The true challenge lies in the definition of the problem, which should be natural to you after years of physics, math, and core discipline courses.

Lastly, you don't even need to be Linus Torvalds at the workplace. In fact, as an engineer you can even settle to be a really bad programmer. Break all the rules and write sloppy code, as long as it works. Remember, you are basically creating a short term solution to a problem. I do encourage you though, to write clean and efficient code, because it can only help you in the long run.

You don't want to be that guy doing a million things manually. To program or not to program, that is the question.

How to Electrify Your Interview Skills

Posted January 9, 2009

Update:  I've posted some books here which can hopefully be of some use, specifically in this book.

I just landed a new job a few weeks ago, and I'm in the process of moving and getting settled for the new position.  Here are a few important things I brushed up on, which I've learned from my 5 or 6 successful interviews, and some new ones I learned, for this interview.

All the basics.  Dress well.  Always overdress.  Suit, tie, all that jazz.  As an engineer, you probably know jack about fashion so take a girl with you to help you find a good suit that fits.  You've got one shot to look like you're not an anti-social bum, use it wisely.  Be there 15 minutes early.  Firm handshake, make so much eye contact that it's almost (but not quite) uncomfortable.  Speak slightly louder and slower than you think you need to, you're the man (or woman) and you know it, so they'll want to hear what you say.  Make sure they can.

Before the interview, prepare just like you would for a final or a huge presentation at work.  Good preparation in advance (even if none of it ends up being useful) will give you confidence and should help you to relax on D-Day.  An excellent resource, whether you're an alumnus/alumna or a current student, can be your campus career center.  I attended Arizona State, and they offered mock interviews, and an excellent online source of possible questions and even resume tips.  These can all be found here: http://www.asu.edu/studentaffairs/career/ .  When practicing in advance, make sure to include practice of actually verbally answering the questions, with either a friend or tape-recorder listening.  Remember, speak slowly, clearly, and loudly (within reason).  Some excellent questions to be prepared with are the open-ended, essay, behavioral type questions which seem to be gaining in popularity.  Examples include "Tell me about a time you didn't get along with co-worker", "What achievement, within the past year, professional or personal, are you most proud of?" "Why do you want to work for us" etc.

When answering a question, try to answer in the STAR response method.  Using this, you talk about the Situation, Task (that you performed), Action (what you did and why it mattered), Result (did you save the boss money?  catch a thief?  increase production?).  Lastly, never, ever "BS", lie, or inflate.  Maybe you'll get away with it, but while interviewing interns I could smell it a mile away, and there's no way I would ever hire someone who did that.  Never fear the truth, if you lose a potential job because you were honest there's nothing to be ashamed of.

During my interview for an internship with a satellite manfuacturer, when I was very young, I was asked what the difference between TTL and CMOS was.  I answered, "Frankly sir I don't know, I haven't had a course on that yet."  Followed with "What is VHDL?", to which I answered "I've never even heard of VHDL."  I walked away thinking I didn't stand a chance, but ended up getting a job and within a few months knowing what both of those were.  There's little doubt in my mind if I would have tried to lie through those situations I wouldn't have landed the job.

I'll be posting some resources soon, and some good books which should help you prepare even more for your interview.  Best of luck.

Motor Soft Starting

Posted January 7, 2009

Doing some modeling as of late, I thought it might help to share some experience with motor soft starters since I was exposed to little to none of this in school.  In distribution systems, particularly rural distribution systems, reduced voltage starters are a common device used to decrease “flicker” on the electrical system.  There are 5 popular ways for the motors to start like this.

1.  Auto Transformer

This is the most common type of reduced voltage starting.  For light duty motors you’ll typically see taps at 30%, 37.5%, and 45%.  Medium duty units require more power to get them turning, so you’ll typically see taps at 65% and 80% for units up to 50hp (37kW), and 50%, 65%, and 80% for units above 50hp.  For motors in the heavy duty range, typically the taps are specifically designed for the application and the motor, since the motors get so expensive and their demand grows quite large.

Typically the motor is turned on with reduced voltage, under no or very light loading, and allowed to reach full operational speed.  Then the motor is turned off (while being allowed to free-wheel) and then reconnected with full voltage.  With a motor, V does not equal IR obviously, so the line current is reduced by the square of the voltage reduction. 

2.  Reactor Starting

Reactor starting uses a reactor in series with each phase of the motor, again until the motor is at or near full speed.  At this point the reactor is removed from the circuit; typically they are just shorted and not physically moved or transferred.  With reactor starting, line current is reduced proportionally to line voltage reduction.

