Schiit Happened: The Story of the World's Most Improbable Start-Up
Aug 1, 2018 at 8:56 AM Post #35,911 of 145,687
I keep getting adds from Mouser in the Raspberry Pi 3 with various accessories, I am going to have to check one out sometime. I have an Ipad Pro and various devices for outside music sources but I like the idea of something less expensive dedicated as flac storage.

Perhaps I could even buy all the parts and have someone set up a nice music source for me and I could repay them with some work of my own? Maybe someone from Texas with a vast amount of computer knowledge?:)

Finally getting some ripened home grown tomatoes so caprese salad today!
 
Last edited:
Aug 1, 2018 at 9:10 AM Post #35,912 of 145,687
[QUOTE="wink, post: 14395296, member: ...

Catch and release works every time..... :ksc75smile::ksc75smile::ksc75smile::ksc75smile::ksc75smile:[/QUOTE]

I spent too many years, too many hours, and too much money, to land this one and keep it from swimming on my wall.

Besides, releasing a fish this size after a 30 minute fight usually means its over for the fish anyway, especially if you drag one up from from the deep, and this was a 4 year old sport so at this point they find a stream to swim up in the fall and die without mating anyway. The smaller the fish the better its chances when caught and released.
 
Aug 1, 2018 at 9:18 AM Post #35,913 of 145,687
Sacred cows make the best hamburgers.

I recommend Yggy based on sound- nothing more, nothing less. You'll note I have Freya and Vidar (x2), so my suggestion comes based on knowing how they all work together. Your speakers will certainly offer far more detail and precision than mine, further reason to orient towards the Sound. A suggestion offered in friendship.


Never say never!

Thanks
 
Aug 1, 2018 at 9:36 AM Post #35,914 of 145,687
Perhaps I could even buy all the parts and have someone set up a nice music source for me
1. Setting up a Pi-based (networked) source; the only difference in my work setup is that I have a Pi 2 Design 502DAC HAT, which is a bit hard to get (it's made in small batches when the Sun shines :wink: ) but very good. Allo.com has the comparable DigiOne HAT, which I've not used, but I have Allo's USBridge on a different, networked, streamer.
2. Installing custom piCorePlayer software for not networked, local disk use.

Happy to provide more info and building/troubleshooting advice.
 
Last edited:
Aug 1, 2018 at 9:47 AM Post #35,915 of 145,687
Hmmm I may have to get involved with these some. Surprisingly my son has not played around with one yet and he has been programming since an early age.

I appreciate the offer, and those who know me here know I will generally help out in specific areas if ever asked. I have been known to build a Coaster amp or two and even an assorted cable on occasion. As far as electronics I can get around pretty well, mainly audio and video and I have even constructed usb and hdmi cables. (nothing mini or micro though).

I am basically looking for a portable music storage device with a small color touch screen and perhaps AVS capabilities. (I would have to put up a wi-fi antenna or set up a personal hot spot).
 
Last edited:
Aug 1, 2018 at 10:18 AM Post #35,917 of 145,687
[QUOTE="wink, post: 14395296, member: ...

Catch and release works every time..... :ksc75smile::ksc75smile::ksc75smile::ksc75smile::ksc75smile:

I spent too many years, too many hours, and too much money, to land this one and keep it from swimming on my wall.

Besides, releasing a fish this size after a 30 minute fight usually means its over for the fish anyway, especially if you drag one up from from the deep, and this was a 4 year old sport so at this point they find a stream to swim up in the fall and die without mating anyway. The smaller the fish the better its chances when caught and released.

Maybe you should put a ruler and thermometer next to it for our reference.
It could be a harmless guppy on that picture for all we know.
 
Last edited:
Aug 1, 2018 at 10:50 AM Post #35,918 of 145,687
2018, Chapter 8
How to (Maybe) Succeed in Engineering (Without Going Insane)



Does it seem like there’ve been a lot of engineering chapters lately?

Why yes, yes there have.

