How do you learn to design like a professional?

[ElbowPics]

I'm starting my final year of my electrical engineering degree. My grades are good, I feel like I understand the material, but I've had very little exposure to actually using the information to make anything.

How do I learn to tackle an electrical design problem like a real engineer does? Specifically, what's the process to go from idea to a real circuit on a board?

I've been spending more time making my own circuits, and that's helped get a more practical feel for components, but I never know whether my design would cut if professionally.

P.S. I've thought about looking at schematics for real products online, would that be a good way to learn how the pros make things?

[Cerebus]

A friend of mine who writes novels for a living says the only way to learn to write is to write, every day, and to read a lot. I think it's the same with any design/creative led process. You have to practice, practice, practice and you have to look at other people's work.

I'm a real fan of finding older service manuals for electronics equipment (newer service manuals are conspicuously schematic free) and studying the other guy's designs. Ask yourself what the choices were for each subsystem, why did they chose the alternative that they did, question all the choices and see if you can second guess the designer. Old HP test gear manuals are great for this, also Keithley, Tektronix, and I'm sure you can add your favourite vendor to the list. Also hunt down articles that describe the design of equipment - the, now defunct, HP Journal is great for this, there's an archive online.

[Electro Fan]

I'm starting my final year of my electrical engineering degree. My grades are good, I feel like I understand the material, but I've had very little exposure to actually using the information to make anything.

How do I learn to tackle an electrical design problem like a real engineer does? Specifically, what's the process to go from idea to a real circuit on a board?

I've been spending more time making my own circuits, and that's helped get a more practical feel for components, but I never know whether my design would cut if professionally.

P.S. I've thought about looking at schematics for real products online, would that be a good way to learn how the pros make things?

Welcome to the EEVblog forum(s).

That’s a good question.   

I think Cerebus got the answers rolling with the writer analogy.  This is going to be a good thread.

[penfold]

Since this is such a big question, I'll try to stick to one point:

With regards to the "mechanics" of what happens in an organisation to go from an idea to a product, I'd recommend doing some research into "quality assurance", standards such as ISO9001 (bare in mind there there's a lot in those standards which isn't engineering related), they essentially dictate a framework of processes that (when actually followed correctly) should guarantee that the product is representative of the idea and that the product should be consistently manufacturable. The execution of such a standard will take many different forms and may be unrecognisable from company to company... but elements of them will be there.

[Siwastaja]

Tear down as much as you can. I did that a lot as a kid and luckily my father understood this is actually a constructive hobby so hoarded me broken, sometimes even functional electronics just so that I can tear it all apart.

You don't even need to understand everything first, that can come later. You get an intuitive idea of what kind of solutions have been proven to work in real world.

Then just design, design and design. Best case, you find clients you work for who are understanding if you sometimes promise too much in too short of time. This way you can pick up challenging design tasks and learn.

Also repair. Repairing not only gives you insight how real products are designed, but also shows the common pitfalls where the designers failed to make a reliable product, so you can do better.

[rstofer]

In the professions like civil, structural, mechanical and electrical engineering, the new grad, having passed the Engineer In Training (EIT) exam, will spend 4 years working under a registered professional engineer before even being eligible to take the Professional Engineer license exam.  Nobody is expecting a new grad in any of these fields to be able to design anything!

I would expect something similar for electronic engineers even though it may not be as rigorous.  No company is going to hire a new grad and expect them to design the newest widget with cutting edge technology.  Instead, they will team the new grad with one or more experienced engineers for some fairly long period of time.  There will be a huge difference in salary which may close up over a long period of time - several years, not months.

Summer internships are another way to gain experience.  Most major companies have a program for taking on interns over the summer break.

If, upon graduation, you can handle the math, the rest will come with time.  Companies will expect the new hire to be competent with math and probably programming.  MATLAB (or Octave) experience may be expected.  Python will be on the list as will some C++, I suppose.  You simply can't slide on programming.

[tggzzz]

Learn the basic theory of how ideal circuits work. This involves a lot of maths and abstractions, which may not seem realistic at first. University is the best place to get a solid theoretical grounding, and that will last you a lifetime.

Learn how components and construction techniques deviate from the ideal. Read many data sheets, application notes, and component catalogues.