3.  Resistance Starting

Resistance starting is similar to reactor starting, however unlike reactor starting the losses are real (i.e. massive amounts of heat) and are typically more economical than reactor starting.  Further the power factor of the motor is not adversely effected like it is with reactor starting.

While working at one facility, I came across a very novel concept of reactor starting involving moving blades and a conductive liquid.  At the beginning of motor start, the blades were entirely out of the liquid, and the motor was stopped.  As the blades entered the liquid, a circuit was completed, and as they moved through the liquid the resistance was lessened as there was more surface area for the conduction to occur.  In this way the motor was “ramped up” to operational rotational velocity, and once there the circuit was shorted across and the motor was brought fully online. 

4.  Part-Winding Starters

Part-Winding starters use windings which are typically connected in parallel.  This reduces starting current, and depending upon their design and construction can be either “full acceleration” or “increment” starters.   Full acceleration uses one set of windings to bring the motor to operational speed.  Increment starters use a series of windings to progress the motor up to speed.

5.  Wye-Delta Starting

This method connects the motor in wye during starting, and then switches to a delta connection after acceleration has been completed.  This reduces the voltage to about 57% of nominal.  Torque is however only reduced 33%.  When the motor is reconnected in delta, you can see large transients however.  I’ve never come across this method and seems to be more popular abroad. (non-United States)

Finding a Job (For College Kids)

Posted December 22, 2008

Having just recently landed a new job, I thought I’d share my experiences with some of you here on finding and procuring employment both at companies that will be rewarding, and in places that you enjoy.   This is the first part of that series.

Finding a job, and landing an interview, can be one of the most challenging steps in the entire career journey, however it isn’t the first.  To be a marketable asset, which is all an employee is fundamentally, you have to be capable of doing something that returns value to a company.  As engineers mostly we design and analyze, and possibly supervise and create documentation.  So as a student, anything you can do to help you learn these tasks will help.  At first take a wide swath of courses, to determine what you like and what you’re good at (not always the same things).  Further, work! 

Getting a job not only gives you some pocket/beer money, but you gain valuable experience that you can’t get listening to your professors, most of whom have been out of the industry for years, sometimes decades.  This isn’t to say that your professors are invaluable, but obtaining a job in an industry that you’re interested in working in gives you a chance to see firsthand how the engineers are working, and what they actually use.  More importantly you build a priceless network of coworkers, which you can draw on after graduation for employment, and if you and the company you work for get along you may even have a post-graduation offer.  Finally, demonstrating both school knowledge, through a nice list of “completed courses” on your resume and a decent GPA, and practical knowledge, through a list of previous employment, will gain you far more opportunities than someone with simply a stellar GPA.

During my college career I worked basically four jobs, one at the world’s leading microprocessor manufacturer when I was interested in computer science and computer engineering, one at a satellite manufacturer when I was interested in VHDL and solid state electrical engineering, one at a large IOU (Investor Owned Utility) in Arizona, and finally as a research assistant for one of my favorite professors.  My senior year I took 19 credit hours, and worked two of those jobs concurrently, and graduated with 170+ credit hours, most of which were in courses I just enjoyed.  You might think that “wow this guy is just nuts” or a glutton for punishment, but that’s not the case at all.  I paid for all this with a sub-par GPA, below 3.0, and to tell you the truth I wouldn’t change a thing!  I could have had a much better GPA had I simply studied full time. 

I also know without a shadow of a doubt that my first post-graduate job was gained through my network and work experience, with little thought given to my GPA, and I also know that after your first job GPA matters little.  I’m not saying don’t try in school, far from it work as hard as you can, but speaking from my own personal experience if you can sacrifice some study time for some valuable work time, do it!

Now, landing your first job during college might prove to be difficult, since you’re young, you don’t have any experience, and you don’t even have much schooling.  But landing a first job (and doing very well there) is paramount, and you should work very hard at it. 

There are several ways to find jobs:

·         Your campus career center.

Your best bet in my experience.  Local companies often go here to find young talent, and it’s a good deal for both of you.  $15/hr for an engineering intern is cheap labor to do paperwork and less important task, compared to the $50/hr that they’re paying the senior engineer in the next cube.

·         On-line websites (hotjobs, monster, etc)

Your worst bet in my experience.  Typically not many companies go there looking for interns or engineers with little experience.  Go ahead and place your resume here and peruse the opportunities, but be wary of things that sound too good to be true.

·         Rock myspace, linkedin, and facebook.

At my current job, I hired my last intern by doing a search on facebook for students in my area, majoring in EE, and looking at their list of experience.  Then select candidates I sent a message to inviting them to apply to a position not otherwise posted online.  As time goes on, I predict more and more employers will be realizing what a valuable tool these things are.