This is partially a reflection of this year. 2018 is, for me and the whole team, a heavy engineering year. Mike is busy working on a whole lot of long-term things that will ensure that we can continue to provide unique digital products (translation: to our own whacky preferences and whims) with very high value (translation: cheap.) I’m chasing down a couple of new ideas on the analog side, new ideas that will allow us to extend our unique analog approach (translation: our obsessive insistence on discrete and linear stuff), and also allow us to increase efficiency (translation: even more cheaper-er.)

And, at the same time, we’re both working on new actual products, and dealing with the inevitable tweaks and changes needed in a world where the “global parts shortage” is a real thing. Many of our standard parts have to have “alts” qualified. For the most part, these parts are not super-critical, things like bypass caps and stuff like that. But it could be worse. We’ll see.

So yeah, engineering is top of mind.

Plus, I just wrapped up a series of chapter on the engineering on the coaster amp, going through the design goals, decisions, and trade-offs for one very simple product, and running into problems that aptly illustrate the complexity of even the most basic designs…

…because there’s nothing like putting more engineering on top of engineering, when engineering is a full-time job.

Hey, I never said I was the brightest person in the world.


So, How Do You Keep From Going Crazy?

When I was having one of my worst days—multiple requests to qualify alternative parts, a prototypes that just plain refused to work, the discovery of a new intractable problem in a very late-stage prototype, and the dawning realization that I didn’t have enough time to devote to the turntable, I remembered some of the stuff that Ed Miller told me when I first got started at Sumo (which was really an into-the-fire kinda job), and I realized three things:
  1. Every engineer has days that suck absolute balls.
  2. Dealing with those days can be absolutely humiliating.
  3. No matter how bleak it seems, you’ll make it through.
And so, I thought…why not a chapter for engineers (and engineers-to-be) on how to deal with those days without going completely insane? Or, maybe a bit more…what to expect from the job on both the best and worst days? Heck, maybe even discuss what engineering is and what it isn’t, for those people who don’t know if they really want to commit to it as a degree or as a career?

So yeah, a chapter on “How to Succeed in Engineering” or something like that. Except that sounds way too rah-rah, way too Dale Carnegie, way too self-help to use without throwing up on myself a bit.

Plus, engineers who say they have all the answers are usually the ones you want to stay far, far away from.

Aside: There are no more terrifying words from a new team-member than “Yeah, I’m a great engineer!” Especially when said with dead-seriousness. Run. Fast.

Because, like it or not, this job is humbling.

So, maybe something like “How to Succeed (or at Least Survive) in Engineering.”

Yeah. That’s better.

Let’s do this.


Eyes Wide Open: What Engineering Is (and Isn’t)

If you’re expecting me to regurgitate the definition of engineering from the OED, forget that.

Heck, I grew up in a world where “engineer = runs a locomotive,” at least according to the popular media. Yeah, a guy in dirty coveralls, running a nasty coal-caked cast-iron behemoth from a prehistoric age. Not something I wanted to be. Not an appealing image at all.

Aside: when I started my major in engineering, I still had friends who asked if I was going to run a train. That image was engraved on people’s minds. That’s the power of programming (via a television, thanks to lack of content requiring the re-runs of ancient shows.) It’s a very, very different world now.

Now, growing up, I also knew my dad was an engineer, and that he did some stuff with electronics (at a number of companies, culminating at Pacesetter Systems, which made Pacemakers. I didn’t know exactly what he did, but some of the stuff seemed cool, especially the tiny circuits (you should see a pacemaker from the pre-surface-mount era….wwwwwwoooozersssss!) Hell, I didn’t know that one of his earlier jobs was for a contractor working on the US space missions, and that he actually had a commendation from NASA for his work on the Apollo programs.

Very, very cool stuff. But it took him passing awayfor me to find that document. And, in some ways, that’s a sad metaphor for engineering: you get to do some really terrific stuff, but it can be lost in the day-to-day grind…or the terrible days when nothing goes right.

So, if this can help you get through those bad days, and focus on the cool stuff, then this chapter has done its job.

“But you were going to define what engineering is, right?” someone reminds me.

Yep. Absolutely.

Here we go: engineering is where you get to design cool stuff, and make it work.