Pick something to build that will stretch your understanding and capabilities. Think how you can design, implement and test small parts of it in isolation, and then assemble the parts into the complete system. Do that. Make mistakes. Learn from your mistakes. At an interview that will demonstrate that:

• you do more than the minimum, because you enjoy it (and hence will enjoy working for the employer)
• you can set goals that you believe are difficult (for you) but possible
• you complete your chosen tasks
• you can state how you would do it better next time

Try to avoid spending too much time learning "which button to press to cause the widget to rotate and turn white". Such knowledge almost certainly won't be relevant to your employer, and will be obsolete in 5 years Basic theory lasts a lifetime and is useful to "all" employers.

Speedread Asimov's "Profession" http://www.abelard.org/asimov.php  It significantly influenced my career, and is still relevant 60 years later (if not even more relevant!).

[Cerebus]

Learn the basic theory of how ideal circuits work. This involves a lot of maths and abstractions, which may not seem realistic at first. University is the best place to get a solid theoretical grounding, and that will last you a lifetime.

Learn how components and construction techniques deviate from the ideal. Read many data sheets, application notes, and component catalogues.

Pick something to build that will stretch your understanding and capabilities. Think how you can design, implement and test small parts of it in isolation, and then assemble the parts into the complete system. Do that. Make mistakes. Learn from your mistakes. At an interview that will demonstrate that:

• you do more than the minimum, because you enjoy it (and hence will enjoy working for the employer)
• you can set goals that you believe are difficult (for you) but possible
• you complete your chosen tasks
• you can state how you would do it better next time

Try to avoid spending too much time learning "which button to press to cause the widget to rotate and turn white". Such knowledge almost certainly won't be relevant to your employer, and will be obsolete in 5 years Basic theory lasts a lifetime and is useful to "all" employers.

Speedread Asimov's "Profession" http://www.abelard.org/asimov.php  It significantly influenced my career, and is still relevant 60 years later (if not even more relevant!).

I think the point is being missed in two ways here:

I'm starting my final year of my electrical engineering degree. My grades are good, I feel like I understand the material, but I've had very little exposure to actually using the information to make anything.

i.e. He's learned the theory, so advising him to learn it is a bit odd.

How do I learn to tackle an electrical design problem like a real engineer does? Specifically, what's the process to go from idea to a real circuit on a board?

He's asking for advice on how to do it, not how to get employed to do it.

[TimNJ]

Agreed with above about looking at other people's work. (FYI, I'm only 4 years into my career in electronics design, post graduating college. Hobbyist since I was about 14 y/o.)

For a particular set of requirements/constraints, there are lots of ways to go about it. As an inexperienced designer, your brain can only think of a limited set of ways to approach and solve the problem. When you start to look at other people's work, you start to see ways of design that seem "obvious" in hindsight, but you probably wouldn't have thought of given your level of experience and exposure.

That's not to say that all designs are good designs. (Obviously not. Use some common sense when evaluating whether someone else's design is actually good.)

Teardown videos and photos were a real eye-opener for me. I was lucky enough to get started when Dave, mikeselectricstuff, etc. were also getting started. Now there's many more. These videos often don't go into low level analysis, but they can still help you build an understanding of the general architecture of various  products, construction techniques, and may introduce you to components that you didn't know about.

[ebastler]

With regards to the "mechanics" of what happens in an organisation to go from an idea to a product, I'd recommend doing some research into "quality assurance", standards such as ISO9001 (bare in mind there there's a lot in those standards which isn't engineering related), they essentially dictate a framework of processes that (when actually followed correctly) should guarantee that the product is representative of the idea and that the product should be consistently manufacturable.

I beg to differ. Do yourself a favor and do not look into ISO 9001 (at least not now). That standard is not technical at all, but entirely process-oriented. And I'd say it states the obvious in so many different ways. "Do risk management! Ahead of doing whatever you intend to do, think about what could go wrong", and things along those lines.

If you don't want your motivation thwarted before you even begin your professional career, stay away from ISO 9001 for now. 

[penfold]

With regards to the "mechanics" of what happens in an organisation to go from an idea to a product, I'd recommend doing some research into "quality assurance", standards such as ISO9001 (bare in mind there there's a lot in those standards which isn't engineering related), they essentially dictate a framework of processes that (when actually followed correctly) should guarantee that the product is representative of the idea and that the product should be consistently manufacturable.

