IEEE Benefits Bulletin Links to Questionable Banking Advice

December 18th, 2011

IEEE e-mailed a “benefits bulletin” last week.  I love IEEE and am an officer for the Madison, WI Section, but I have never found anything useful in the financial services IEEE promotes.  Of the non-officer snail mail I receive from them, most of it promotes insurance products.  Based only on their snail mail and discounting all the good stuff I do with them online and in person, IEEE feels like an insurance company promotion organization with a side focus on engineering.

The most recent e-mailing links to a US Bank website extolling the benefits of a good credit score.  Their argument goes something like this:

We’ve developed a scoring system for individuals’ trustworthiness when it comes to paying back money.  The way to score high is to use our product.  The benefit of scoring high is you can use more of our product.  Even if you don’t like using our product, the score is being used by other industries.  We’re hoping this number we created will become widely accepted as a general indication of trustworthiness in all areas of life.  So use our product!

That’s a great deal for them.  It’s no wonder some people feel such anger toward financial institutions.  That anger is unnecessary, IMHO, because banks are not as powerful as they fancy themselves.  Their scoring system is not widely accepted, except by banks and now insurance companies.

I am not categorically opposed to a single number that attempts to distill someone’s trustworthiness in paying bills.  I’m opposed to the banks FICO score, though, because a) the scoring method is not open to the pubic and b) much of it is self-servingly based on using their product.  I find this doubly offensive when it’s presented as tips for responsible financial management.

I am positive overall on the concept of finance and have encouraged engineers to go into the financial industry if they are so inclined.  When we deal with any company promoting a product and they say “How can I help you?” we are immediately cautious they’re trying to sell us something.  Financial institutions should be treated no differently.  IEEE should not present their propaganda as tips for responsible living.

Thermal Relief Error Costs 10dB

July 16th, 2011

A few years back my layout package omitted a thermal relief on the pad of an RF part placed with a ground pour.  The software is supposed to put small traces from the pad to a pour of the same net wherever it can.  For some reason, when I did a minor rev of this board, the software omitted one of the thermal reliefs.  This resulted in a pour ground connection on one of the two ground pads on an RF connector.

At DC this is fine.  At 2.4GHz the effect is negligible.  Between 5GHz to 6GHz, the error results in 8 to 10dB of signal loss.  If you simply scrape away a bit of the soldermask and bridge the pad to the pour, the footprint works perfectly.  It is amazing that that tiny mod results in 10 times the power being delivered to the antenna.

I suspect the reason for it has such an impact is the center conductor goes to a coplanar waveguide, i.e. a trace with grounds below it and on both sides of it.  The ground plane on the left side is connected to the ground on the right side through ground vias, but there is no path not through a via.  If there had been copper on the same layer as the connector connecting the two sides together, the missing thermal relief would not have mattered as much.

The cautionary part of this story is the length of the coplanar waveguide is less than 1 cm.  The fact that the length of the transmission line is shorter than the wavelength being transmitted does not make it immune to transmission line problems.  The entire path through the connector and connected coax is longer than a wavelength, and a discontinuity at one point can drastically effect power transfer.

This all happened years ago, but every once in a while someone new at the client asks me why we rev’ed this board, and I have to go over this infamous mistake.

Engineer Struggles to Understand IDEF0 and Process Modelling

May 19th, 2011

David Marca presented at today’s meeting of my local section of IEEE on the topic of modelling systems.  David’s focus is not on a modelling something like a filter or a channel but rather the context in which an enterprise operates.

If the team doing the work and the customer both understands the problem being addressed on a project, Agile Methods can handle the specification.  Usability Engineering is when someone on the team works closely with the customer to get feedback.

More complicated problems or problems that are new to the team and the customer can be modeled with a Universal Modelling Language (UML).  SADT/IDEF0 is one such method.

My understanding is SADT/IDF0 is a model of an asychronous state machine.  While it’s in a given state, the system acts as a time-invariant continuous-time system.  Every time the “state” changes, the transfer function of the time-invariant system changes.

I would love to hear feedback on whether my DSP-ish understanding is right.

TI Switcher Module Touts “Higher Efficiency than a Linear”

May 18th, 2011

This must be a case of a tech writer generating a product description after talking about the part to an engineer.