“Well, that seems generic,” you might say. “Aren’t there designers that, like, work on what stuff looks like? Aren’t there engineers who do, like testing and quality and stuff like that? And engineers who do all the documentation?”

In short, yes, yes, and yes. But if you’re talking about very specific engineering tasks, you’re talking about large engineering groups…and those groups are designing cool stuff, and making it work.

But let’s break the statement down:
  • Get to design: engineering is a creative endeavor. You are creating something. Typically something new. Yes, even if it is a next generation of an existing product. There’s always going to be a different angle, a new technique, a cutting-edge component to consider. If you don’t think engineering is creative, then you’re missing the point.
  • Cool stuff: engineering is where you apply your knowledge (of electronics, materials, chemicals, etc) to make something cool. Note that cool doesn’t have to mean “a brittle phone that easily slips through your fingers.” It can also mean, “A product that can take the abuse of the Rubicon” or “a new coating that makes the product feel unlike anything else.” If you’re not shooting for your version of “cool,” why bother with engineering at all?
  • Make it work: engineering means that it doesn’t end at “design” and “cool.” It’s not a poster, a piece of art, or a concept model from the industrial design house. It’s not a spacesuit as shown in The Orville, where you’re wearing it in full atmosphere and the biggest danger is the actor getting too hot, it’s a spacesuit where you’ll die if it doesn’t work. And this is the sting—if you aren’t interested in taking it through to the “make it work” phase, maybe best to stay on the concept art side. No shame in that.
Aside: There’s actually a shocking amount of engineering that goes into practical effects on sci-fi shows. Hell, they built those ugly person rover things for Lost In Space, too bad the whole thing hinged on the cast being dumb as posts. When the friggin ship is sliding off a mountain, grab the bag and run, don’t just stand there staring at photos. And the light teepee. Holy hell, my wife (who is a lighting designer) yelled in frustration and left the room.

Aside to the aside: but I digress. What I was saying was that there is quite a bit of engineering in practical effects. I was asked by a friend with an effects company to come up with a better way to light up electroluminescent strips for the Tron sequel, because the old suits were, ah, shocking the actors. I didn’t have time to complete the project, but yeah…boosting them to 100VAC could be bad news when the actors started sweating. The same friend also did design work for the Tesla spacesuits and capsule interiors, so there is crossover…but he also worked with engineers for that.

“Well, I still think your definition is pretty basic,” you say.

Yep. Cool. Okay, let’s look at this from another angle. Like, say, what engineering is, and what it isn’t.But let’s start with what it isn’t.

What engineering isn’t:
  • It’s not science. Surprised? It uses science, yes. But if you think you’re going to be doing basic, fundamental research, or doing something like, say, creating your own transistor design, you’re sorely mistaken. Move on.
  • It’s not touchy-feely. Or, more precisely, touchy-feely stuff about how everything is relative and it’s not fair that the simulation doesn’t work won’t go down so well. Yes, the laws of physics are inflexible and will not be bent (at least on the level you are working at—no matter how good your argument is, no matter how much you wishit to be.) Remember: and make it work.
  • It’s not as hard as you think. Going through school, you might be put off of engineering forever. Differential equations. Matrices. Fourier transforms. Z-domain. Thermodynamics using calculus. Probabilistic systems. The reality is that you may never use any of those. You’ll use the principles behind them, absolutely. You’ll use the shortcuts that keep control systems stable, sure. But if you think engineers do differential equations every day, you’re profoundly mistaken.
  • It’s not as easy as you think, either. Breathing a sigh of relief? Don’t. Remember: and make it work.Depending on what you’re engineering, “and make it work” can mean, “make it so it doesn’t fall out of the sky at 80,000 feet,” or “make it so we can make 1,000,000 of these less expensively than the competition, or “make it so it benchmarks better than everything else coming out at the same time—oh, and by the way, you don’t know what all those other engineers are designing.”
What engineering is:
  • It’s subjective. Surprised? Again, you shouldn’t be. There is no One True Way to engineer Your Cool Product. There are a million ways to design something…and a million tradeoffs. Engineers choose radically different solutions based on their personal preference, experience with past designs, knowledge of current and upcoming technologies and techniques…and also based on the size of the company, target price of the product, time given to turn the design, ergonomic factors, and many other intangibles. None of these million ways to design something are 100% wrong. (But then again, none of them are 100% right. No, not even with unlimited budget, time, and target price.)
  • It’s a rush. If you love creating things, there’s nothing like the rush when a design runs for the first time. It’s something you made! It’s completely new! It may be a little wonky, it might run a little hot, it might oscillate a bit, oh hell it’s on fire! But it ran! It really ran! Sooooooo cool! I always take a photo of a new design when it’s finally running. Which means my photo album has pictures of really ugly PC boards interspersed with cool hiking shots.
  • It’s more than a job. Engineering teaches you how to think analytically—and, even more importantly, it teaches you how to make things work. This reflects on your entire life. Most of the engineers I know are very much can-do people who tend to do well, even if they’re thrown a curve-ball. I can’t imagine not being able to work on my own car and house, even if I sometimes don’t have the time and inclination to do so.
  • It’s profoundly humbling. And yes, there are those days where nothing goes right. Nothing works. Nothing runs. And the work just keeps piling on. And, when you love to design stuff, this is the worst time ever. You’re an idiot. Worthless. Can’t do one single thing right. Might as well have taken that car sales gig. Don’t get into engineering if you aren’t comfortable failing, because there will be some soul-crushing times.
So, what do you do on those special, special days? Those days when you think—no, you know—you’re the worst engineer in the world?