I beg to differ. Do yourself a favor and do not look into ISO 9001 (at least not now). That standard is not technical at all, but entirely process-oriented. And I'd say it states the obvious in so many different ways. "Do risk management! Ahead of doing whatever you intend to do, think about what could go wrong", and things along those lines.

If you don't want your motivation thwarted before you even begin your professional career, stay away from ISO 9001 for now. 

I do agree with your dis-agreement, nobody should read the standard, my reference is to quality assurance/management standards in general, ISO9001 is just the first example which came to mind. Some awareness of how it impacts your life as an engineer is almost the fundamental difference between a professional and a non-professional EE. Awareness of quality management is a requirement of IET accredited degrees... I found it useful to know a little about before beginning work.

My suggestion is that the "high quality", well presented, strongly performing designs one would see in (for instance) HP user manuals are not the result of a single design engineer with individually good skills, but the work of many people within a organisation and one of the means by which such teams can possibly cooperate is by a collection of standard processes... many features of such processes are described by various QMS standards...

[Electro Fan]

Since I liked the OP's question so much I'm going to try to help with the answer or provide some thoughts on how to get to the answer(s).   

I think it's somewhat safe to say that in most businesses technology exists to support business objectives.  So while sometimes pure R&D drives new products and services, generally someone starts with a business objective (such as increasing profit by decreasing costs and/or increasing revenues). 

Once someone has done some high level analysis that indicates if some such product could be brought to market for X investment and produce enough Y profit over the life of the product to drive enough Z ROI to make the project worthwhile vs the other potential investments available to the company, then someone says, "well that sounds good financially but what are the parameters of this opportunity, ie:  what are the 'market requirements?'"

At this point someone should have already gone to the marketing and sales teams to say "if we bring you this new product with these parameters, are you sure you can sell enough volume at the projected price in the forecasted timeframe?  Because unless you can, there is no sense in the company spending NRE and other resources to even go down this path." 

Once the sales, marketing, and other people are pretty sure that the new product with the key parameters will be successful, then the challenge is given to engineering (and sometimes other departments) to design something that can make it real with a high degree of specificity.  At this point, somewhat early in the overall process, there is only a very high level specification, plus a preliminary budget and schedule.  These documents are targets that are close enough to potential reality that it intrigues and motivates the business people but it's still somewhat vague for the engineering team.

After the market requirements are outlined by the business people ("it has to be this big or small, cost this much to manufacture so we can mark it up by this margin and still stay under this selling price, it has to provide these functions, consume this much power" or whatever), the market requirements then land on the desk of the engineering community who is tasked with turning the market requirements into the more detailed product requirements.

The workflow can vary from company to company but at a high level it generally (but not always) goes from the business requirements to technical requirements.  In big companies the process might be nearly overwhelmed with rigorous bureaucracy, in a smaller company the process might be much more informally driven by the instincts and power of personalities, but generally it starts with a set of business requirements that evolve into an engineering design.

I think for the purposes of helping the OP see the basic design process (and there are truly are many variations, so we're just talking "typical" if there is such a thing) it would be good to start at the point where the market requirements (including ISO or any other QA/QC requirements and any industry standard practices) have been largely established by management, the "product manager", etc. and the market requirements are now in the hands of engineering management.  So in addition to the key parameters of the new product the market requirements document probably also highlights the key competitive products - so the engineering community probably has a rough idea of what the alternative designs look like, cost to manufacture, etc.

So, starting with "it's bigger than a breadbasket but smaller than an automobile, these are the "market requirements", so how do we design something that will give us the details of a good "product requirement" document that would be complete and clear enough that it would be sufficient for manufacturing to manufacture – and that will eventually meet the high level "market requirements" which will enable the sales department to be successful and the customers to be happy?"

For example, if the OP is working at a radio manufacturer and the market requirement says the objective is to design a radio that can be manufactured for $x of parts and labor and that supports the following the high level parameters - what does engineering do from that point forward to design the product?

No doubt there will be some ongoing iterative discussions between the engineering team and the business team, but once the business team says here are the parameters of what we are trying to accomplish, and if the OP is assigned to design something that meets the parameters, what's the process from there?  Or what are the top 10 things engineering is going to ask or try to solve as part of the process?  These might be more process questions/suggestions, or they might be more technical questions/suggestions - but let's assume we know roughly what someone wants built and now we need to make the electricity work in circuits.