·         Cold calling.

I personally have never had this work.  But don’t let me stop you, this may be something that you need to resort to!

Finally, if you’ve exhausted all your options here and not found anything you are interested in, one viable alternative is finding a volunteer job at a non-profit.   Even if it’s not as an engineering position, this can be very valuable to you as it does many things.  First it widens your network, which can be instrumental in the future.  Secondly, you prove to future employers you know how to get along and otherwise work well in a real-world environment.  Third, you get something that you can place on your resume that proves you’re a good citizen interested in the community, an asset to a team, and just generally a super-duper guy/girl.

I’ll be continuing this series soon on landing a good career, stay posted for more tips.

EHV and Cancer

Posted December 19, 2008

Researchers in China may have found a link between EMF produced by EHV lines and transformers and leukemia.  For years it's been thought that the EMF produced by low frequency (50-60Hz) power lines and transformers caused childhood leukemia by breaking down bonds in DNA structures.  Further it was shown that in certain regions, clusters of childhood leukemia cases were found around transmission lines and substations.  However in laboratory tests none of this could ever be replicated and no DNA structure dissolution could be seen.

However Xiaoming Shen and coworkers at the Jiao Tong University School of Medicine in Shanghai have discovered that only children with a genetic variant, known as polymorphism (aka “snp” pronounced snip) are more predisposed to this condition.  The found that children with this condition, who lived within 300 feet of transmission lines or substations had over four times the incidence of leukemia than similar children in similar environments without snp variants in their genes.

Children with these snp variations have difficulty repairing DNA structures after they have been broken.  Shen and Tong theorize that these children near EMF sources aren’t experiencing increased DNA restructuring, they are just predisposed to have problems rebuilding DNA, and then the EMF further hampers their reconstruction and cause increased incidences of leukemia.

An abundance of further information about the details of the research and technical causation of  can be obtained by reading the article in this link:

http://www.informaworld.com/smpp/content~content=a906347140

My question, as an engineer, is how EHV lines contribute to this more than household wiring.  A transmission line in the system I’m working on currently is typically around 138kV and carries around 250A.  While these transmission lines carry much more power than your typical household wiring, the amperage is not a whole lot more (in this case it’s carrying about 10,000 more power, while only carrying 10x more current).  The reasoning behind this is due to voltage transformation, the higher the voltage the lower the current.

So the current is not significantly higher, and we know from fundamental physics that the EMF from this line of charge (I say that the model for a transmission line in this case is best suited to an infinite line) drops as a function of 1/r2 where r is the distance to the source.   Further we know from amperes law that the electromagnetic flux (B-field) is equal to μ0 times the enclosed current.  Therefore, using the numbers already used here a child 150ft (half the 300ft distance stated by Shen et al) away from a EHV line carrying 250A, will be exposed to roughly 4.58*10^-8 Tesla.  However, the same child in a house with a wire carrying enough current to power an average stove, will experience 9.162*10^-6T, which is about 100 times GREATER exposure of EMF than the child sitting 150ft from a transmission line. 

I use these numbers because I had a customer complain once about EMF from a line, and after going out to her house with an EMF meter and proving to her that the field was greater in her house from the EMF than it was outside near the line, she was assuaged of the transmission line’s danger to her.

The 5th Element?

Posted December 19, 2008

Ok well... not really a 5th element, but scientists seem to think that they've found a 4th fundamental electronic device.  The first 3 fundamental devices being resistors, capacitors, and inductors.  This is an interesting concept, and it will be fascinating to see what new sort of technologies and devices we're able to build from them.

http://www.spectrum.ieee.org/dec08/7024

 

Work Plan Creation

Posted December 19, 2008

The past two weeks I've been working on constructing a work plan for a small cooperative.  This cooperative has a model, built in WindMil, however no load has ever been exported from the billing or AMR into the model.  It looks like it just has load allocated using the REA method (a method popularized by the Rural Electrification Authority 50 years ago, which is fairly accurate and distributes kW demand based on kWh usage) with substation loading distributed amongst the line sections.

As with many other tasks in engineering, a work plan requires good information input to get good information output.  Getting these data (specifically system information and billing load) from the customer or your employer is vital, and will require strong personal networking and ties.  I’ve learned that getting them to understand the accuracy and vitality of these numbers is not easy, but will pay off big in the long run.

Currently I’m trying to determine the ordering of “most important” to “least important” system improvements, which is difficult for me to do as the model is telling me that in many locations customers are seeing very low (in the 108V to 109V range) voltage.  It seems that the engineering task at this point shifts away from scientific or objective, to a more artistic and experience based approach which is pretty challenging and pretty fun.