Thanks everyone for joining in for this delightful chat. You've given me loads of ideas for future articles...which I will assign to Max. :-) Don't miss Max's radio show coming up in three weeks, where he will ponder the intracacies of embedded.
Have a good weekend all and happy analog-ing, or plumbing if you prefer!
@zeeglen "I used to work at a place like that with summer college interns. A couple of them specifically asked me for analog training. Wrote a blog on that once, might have been Scope Junction or the Connecting Edge, can't remember for sure."
I'm always seeking to hire Analog/Digital engineers and I consistently find that newer engineers are short on the analog side. I'll still be hiring 3 to 5 years from now so it'd be great if I could point these [strong] engineers to a house of training.
Analog design has helped me to be able to wade throught the cloud of noise in the middle of the two extremes and I'm glad I've done it. I also like coding MCUs and doing power. A little pitch for all the recruiters on line.
Analog will survive, but not as we now know it. It is morphing into a more well-rounded engineer/designer who specializes in analog or power or RF as well as knowing how to work with the other disciplines of digital and software design, etc.
Mentors will be found on line in tech publications, the local IEEE community and Webinars and even YouTube as well as live events
Historically apprenticeships had issues. Most were used as drudge servants (basically worse than slaves. emptying chamber pots, cleaning) Masters were lothe to share secrets. At best one could marry the master's daughter and hope to join family. Usually what happened was that the apprentice, physically assaulted the master when the master got too old to fight back.
@ Wintz I was thinking of a more streamlined location where aspiring analog engineers could gain some hands-on industry experience, tips, examples.
I used to work at a place like that with summer college interns. A couple of them specifically asked me for analog training. Wrote a blog on that once, might have been Scope Junction or the Connecting Edge, can't remember for sure.
@steve.taranovich "IoT has analog signal conditioning and RF and power management---there is a great deal of in-depth analog needed unless you use modules. Modules are fast and easy, but costly---your competitor will kill you with their discrete analof, RF and power designs"
So all these guys on Kickstarter who want to be the next Apple, are you saying that the winners will be the ones who have the best analog engineering design on their robotic bartending machine or Internet-enabled medical device?
@Junko: Speaking of IoT trend, how much in-depth analog expertise is needed for that?
The IoT is just a mirror-reflection on the rest of the world. Parts of it can be designed with off-the-shelf modules, in which case you don't really care if they are analog or digital so long as you know how to plug them together -- the trick comes when you are designing something from the bottom up -- and that something needs analog expertise -- at which point everyone starts running around in ever-decreasing circles shouting "Don't Panic!"
@Davd Ashton - what happens when the thermocouple is non linear and the wong lookup table is used? Or the slapmonkey does not bother with calibration becouse they do not understand the real world and xeno's paradox.
@junko--Speaking of IoT trend, how much in-depth analog expertise is needed for that?
IoT has analog signal conditioning and RF and power management---there is a great deal of in-depth analog needed unless you use modules. Modules are fast and easy, but costly---your competitor will kill you with their discrete analof, RF and power designs
@davidashton "Or companies need to offer apprenticeships" "I think you've hit something there - companies these days are too lean and mean"
I think the challenge is the dynamic between employees and employers today: Until we run out of engineering talent, there isn't a lot of incentive to train people that ultimately leave. Although on the other hand, it's not like this is a new phenomenon. When I worked at TI in the mid-1980s, they had lots of blue badges (under 5 years) and lots of gold badges (over 15) and a big yawning gap in the middle.
@Sheepdoll "Are we too dependent on such tools such as Matlab, and pSpice"
Try this one: I asked two engineering managers - one HW and one SW, what they thought of the concept of going on line to find canned solutions. I was astonished that both of them answered very positively about that prospect. My response: then why bother hiring/paying engineers? Just find the most resourceful and computer savy seekers of information you can find for min wage. They're reply: And then what? Hmmmm.
One big trend is the Internet of Things (IoT) -- I think we are poised on the bring of an exponential explosion in the use of sensors -- and many of these sensors are analog in nature and are measuting analog quantities -- someone has to knwo how to design these sensors, model them, and process the information coming out of them.
There's also a trend to pussing intelligence as far out to the edge of the net as possible -- don;t send huge amounts of raw data back to a central location -- process it at the sensor and then send information back -- this all goes to Steve's point about analog engineers also having to knwo microcontrollers and suchlike...