Wait a bit. We’ll get to that. First, let’s talk about most days.


Rules (or at Least Suggestions) for Most Days

First, a definition. “Most days” are typical days in your engineering career. You’re working on something, but there’s nothing exceptional going on. There’s no gigantic crisis, no huge breakthrough, no successful launch, no intimation of an uncrushed bug like the thermal throttling on the new MacBook Pros (but I digress…)

You know, just another day. Now, these days can be fun, because you can finish parts of stuff, figure out an elegant way to solve a small mechanical problem, build a new prototype, or lots of other things engineers find interesting (yeah, I know, if you’re not an engineer, you probably don’t get this—that’s cool, I don’t understand accounting, but I know some people think it is loads of fun.)

On these days, I go by what Ed Miller might have called the “Three Rules for Sanity in Engineering,” if he lived in an era of listicles and YouTube videos.

These rules were as follows:
  1. Take the win.
  2. Take the hope.
  3. Take the lose.
Huh? Let me expand.

Take the win. Okay, you finished putting together a new prototype. Or maybe you finally figured out how to get that &^$&^@!* heat out of the system. Or the first article of a new product turned on and ran without issue. Those are all wins.

You feel good about yourself! You feel, briefly, invincible! And, feeling invincible, you want to go a bit further. You want to spark up that new prototype and see how it goes. You want to draw up that thermal solution in detail. You want to fully qualify that first article.

In brief: DON’T.

Stop, go home, have a beer if that’s what you’re into, spend time with your family or friends if that’s your thing, and bask in your small victory.

Because if you go farther, you may find the prototype goes up in smoke, the thermal solution isn’t workable in detail, or the assembly house boned the first article. Then your good mood is shot. You’ll work too late trying to recover that good mood, then you’ll come home late and pissy and irritate the hell out of your friends and family. Then you’ll toss and turn at night, not sleeping, so when you come in the next day, you’ll be even more irritable and sleep-deprived, and in no state to work on anything.

No. Take the win. Go home happy. You can piss yourself off tomorrow.​


Take the hope. This is when you’re working on a difficult problem—say, one of those prototypes that just refuses to work correctly. You’ve been beating your head against the wall for days (or at least it feels like it). You’ve checked for bonehead mistakes in your design. You’ve done all your normal tricks. But it still isn’t working.

Then, the clouds part, and god (looking like a Monty Python cut-out, perhaps with Mike Moffat’s head superimposed) reveals an Entirely New Way of Looking At Things, one that is almost Certain To Sort Things Out.