Net, net:  I think it's good for new engineers to recognize that a lot of stuff happens upstream but in this case I think the OP is also trying to understand what happens after the business objectives have been largely determined.  How do you design power supplies, analog and digital circuits, etc.?  Do you start with inputs and work forward, or outputs and work backward, or some of both, or something else?


PS,  ElbowPics, this is just my guess at interpreting your question(s).  You should feel free to jump in here and tell us more about what would be most valuable for you to learn/better understand.  My guess is that in the post college world you will come up with plenty of ideas on your own but that as an engineer you will also often be taking inputs from lots of other people and trying to turn those inputs into detailed designs.

[Buriedcode]

I'll add my two cents.

Make a list of projects, or problems that are amenable to being solved with technology. 

Large, complex, ambitious projects are good in that it will force you to research each part, and highlight areas you're not experienced/confident in, but with the big negative that it'll just be too much so you might not get anywhere.  Projects will often stall because one doesn't know where to start, how to break things down into smaller problems, and where to source parts etc.. etc..

On the other hand, smaller simple projects won't require you to research areas (sub circuits, sub fields, like filters, analogue, uC programming etc..) won't push you in anyway.  But.. will provide you with experience in construction, prototyping, soldering, housing, PCB design (eventually).  Not to mention that you can actually finish a project, tick it off the checklist and have something you've made that is actually useful.
A part of "making stuff" I don't enjoy much is finding and machining enclosures, and making things sturdy/practical, but it is handy to have some "favourite" enclosures to use so you get used to housing stuff, so you don't have a very delicate breadboard design that can be flakey in operation (which makes debugging difficult, but... another thing to learn!).

So, you could try and find a project that is a happy medium or just a long list of ones that vary in compliexity, from something as simple as say, an LED flasher/fader, or a headphone switch box (which may have no active electronics, just a switch and sockets) all the way to a large project that involves something else you're interested in.  You can then order these in terms of complexity, and work on them one by one.    Each time you come across a module or subcircuit you don't know anything about - google it and maybe knock up prototypes, as-and-when you need to, rather than picking a subject and trying to learn everything about it.

If you're a guitarist, guitar pedals are a great introduction as they vary in complexity from simple single transistor boosters, all the way to hugh arrays of filters.  Making things that you need, and will use is much more rewarding than just making stuff for the sake of learning.  It also provides motivation - having a purpose. 

Lastly, don't be a perfectionist.  It's great to have things that are very precise, well made, or over spec'd, but ultimately, first get something working, then you can make it lower power, small, cheaper, or "better" then make it pretty.  If it doesn't function well, or is unreliable, then that will make you investigate why, and that will only make you a "better" engineer.  I think it's all about experience, you can't get that just by reading (but reading is still required when you need to!).
 

[gooligumelec]

I've been spending more time making my own circuits, and that's helped get a more practical feel for components, but I never know whether my design would cut if professionally.

I think a key here is that word "professional" - what does it mean?  Obviously different things to different people.

For me, it means being aware that I'm doing this design for someone else.  Not just my customer or employer, but their customer(s).  And not just now, but in the future.  In other words, if I'm designing something that will be made in the thousands, I want every one of them to work.  And not just now, but still be working years from now, even in the cold or heat.  Or even if the user plugs it in backwards.
Sure, those reliability goals have to be delivered within constraints - cost, size, battery life.  And real engineering is about juggling constraints and making seemingly endless trade-offs.

So I guess what I'm trying to say is that, when a professional has a basic design concept, a preliminary circuit, they will stop and think "what could go wrong?"  How likely are those things, and can we address them within our cost/size/etc. budget?  Does the design rely on exact component values?  What if they drift over time/temperature?  What happens if the user does something stupid?  Or the input isn't within the bounds we were told (happens a fair bit in industrial design, e.g. don't assume your car battery is 12V...).

Back in university, I was happy to have a one-off design that worked well enough, if you looked at it right.  And it didn't matter how much the parts cost.  Or whether it could be manufactured.  Or tested.