The part in question is a TI switching regulator module.  When you go to the PTH08080W product page, the second sentence you find is: The PTH08080W sources output current at a much higher efficiency than a TO-220 linear regulator IC, thereby eliminating the need for a heat sink.

This is like saying the car we’re selling is much faster than a flat-bar hybrid bicycle, thereby eliminating the need to carry a water bottle.

It’s bizarre that they compare their switcher’s efficiency to a linear regulator, which is the least efficient way to step down voltage.  The whole reason to use a step-down switcher is to avoid this inefficiency, so of course their part is more efficient.  It goes on to say this eliminates the need for a heatsink, which is sometimes true, although whether you need a heatsink on a linear regulator depends on how much power your dissipating in it.

My guess is TI figures most people go to the page to scan the specs and get the datasheet, so it doesn’t matter if the description makes sense.

AirFire Mobile: More Olive Tree than Lexus

May 14th, 2011

I love my Wisconsin-based mobile phone provider, AirFire Mobile, formerly Einstein PCS.

  • I can get through quickly to a Wisconsin-based employee when I have a question.
  • $35/mo for unlimited calling plus unlimited low-speed (64kbps) Internet access
  • They help you set up any phone you bought on E-bay just as eagerly as they help you with something they sold.

I have been having a problem using their service, though, when I’m in a certain area of Madison.  The phone signal indicator goes up and down as if it’s trying to link to several towers.  This drains the battery over several hours.  I cannot make a phone call from this area unless I move to higher ground, where apparently I gain line-of-sight to a tower that my phone can use.

The people at AirFire are still trying to work out why this is happening.  Yesterday I was on the phone with them about this issue, and I got disconnected.

Me: I got dropped while someone was helping me with an issue.
AirFire: Did you get his name?
Me: No, I don’t think he said it.
AirFire: Hmmm… Do you want to go over your problem again with me?
Me: He had faint accent.  I think he was from California.
AirFire: That’d be Charlie.  Let me get him.
She found the correct person and put him on the phone.

Borrowing Thomas Friedman’s metaphor, AirFire Mobile is much more Olive Tree than Lexus.

I hope they can solve my problem because I think it’s something idiosyncratic, not necessarily a flaw in AirFire’s system.

“Open for Business” Is a Bad Tagline

March 21st, 2011

I am sorry I’m not going to tomorrow’s Wisconsin Innovation Network meeting, where the talk will discuss the Wis4Biz initiative and how it will “reinforce Wisconsin’s Open for Business message”.  This slogan is not something we should want to reinforce.  I would like to learn more about how anyone thinks this campaign makes sense.

“Open for business” sounds like something you say when you’re just barely functioning, as in “the building was severely damaged in the storm, but we’re still open for business”.

It also sounds backwater.  (Nothing against West Virginia, but they use the same slogan.)  The dollar-store-style sign furthers the backwater image.

The slogan sounds like something that would appeal to someone who a) is not formally educated, b) worked mostly in government not business, and c) knows business is very important but does not understand how it works.

The bizarre thing is that Wisconsin already had a good brand image of good work ethic, good values, and stoicism.  Wisconsin was specifically mentioned as a good place to locate a shop in the book “Execution: The Discipline of Getting Things Done” for these reasons.  We should be building on this already-existing image instead of something taken from an off-the-shelf sign.

Will Information Technology and Automation Result in a Long-Term Shortage of Jobs?

March 16th, 2011

Dave and Chris of the Amp hour asked on their last show whether technology will eventually cause a shortage of jobs.

If computers could replace almost all human jobs, it would be a good thing because that would mean we could have the same goods and services we have today without human work.  This is something people have dreamed about since time immemorial.

When the industrial revolution started, it meant people theoretically could work much less and have the same lifestyle they would have had in agricultural economy.  They could spend the rest of the time relaxing with their families.  Instead, though, society chose to keep working and produce way more goods and services.

Something similar happened when we went from a hunter-gatherer society to an agricultural society.  We may be seeing something similar today going from an industrial society to an information/automation society.  If it plays out the same way, the amount of goods and services we can produce will continue to increase.  We are right to be asking the big questions of how we want that go forward and what we want to do with our new-found production capabilities.