Now you’re happy! And now, your natural reaction to this is to try out this new thing right now! Because that’s gotta be it! It’s exactly what you need to do! That’ll fix the problem!

Yeah, well, except you missed the “almost” in “almost Certain To Sort Things Out.” And now your one great hope has been dashed, and your happiness implodes like a beer can in the Marianas Trench.

Again, DON’T. Take the hope, and go home.

Enjoy your time of certainty. Be happy that you know exactly what you’re gonna do tomorrow. Have that beer (or wine, or jun, or milk, or whatever the heck you drink to celebrate) and smile at your family and friends.

Because tomorrow is another day. And you could be right!​


Take the lose. This is when something doesn’t go right. If you’re an engineer, you know what I’m talking about. The new run coming off the line isn’t meeting spec. The prototype went up in smoke on first turn-on (durr…get a variac, bonehead). The rules changed for the new product design. There are plenty of things that won’t go right. That’s part of the job.

In this case, your first impulse might be to deny it. The spec is too tough, the person who assembled the prototype is an idiot, the change of rules isn’t fair.

DON’T. Accept it. Again, don’t get into engineering if you aren’t prepared to fail.

But even if you’ve accepted it, your natural reaction may be to throw yourself into solving the problem right hereand right now, even if it happens to be 5:30PM on a Friday evening. Or, if it’s mid-day, you may decide I’ma gonna fix this sucker by the time I leave, period, no exceptions.

Again, DON’T. You’re only going to make yourself miserable.

Sure, work on the problem if you’re still wide-awake and fresh. But if you’re already dead-tired and done, you may make it worse. Or, at best, you may make no progress at all. And then you’re even more tired and irritable, with no solution.

So yeah, there are times you take the lose. Do what you can not to dwell on it too much, without drinking yourself into a stupor. Don’t take out your frustrations on friends and family.

Because tomorrow, again, is another day. And that day you may be taking the win.​


Now, Ed never really codified these Three Rules into anything formal. He just stated one of the three rules whenever appropriate—when I was happy something worked, or when I thought I knew how to fix something, or when something had gone pear-shaped.

And, at the time, I never realized that, taken together, his Three Rules really were nothing more than excuses to go home on time.

But, you know what? He has a point. Engineering can eat your life. Get some perspective. Be with your friends and family. Go home. Enjoy your life.

“So is that it?” You might ask.

Nope. I’ll add a few rules of my own. These work for me. YMMV.
  1. Keep accessible notes.
  2. Be methodical, at least as much as you can be.
  3. Give yourself permission to explore.
Again, let’s expand.


Keep accessible notes. Oh god oh god oh god. No notes? How the hell do you ever get anything finished? It’s impossible to remember all the details—especially when you’re chasing down a particularly intractable problem.

For most of my engineering career, I kept engineering notebooks. As in, actual pen-on-paper notebooks. These were great to refer back to and see what I’d tried, and what the results were. But in an age of Evernote and OneNote, this approach sucks for many reasons:
  • If you’re working on multiple projects, how do you keep them separate? Separate notebooks? Oh hell that’s unwieldy.
  • What happens when you get an inspiration in the middle of a project? How do you find it later?
  • How do you add data to it, like analyzer output, test results, etc? Paper clips? Printouts? Arrrghhhhh.
  • How do you add datasheets to it? How do you add useful links to it?
Hence, keep accessible notes.

Using a note-taking platform like OneNote or Evernote allows you to add links and datasheets and links and photos you took and thermal camera images and Audio Precision reports and dozens of other things.

It also allows you to separate your projects more logically, as well as search your notes and find that whacky idea you had a few months back, but were too lazy to break out.

Sounds like a lot of work? It isn’t. Both of those note-taking platforms are available everywhere—on your PC, Mac, phone, tablet, etc. If you like scribbling, they’ll happily take stylus input from a Surface Pro or iPad.

I waited farrrrrrr too long to switch over. Don’t be lazy. Just do it. You’ll thank me later.​


Be methodical, at least as much as you can be.Yeah, I know, you’re The Nutty Professor. Or whatever. That’s cool, if you get results and your company is doing well because of it. But for the rest of us, some methodology can help weed out stupid ideas, half-baked solutions, and didn’t-find-it-gotchas.