To sum that all up: "professional engineers" stop and think.  Then think again.  About lots of stuff.  Hard to define until you've experienced it, but thinking in terms of customers is a start.

[twospoons]

I've been spending more time making my own circuits, and that's helped get a more practical feel for components, but I never know whether my design would cut if professionally.

Do lots of this. And if you can, find a good mentor who can critique your work.

When you get into a professional environment attend every design review you can, and listen carefully, and question anything you don't understand the reason for.  Ignorance is easy to fix - and I have much more respect for someone who will speak up when they don't understand than someone who pretends they know what they're doing and then makes preventable , and possibly costly, mistakes.

There are plenty of experienced engineers happy to provide pointers to grads, because we were grads once too and we got tips from the old guys we worked with. I feel its my duty to pass on what I've learned over the last 30 years to help the next generation of engineers.

And us oldies love to tell our stories, which usually have a lesson in them somewhere.

p.s. The simple fact that you're here asking this question says to me you will do well professionally.

[Cerebus]

Or even if the user plugs it in backwards.

If your ultimate customers are the police, the military, actors, or musicians: "when the user plugs it in backwards". 

[wizard69]

You have gotten some really good answers in this thread and one of the best was from Cerebus in the first response.   I'm more of an automation technician so can't reply base don being a hard core electronics engineer, however I can offer up some points after working in industry for years.   Some of this is repeating what has already been said.

You don't be afraid to learn outside your specialty.   If you are designing digital circuits gain an understanding of power supplies.   I say this because poor power supply design or selection seems to be a weak link for many products.   Those products that don't have problems have obviously well designed power supplies.  In a more generally sense get an understanding of how your products will be working and what they will be working with or on.   If you land a job designing controls for an interferometer that doesn't mean that you need to earn a degree in physics.  You should however have more than a passing knowledge of how that device works.  Education never stops.

I'm rather shocked that you have gone through an electrical engineering program not designing anything.   I almost want to know the name of this school to make sure I don't suggest it to anybody.   This actually seems extremely odd to me.   If the school does any feedback you might want to imply that even a little bit of practical would have helped.   Back when I was in school the full time students did work study as part of the program so that the students would have a better idea as too what the career entails.   

Ultimately you need to read and design on your own and maybe blow a few things up.   The various manuals and datasheets vary from barely there to being a wealth of information.  In any event building things in an engineering way, again and again, is how you develop skills and maybe more importantly insight.   One of the reasons I suggest to people new to electronics to build their own DC power supply for their digital circuits is because that is a good low cost place to learn skills as a beginner.   Even if somebodies interests is primarily in digital circuits, building a small DC supply develops skills outside of electrical design also (soldering, component placement, heatsinking, housing design, AC systems & etc.).

If you haven't as part of the program learn some of the basic of computer programming!!   This was hopefully part of the program but you could be called upon to help out with code, or maybe more importantly can leverage a computer to do some work for you.   Like wise learn to really use and understand computer operating system.   That includes the UNIX likes (BSD, LINUX and Mac OS) as well as Windows.   I'm not sure how they got their degree but I've seen engineers stumble badly when face with a PC of some sort.   Then you have the guys that standout as absolute masters of their machines.   Learning to use a PC well is as important as learning to use an O'Scope inside out.

I mention the stuff above because you don't want to look like an imbecile when you get that first job.   That first job will entail working with a experienced person and you might get assigned to different departments to learn the business.   If the facility has a production line don't be surprised if you get assigned to work with production for a few weeks.   The better companies will expect you to understand their business and how your work will impact other areas of concern.   So if this happens don't dismiss it, rather use it to learn as much as you can.   That is some companies in any case you will work alongside the established engineer for some time.   Don't be surprised if you end up going through design reviews and have to justify your approach.

Don't be surprise by canceled project and don't always blame yourself.   If your engineering was the reason something got canceled you will be told.   What is frustrating is putting your heart into something and have the project evaporate before your eyes never to see the market place.    I've seen companies spend in excess of $300 million on a project that never produced a product for sale.   It can be extremely frustrating to have the rug pulled out from under you - do not take this personally.   

It is actually well past bedtime so I hope this doesn't sound like a ramble.  The point with some of this is that you will also learn what it is like in the real world.  By the way if the engineer you get teamed up with isn't meeting your expectations or you are not getting the work to develop your skills say something.   Communications is important.