@kfield---the problem young engineers, who want to be analog, have is they want to specialize---that's OK, but also know and learn more about the other parts of engineering design like digital and power
@prateekvr "For students, its extremely hard to get into the Analog industry, most positions requires experience! 2+ years. With debt, many are spooked and end up focusing on Programming to fill their stomachs and clear the debt. More opportunities for freshers needed. Will there be?"
So we have young engineers who can't get into the field and presumably learn from the generation retiring out, isn't this going to leave a big gaping hole?
@kfield: ...isn't the problem with some of these tools is that they can encourage plugging-and-chugging without any real understanding of the basic principles?
That's sort of true of any tool -- a lot of today's "digital designers" wouldn't know the difference between a ripple adder and a fast-carrty adder if one were to crawl up their leg and bit them on a very unfortunate part of their anatomy
@sheepdoll--To your point"Regarding the demise of analog engineering. It seems that analog circuits like the example posted are easy to solve with field equations in a simulator (Ironically digitally implemented.) Are we too dependent on such tools such as Matlab, and pSpice which can add to the cost of the project?"
Expertise in analog layout and design is still the key to a good analog designer. My analogy of a doctor: you go to medical school and learn it all and then you specialize----engineers need to do that better than we do now and continue to learn always
Regarding the demise of analog engineering. It seems that analog circuits like the example posted are easy to solve with field equations in a simulator (Ironically digitally implemented.) Are we too dependent on such tools such as Matlab, and pSpice which can add to the cost of the project?
It's evolution leading to demise. Remember the "logic designers"? They were replaced by Verilog and fpga's. That job is gone, but engineers moved forward to higher level design. Analog design has suffered from fads, mostly the fads of college instruction. Today's fads are RF, ADC, and power-supply design. Meanwhile the rest of the technology is allowed to fade,,,
For students, its extremely hard to get into the Analog industry, most positions requires experience! 2+ years. With debt, many are spooked and end up focusing on Programming to fill their stomachs and clear the debt. More opportunities for freshers needed. Will there be?
If by digital engineer, you mean logic gates and nothing else, then, yes. I'd agree. Purity in electrical engineering is probably a lost concept. You need to understand digital, analog and software to reall get by.
@kfield---I was solely an analog designer for 28 years until I joined TI in 2000 and was forced to look at the bigger picture----How microcontrollers, DSP, software and power management fit into the design. There are very few experts just analog today
I think the evolution of engineering is a great topic, especially given all of the advancements in technology and tools with the generation of engineers that started their careers during the digital revolution
@Harvey---the proctice comes with using evaluation boards, reference design boards, etc. And call the tech line at the companies that supply these----ultimately, if you can---go to seminars and electronics shows with good tech sessions----the analog mentors are few and far between nowadays
At this point it makes more sense to learn analog and then double back for digital - get the strong analog training and knowledge transfer while you can. I learned a lot in college but it's the hands on training and "rubbing elbows with other engineers" that will allow you to gain the engineering "wisdom."
@prateekvar: So for someone who is planning to specialize in Analog Design, what skills are a must then?
You have to really understand the characteristics of the various circuit elements (resistors, capacitors, inductors -- and diodes, transistors, etc.). You also have to understand that in the analog world you also have to deal with these things as parasitic elements associated with interconnect and package pins and voas on circuit boards and stuff. You also have to be able to use tools like analog simulators (an art in itself)
That's exactly how companies are shooting themselves in the feet. There is no serious focus on knowledge transfer or cross-training. One day, a function/task will make more sense via analog and there'll be no one who can make that happen.
@maxthemagnificent "So when we talk about "analog designers" here -- are we talking about people creating analog circuits like audio amplifiers (hence working at the board level with discrete parts) -- or people creating the PHY interfaces on ASICs/ASSPs/SoCs ... or the people creating MEMS components ... or."
I think you hit the nail on the head here, all analog engineers are not the same. But as @prateekvr asks below, what does an engineer need to know today to stay relevant?
@kfield "No one wants to take on an employee they have to train"
When I started my first job as a member of a team designing CPUS for mainframes at International Computers Limited (ICL) in the UK, each new engineer was assigned to an older engineer who was to be his/her mentor -- I have to say that I learned a HUGE amount from my mrentor -- Dave Potts was his name
@kfield--The foundations for analog nowadays are being made from the ground up, not in engineering schools, but online with webcasts, white papers, live seminars, etc. The audience is usually not analog engineers anymore but designers who need to know analog
@Wintz "The problem is that companies are requiring the "Classic" Analog engineers to design Digital circuits (and oftentimes write code) as well."