For example, I have a pretty standard methodology for evaluating prototypes. I have a template where I take the same measurements and compare them to stuff that we know works in production. These measurements include:
  • Thermal measurements—how hot do the components get? How does this compare with what the datasheets recommend? Is it acceptable to the owner (transformers are insanely resistant to high temperatures, but a design that relied on running a transformer at 125 degrees C, while OK per the manufacturer, wouldn’t work very well when your customers ended up in the burn ward.)
  • Regulation measurements—how much headroom do the regulators have? Where does it go out of regulation?
  • Noise measurements—starting subjective then on the analyzer.
  • DC measurements—starting on the meter, in this case.
  • Rise time/square wave response—to see if it’s overdamped or underdamped, very useful in tweaking compensation, because compensation is almost never what you simulate, due to the real-world parasitics of the board.
  • Standard analyzer suite—THD, power output, noise, etc. Although sometimes our products won’t measure well on a traditional scale (comparing, say, a Vali 2 with a high-feedback, op-amp based design, for example), we need to know how they perform. And if they don’t perform to expectations, something is wrong.
I also have a pretty standard methodology for troubleshooting, when things go wrong on the prototype side:
  • Check the parts. Usually it’s because someone put the wrong part in the wrong place. Usually me, but both Tony and Naomi are building prototypes now, so it’s not always me. This can be kinda a sucky job due to cryptic (or missing) surface-mount markings.
  • Check the schematic. Did I screw up the layout, maybe stick an NPN where there should be a PNP? Yeah, it happens.
  • Check the layout. Do the component pinouts actually, like, match the datasheet? Yeah, this happens too. Especially when I’m thinking, “hmm, maybe JFETs, hmm, maybe BJTs…”
  • Check the volts. If it’s stable enough to bring up to full voltage (that is, not getting hot/smoking), it’s probably due to a bum power supply. It only takes swapping two resistors to take down a supply.
  • Check for oscillation. Believe it or not, you need a decent scope to find some forms of oscillation. And oscillation can hide. We had a recent case of it hiding between a regulator and a cap multiplier—the regulator was singing at 2MHz. The only clue was lower output on one side. The 2MHz didn’t make it into the main supply.
  • Check for variance. What happens when we swap parts? What happens when it gets hot? What happens when a value is a little off? Things vary in production—but the products need to work the same.
And yeah, there’s tons more to check, but I think that’s tons—plenty to get you started.

Give yourself permission to explore. Engineering can be a grind—if you let it. You probably got into engineering because you wanted to make cool stuff.

So…make cool stuff.

Try new things. Use the latest parts. Or use the oldest parts. Do something totally different than what you usually do. Explore the new hot topology of today. Or dredge up an old one from the past.

If you can’t do it at work, try it on your own time. It doesn’t take much to set up a simple workstation at home. Test equipment has never been cheaper. Design software is as cheap as free (though yeah, good, usable 3D CAD is still pricey, especially if you want to do sheetmetal.) Boards have never been cheaper (or faster.) There are more options for one-off mechanical stuff than ever before, from 3D printers to on-demand, bespoke CNC.​

You’re an engineer. You design cool stuff and make it work. So do it! Give yourself permission. Have some fun.​

(But…remember Ed’s three rules.)


Getting Through the Worst Days

There’s no way to sugar-coat this. There will be days when everything goes wrong.

Your latest prototype went up in smoke, production has multiple issues with current products, the competition just introduced something that leapfrogs what you planned to release next quarter, one of the parts you’ve been betting on just went NRND, and you have a dozen parts that need alts because the lead-times just went out to 37 weeks.

And yeah, your self-confidence is shot. You’re an idiot. The worst engineer in the world. How do you possibly recover from all of this? It’s literally the end of your career!

Well…probably not. Because every engineer has worst days.They’re like death, taxes, and metal vendors screwing up.

So what do you do?