[ElbowPics]

Wow, thanks for all the replies. I'm definitely going to check out more application notes, any schematics I can find, and work on building more of my own stuff.

To clarify I'm specifically asking about how to go from a description of what a circuit should do/accomplish to actually having the circuit. I don't feel like I have a good grasp on two main things:

1) Practical differences in real components
Placing filter capacitors as close as possible to IC power pins (just happened to have a lab demonstrator mention it one time)
Choosing one type of capacitor over another
Which parameters in datasheets are stable vs. which are likely to have a large uncertainty, change with temperature or frequency etc.
If you need to amplify a signal are you going to just automatically go for an omp-amp unless there's a reason you need discrete transistors?

2) An actual process to follow
Do you make sure to simulate everything beforehand?
Do you get an over-engineered, way too expensive version as a prototype first and then begin replacing sections with slimmer designs?
Is there a general rule for choosing between buying an IC to accomplish somethings vs. making your own circuit? How complex does it have to be to justify just buying a premade IC?
When selecting components do you have certain transistors you default to for certain tasks even if a slightly better one might exist?

I've never seen anybody go through designing a circuit, and while I can and have designed circuits I have no idea if they're "good" solutions or flimsy ones. Mainly, are they robust enough?

I think a good summary is: What is the difference between the textbook and the real circuits? And what are the general rules of thumb and shortcuts you use when choosing a circuit design?

I know that's a very broad question, and if it could be answered in a forum post electrical engineers would basically be replaced by programs in a year, but any advice would be great.

[Electro Fan]

ElbowPics, you ask very good questions.  You have a very bright future ahead of you and you are going to do very well. 

[twospoons]

Generally, alongside a functional description, you'll have a a set of constraints. Typically these will be things like cost, size, efficiency, availability of parts, time to market, patent issues, standards compliance (thats a huge one).  Your design will have to balance these constraints against each other - e.g. there's a chip that does just what you what but it might be very expensive, or have a long lead time. Do those things matter in this case?   These constraints will drive your system topology and parts choices.

When it comes to real-world parts, remember every passive has resistance, inductance and capacitance.  You just need to check those parameters fit your design. 
Datasheets will usually spell out the tolerances in each parameter, and often the test conditions too, and you will need to relate those to your design to check its OK. 

Choosing capacitor types comes down to which type fits the constraints of your project - size and cost are the most common drivers, but ESR, ESL and stability can come into play in some situations.

Opamps vs transistors: horses for courses again. You would struggle to find an opamp that handles high power, or high voltages or RF. But sometimes a simple BJT works just as well in low voltage, low power situations.

Simulations can be very useful, so long as you remember that simulators use "perfect" components. I simulate a lot of things, but not everything. Some stuff is simple enough that you know the math, and a sim is unnecessary for confidence.

I aim my prototypes to be reasonably close to the final product.  Theres no point prototyping with some gold-plated part you know you can't use in the final product.  Any extras at the proto stage would typically be there to assist with debugging the design.

IC vs bespoke is usually driven by the constraints I mentioned at the start. Frequently it is cheaper and faster to use a ready made chip.  But sometimes you just can't find something that meets all the constraints ...

I often use the same transistors because I know I have some, or another product uses it, simplifying the procurement process. If you've used a device before then you will likely have symbols and  footprints in your EDA, which saves time.   Its often not worth picking a marginally better device.  A significantly better device (in terms of the project constraints and circuit requirements) is worth having though.

The difference between theory and practice is than in theory there is no difference, but in practice there is!

Rule of thumb and shortcuts? Do your research to see how other engineers have solved your problem - you will find some cunning tricks, and probably think of a few improvements along the way.

The one critical piece of advice I give every EE grad is "Current flows in loops".  In other words, every signal has a return path, and you need to manage both. Sometimes parts of the loop are not obvious, sometimes they are unexpected.  Those are the ones that can cause trouble.

[Siwastaja]

"Professional" means: someone pays you for it enough so you make living with it, your profession. Nothing else. There is a temptation to add a lot of connotation to this word, let's not do it.

Many businesses can operate without stringent factory-like conditions, ISO9001 certifications and whatnot, long chains of management, submanagement, and finally you, the fresh engineer, on the bottom. There are other ways.