Buy Your Own Tools

March 14th, 2011

My first job was at a small private company.  They were frugal about spending money on tools.  I recall being frustrated at having to spend time wandering around looking for tools.  I had a notion to bring my own tools, but my colleagues told me it wasn’t a good idea because with tools in short supply they would be easily lost to well-meaning people who would borrow them and forget to return them.  The company eventually bought me new tools, provoking some resentment from other employees who didn’t have new tools.  I had no idea at the time that the on-the-job training I was getting in circuit design and debug was worth way more than the cost of some tools and books.

In retrospect, I can’t imagine why I didn’t use my own tools.  Even at a company that is short on tools, if you keep them in a locked toolbox, they’re not likely to be taken.  A very nice set of electronics tools can be had for under $1,000.  That is a modest amount for an engineer.  Even for people working at full-time jobs that claim to provide everything an engineer needs, engineers are still paid for getting things done.  The employer may not notice time wasted searching for tools, but that time is money/experience lost to the engineer who could be using that time to design new technologies.

All engineers make their money by causing things to be worth more than the sum of their parts.  Tools and books are a cost of being an engineer.

Engineer Barbie

February 8th, 2011

From an IEEE article: Preparing Barbie for a High-Tech Career
[Mattel] let the public vote online in January 2009 on Barbie’s next career in its “I Can Be…” series, which aims to acquaint children with a variety of professions. As more than 600 000 votes poured in, Computer Engineer Barbie took a commanding lead over the other choices, which included environmentalist, architect, and surgeon.

Any publicity is good publicity, so I appreciate the efforts of the IEEE member who promoted this.  If Barbie is getting new professions, engineering should be one of them.

If they were going for realism, they could have given her lighted magnifying goggles and a nice button-down short-sleeve shirt with a caliper and tweaker stuffed in the pocket.  The promotional pictures should have her looking down, not right at the viewers eyes.  I do like the narrow glasses, though, which allow her to look over them for reading markings on parts without removing them.  They could also do a product placement partnership with Pepsi by giving her a Mountain Dew.

My vote for Barbie’s next job would be for her to retire and become an historical relic.  Barbie is a goofy toy to begin with, but juxtaposing her with real professions, esp a male-dominated one, make the dolls seem even goofier.  It seems like a toy that was offensive a generation ago and is now just a joke.

Invalid Altera FPGA MSEL Configuration Causes Big Trouble

February 6th, 2011

A recent board I designed used an Altera Cyclone VI FPGA.  This FPGA’s configuration scheme is set by the value presented to the part’s MSEL lines.

There is a table showing how to strap these lines for each configuration scheme.
Configuration Schemes for Cyclone IV E Devices

When I first tried to bring up my board, I had these lines tied in a way that would have been defined for the GX version of this part, but not the E version that I was using.  When I tried to connect by JTAG, I could observe lots of traffic on the JTAG lines, but the utility gave me the conf_done failed to go high message, which from what I can tell means almost anything that prevents communication with the FPGA.  The FPGA held various I/O lines high or low without apparent order. Within a single bank, some lines would be weakly pulled up, some lines would be high or low, and some would be at intermediate values.  I put resistors on the lines that were high or low and found a 100 ohm resistor would usually pull them most of the way up or down but not all the way.  The way so many lines were pulled up or down made me think that I had a footprint error, major soldering defects, or some kind of fabrication error allowing planes to touch some of the signal vias.

One of the lines being pulled low was my processor’s reset line.  (Maybe it was a bad idea to connect it to the FPGA, but I had spare pins so I went to town connecting nets that I didn’t plan to use on the FPGA.)  I couldn’t cut the line because it ran on an internal layer the whole way.  I ended up removing the processor, thinking some kind of error with the processor was causing its reset line to be held low.  I was disappointed to discover that the line was still being pulled low just as strong.  I was considering removing the FPGA too when I discovered the MSEL error.

After fixing MSEL everything amazingly worked.  The MSEL line is the only jumper wire on the whole board now.

The FPGA datasheet says the MSEL lines must be tied directly (not through resistors or driven by another device) to V[CCA] (not to the I/O voltage).  After this, I will always follow any rules related to MSEL lines to the letter.