Well, if you’re not simply going to say “heck with this” and go sell cars or real estate or start a plumbing business or a bakery, then maybe these steps will help. This is what I do. Again. YMMV.
  1. Realize that it will get better.
  2. Knock out something easy.
  3. Don’t hide the problem.
  4. Bring in some help.
  5. Start over.
By now, you know how this goes. Let’s look at each of these in more (some might say excruciating) detail:


Realize that it will get better. This might be the hardest part, especially when it’s one of the first worst daysyou’ve ever had. But yes, even when things seem bleakest, it will get better. You’ll fix the problem, find the alternate parts, re-spec to beat the competition. That’s what you signed up for (remember that “make things work” phrase).

So. Yeah. It will get better. You will work things out. If you’re near the end of the day, time to go by Ed’s rules and take off. Don’t beat yourself (or your friends and family) up, though. Come back to it fresh the next day, and everything may seem a lot better.

Or, at least, some things may fall into place. And that’s a start.

Once you’ve gotten a few worst days under your belt, it’ll be easier to take a deep breath, sigh, shake your head, and accept that you’ll make it better…

… but, maybe, like, tomorrow.


Knock out something easy. Okay, so you’ve accepted that it’ll get better, but its not time to go home, and all hell seems to be breaking loose. What do you do then?

Easy. Find something easy to fix and get it off your list. That gives you a win, and that alone will improve your mood…and your chance of fixing other stuff.

In the example above, qualify the alternate parts. This is relatively straightforward and eliminates some list clutter. It takes it out of your mind. Then maybe move to the NRND stuff. Is it a real NRND, or is it just a process change? If it’s a real NRND, when is the end-of-life happening? Are there better alternatives that you missed? Knocking this out is another win, and another mood-booster, which sets you up for the harder stuff.


Don’t hide the problem. Do you have other engineers who can help you, or are there people farther up the chain that should know? Once you confirm that yep, your competitor has scooped your release, or your new prototype really, really ain’t gonna be working anytime soon, make sure they know about it. Better they know now than later. And, who knows, they might even have some suggestions.

Bottom line is, you have everything to gain and nothing to lose by being transparent. Hiding a problem is a bomb that may go off in your face.


Bring in some help. You also gotta know when to say “when.” Super-complex systems have a way of baffling the best engineers. And some engineers are much better than others at certain things. You wouldn’t ask me about a digital filter or firmware problem, for example.

If the problem is caused by something that’s out of your depth, it’s time to bring in some help. This may be help from a device manufacturers’ own application engineers (though if you’ve wandered far off their reference design, they may look at you blankly and offer little assistance), or it could be help from within your own company, or it could be from specialized consultants.


The main thing is that once you’re at the end of your experience, it’s time to lose the “not created by” or “not invented here” stigma and see if there’s someone who can help. Remember: “and make it work.”


Start over. I knew of a creative director who consistently got amazing work from his designers (this is on the agency side, not on the engineering side, but hold on—this is relevant), but at the cost of his staff hating him. When a designer was stuck on an idea—maybe a great idea, but they couldn’t make it work—he’d delete those files and make them start over again.

Aside: No, this wasn’t me…eeek, I don’t want to know what turnover was like at that studio.

But, here’s the point: sometimes you design yourself into a corner. And sometimes, you can’t get out.

And in that case, it’s best to start over.

“Oh gawd, that’s the hardest thing to do!” someone cries.

Yeah. I hear you. But when you’ve exhausted all of your tricks, reached the hinterlands of your knowledge, let everyone know you’re stuck and found no ready answers, brought in specialized help and baffled them as well…then, you know what? It’s time to schiitcan the idea and start over.

Yes, it’s painful. And yes, I know how you feel. I did that recently with a new design (I mentioned it in another chapter). But the new design simply didn’t address all the problems with the old design. And it simply wasn’t behaving the way I wanted it to. So, after months of development, I killed it and started over.

I get it. The design is your baby. But sometimes you have to kill your babies.​

So, in short, realize that things will get better, keep your spirits up by doing some easy stuff, make sure people know there’s a problem, bring in help…but absolutely, positively, don’t be afraid to start over.