For example, various audio products (think about guitar effect pedals, etc.) have always been available from small and not-so-small garage-like companies. This is completely "professional" even if it doesn't fit the mental image of "professional" for many. Many of such endeavours produce utter crap quality, but some do not. They all have their business case.

I'm mostly being subcontracted to do design for small startups. It's a whole world on their own, no one has time and resources to play a "real professional factory", that would prevent any progress from being made. The time for it comes later, or it may never come, some companies succeed well doing one-offs or prototype batches in hundreds.

[TomS_]

At the risk of muddying the waters by adding yet another perspective, I will say:

How does a golfer end up winning the PGA? Or a racing driver end up winning the championship of their series? Or an ice skater become an olympian?

Practice! Rarely is anyone going to hit the ground running, so to speak. An "inherent natural ability" (aka "how your brain is wired") will for sure go some way to help - some people just pick things up easier than others, as if they were born to do it - but it will still take years of practice, learning, re-learning, and perhapst most importantly making mistakes to reach the "pinnacle".

You will have to have the perseverance to work your way through the tough early years of this process, and visualise the end goal - Im sure thats what a lot of coaches tell people who become sports stars....

Theres going to be a lot to learn, and some of it you may only be able to learn on the job, or some of what you have already "learned" may only truly make sense when you start trying to put it into practice in the real world.

And one way to put stuff in to practice is to work on personal projects. Maybe they dont look all professional, but as a personal project this is probably not warranted - it is not intended to be a professional looking product that will sell to the masses and survive all of the abuse of the real world. Whether or not you are going for a job, doing anything is better than doing nothing. But when it comes time to land a job, it shows to a potential employer that you are trying to learn and you are practicing.

Keep a hold of everything that doesnt work, too. You'll then be able to use that as an example of how you tried something that didnt work, and how you then iterated and redesigned to overcome issues.

And yes, there is sometimes no better way to learn than to observe what others do. Buy old bits of gear off ebay and tear it down, study the layout, the schematics (if available) and how it has been designed by "so called professionals". Take my word for it or not, but some "professional" products are going to have some very "unprofessional" design choices in them from time to time... See Daves video about a tablet PC that was meant for medical applications I think it was..........

Another thing to bare in mind is that in a professional environment there will often be a process of reviewing work. Senior engineers will review the work of junior engineers, and through this process you will learn a lot by having to re-do some of your work. Some of it may simply be "style" that is specific to one company, but other parts will be "best practices", "industry standards" or simply "you screwed that up" kind of insight that you can use anywhere.

I guess the short form is, you learn by doing.

[tggzzz]

You learn by understanding and thinking, and cement that learning by doing.

Both theory and practice are necessary; neither is sufficient.

[DenzilPenberthy]

My perspective on this falls between the "Get better at circuit design" and "Learn about business stuff" advice that others have already given.

An important difference I'd say is the difference between making one of something that works when it runs on your bench, and making large numbers of things that work reliably in a variety of environments and are easy to manufacture.

Hobbyist stuff typically stops at the point in the process where your circuit works. The professional engineering is then done to make that circuit work at -X to +YdegC with x level of vibration, it works if you plug it in backwards, board manufacure has >X% yield, it doesn't have any single-source parts, it's been cost-optimised, it has a test system designed for it,  it didn't need unnecessary development time because you made a component footprint mirrored on the PCB, etc etc etc..

It's all quite unglamorous work but it's where the engineering happens.  Dave has some great videos on Design-for-Manufacture etc. Robert Feranec has a fantastic set of videos of short tips, and many longer videos about processes for rigorously checking schematics, creating a formal Board Release Procedure, etc etc etc.

 

[Cerebus]

"Professional" means: someone pays you for it enough so you make living with it, your profession. Nothing else. There is a temptation to add a lot of connotation to this word, let's not do it.

I'm just going to briefly take issue with that. Whether you like it or not, in English, "professional" does carry connotations beyond merely being paid for one's work, depending on context. One can be a paid professional and produce amateurish work, one can be an amateur and still produce work of professional quality. "He did a professional job of it." is a compliment irrespective of whether anyone's getting paid or not. "The professions" are occupations that require advanced formal training or study to undertake such as medicine, law, and so on; cf "Professor".