What comes first, the chicken or the egg, meaning does one need to learn analog before digital or the other way around? It reminds me of Paul Rako saying that's too much information for any one person to hold in their head
So when we talk about "analog designers" here -- are we talking about people creating analog circuits like audio amplifiers (hence working at the board level with discrete parts) -- or people creating the PHY interfaces on ASICs/ASSPs/SoCs ... or the people creating MEMS components ... or...
Analog suppliers have teamed up with software and microcontroller companies or have acquired or become acquired to have the complete solution---these companies ar creating the great tools needed for analog design
@kfield "No one wants to take on an employee they have to train"
That's exactly how I feel. I am just starting working in a semiconductor company, hoping to learn, but I feel the company is unable to invest training in me. I am in applications and yes I am learning, but I'm interested in analog design, and the designers tell me "Oh well, it takes years of experience to get a position." I don't even know how to get in.
Answer is that the engineers doing analog nowadays are microcontroller designers and even software engineers who are not analog engineers, but need to design analog circuits
Tell me about. Today's project is hacking in a 5V/3.3V translator into my new test fixture because the digital guy who designed the interface was not aware that his PIC microcontrollerf analod/digital inputs are not 5V tolerant.
There are still systems that are almost completely analog -- like audio amplifiers -- also I guess there are still analog filters and signal conditioning circuits -- but the vast amount of signal processing is now performed in the digital domain (DSP) and there is relatively little analog signal processing (ASP) these days
@Harvey "The shortage is in ready-to-design engineers who are already trained" I agree, this is a problem in many fields today. No one wants to take on an employee they have to train and if they do they're scared stiff that person will leave and someone else will benefit from the investment in training.
I'd like to pose that there is simply a gap in mindset between A folks and D folks. We think continuously and we're happy in the shades of gray. Many of the people who work in our field, and hire us, are not. They're more comfortable with discrete thinking where only logical conclusions are welcome: one-zero, yes-no, up-down, here-there, now-then.
@Maxthemagnificent To your point, engineers not only didn't learn analog, they didn't learn electrical engineering either. BSEEs awarded in the US have dropped by 50% in this last generation, while computer science grads have soared through the roof
When business is easy, companies treat people well, generally. Now, competition is intense and there is no time for innovation- just product design. Companies now want analog design to be turnkey- like digital design has largely become. The shortage is in ready-to-design engineers who are already trained.
I've had the pleasure of spending some time with our analog engineers and I appreciatae what they bring to the table. Unfortunately, I'm not certain that industry will be able to replace many of the "Greats" when they retire.
The existence of the A to D and the D to A converter tells me that Analog is not dead or dying---it is evolving. We will always need to signal-condition the world around us so that the digital realm can further act upon it (I'm the Shakespeare of Analog electronics)
I'll kick things off here by giving you some context to this chat: My original idea for an article was to write about the toughest engineering jobs to fill, inspired by a trip to Bangkok where I met an engineer who said they could not find any C programmers to hire. But in researching more jobs, what kept coming up was analog, analog, analog. So I published the article recently on EETimes, Analog Engineers: Too Many or Too Few? and stirred up a bit of a hornet's nest with the assertion by some that the glory days of analog were over.
Our next live online chat will commence on Friday June 20, 2014 at 10:00 a.m. Pacific Time (1:00 p.m. Eastern Time). You'll have to work out your local time from these clues (you can always use this handy-dandy Time Zone Converter).
Your hosts will be Karen Field (EETimes) and Steve Taranovich (Planet Analog), along with Max Maxfield (EETimes) making a guest appearance. The topic of conversation will be anything and everything to do with analog engineering and the simultaneous dwindling in and rising demand for analog skills.
As always, we will be following our usual practice of leaping from topic to topic with the agility of young, fearless mountain goats, so make sure you're wearing appropriate clothing!
Replay available now: A handful of emerging network technologies are competing to be the preferred wide-area connection for the Internet of Things. All claim lower costs and power use than cellular but none have wide deployment yet. Listen in as proponents of leading contenders make their case to be the metro or national IoT network of the future. Rick Merritt, EE Times Silicon Valley Bureau Chief, moderators this discussion. Join in and ask his guests questions.