A Sanity Mantra

Despite all I’ve written here, some of you will still be at wits end when faced with another of those worst days. How do you deal with it—with failure that seems to be on an epic scale?

Worse, what happens if another engineer starts attacking your work, saying you’re incompetent and irrational and that your whole approach sucks?

Yes, this happens.

Because, remember: engineering is subjective. There are a million ways to do something. Some engineers are going to think your approach is absolutely idiotic. Some will even attack you for it, especially when they’re going through one of their own worst days.

Aside: when I got married, I thought it a bit simplistic and utopian when Lisa called me out on “spreading negativity” when I had a “worst day” (yeah, you get them in the agency biz as well). But you see now I’m counseling people not to go home and make life bad for friends and family because they had a bad day. The reason is simple: she’s right. Spreading negativity comes back to you in spades…hell, it can ruin your life.

So here’s how I stay sane on those worst days. What I call my "sanity mantra." It goes like this:

No engineer is always right. Not even you.

Your engineering manager who just called you an idiot because you didn’t do it his way? Yeah, not always right. Your screwup that started your Worst Day Ever? Yeah, you’re not right all the time either.

Repeat after me: no engineer is right all the time. Never. Nope. Notgonnahappen.com. There is no such thing as a perfect engineer. Engineers screw up all the time. Worse, they’ll judge you on how they’d do it…and that might be wrong too!

Then repeat 10 times: that goes for me as well.

Yes, it really is that simple. Engineering can be a very, very humbling field.

As I’ve said, don’t get into it if you can’t deal with failure.

Get into it because you get (the capability) to design (amazingly) cool stuff, and (get the incredible rush of) making it work.
 
Last edited:
Schiit Audio Stay updated on Schiit Audio at their sponsor profile on Head-Fi.
 
https://www.facebook.com/Schiit/ http://www.schiit.com/
Aug 1, 2018 at 11:03 AM Post #35,919 of 145,687
"And, at the same time, we’re both working on new actual products, and dealing with the inevitable tweaks and changes needed in a world where the “global parts shortage” is a real thing. Many of our standard parts have to have “alts” qualified. For the most part, these parts are not super-critical, things like bypass caps and stuff like that. But it could be worse. We’ll see.

So yeah, engineering is top of mind.

Plus, I just wrapped up a series of chapter on the engineering on the coaster amp, going through the design goals, decisions, and trade-offs for one very simple product, and running into problems that aptly illustrate the complexity of even the most basic designs…" (Jason Stoddard)

I just got an email from Mouser, arrival dates for parts ordered for Coaster amps in June are now pushed back to September-November and that amp does not have that many parts. It is obviously rough right now for larger and more complicated amps. Many are common parts that are in short supply, yikes!
 
Aug 1, 2018 at 11:45 AM Post #35,921 of 145,687
Excellent write up Jason, as a newbie engineer in a totally different field (RF compliance testing). I found your advice useful and very relevant. Start using a pen and engineering notebook and recording everything and anything useful is the best advice I would give to all to be engineers. Gotta take them notes, or you are bound to forget it.

I would love to move into design and R&D one day...
 
Aug 1, 2018 at 11:51 AM Post #35,922 of 145,687
Maybe you should put a ruler and thermometer next to it for our reference.
It could be a harmless guppy on that picture for all we know.

Not a fish story, 30 pounds 4 ounces, 42" nose to tail

upload_2018-8-1_11-46-34.jpeg


upload_2018-8-1_11-48-17.jpeg
 
Aug 1, 2018 at 11:57 AM Post #35,923 of 145,687
Jason thanks for the chapter, good advice for new engineers. I used OneNote for the last 4 years of my engineering career (greatest thing since slice bread) and still use it for day to day notes in my retirement.
 
Aug 1, 2018 at 12:05 PM Post #35,924 of 145,687
Not a fish story, 30 pounds 4 ounces, 42" nose to tail

Kind of strange place to hang it.
In the kitchen above the dinner table maybe?
Now every guest you have sitting in frond of you gets major distracted by that fish, waiting for it to swim on.

Nice Schiit-stack. Almost identical to mine.
 

Users who are viewing this thread

Back
Top