Electronics > Projects, Designs, and Technical Stuff
EMF pickup from amplifier in I2C line causing glitches. (Now with scope trace!)
Monkeh:
--- Quote from: Starlord on July 06, 2016, 12:37:09 am ---Well PartSim is free, and I'd say it's better.
--- End quote ---
Say that when their site is down or they decide to take it away and you lose all your simulation data, and all your experience.
--- Quote ---Also a quick google search turns up a number of free options. Multisim for example looks much more advanced and there's a free edition:
http://www.ni.com/multisim/
--- End quote ---
... for 30 days, afaik. It's a trial, not a free edition.
--- Quote ---I suspect people just recommend LTSpice because it was the first and it's what they're used to and learned in school, because I refuse to believe it's the best of the available free options in this era of open source software.
--- End quote ---
'it was the first'? Not by a long shot, SPICE tools date back over 40 years. You could try Qucs if you like, I promise you it's no more friendly, just different.
T3sl4co1l:
--- Quote from: Starlord on July 05, 2016, 10:28:37 am ---
--- Quote ---If anything, I'd say the FCC ably performed the job they're charged with. Though it's not clear how long it took them to crack down.
--- End quote ---
Well going by the copyright date Someone listed, and the date the Wikipedia article states they took action, then it sounds like they got away with it for 2-3 years, assuming they weren't forced to immediately change their design.
--- End quote ---
And with unknown outcomes -- discard/recall/fine users/fine mfg/etc.
If nothing else, it shows they had to redesign the product; if they had done it right in the first place, they could've saved that NRE. (How much did it really cost? Obviously, not enough for them to tank. Was it less than the initial sales, with no loss of customers, so it was a net win? Did it hurt instead? No idea.)
--- Quote ---Well, I have plenty of arguments to make regarding that:
Consider Uber, AirBNB, and Paypal. And all those handling bitcoin. All businesses which were/are outright illegal under current codes, or whose legality was in an extreme grey area. These companies exist and are worth billions because the people running them didn't let a few pesky laws deter them.
--- End quote ---
Ah, now we're getting somewhere! :)
Gray area, yes. Illegal, no. If they sat there, asking lawyers and politicians whether their thing is right or not, they wouldn't have gotten anywhere. Better to put it out in the market and see how well it does.
And indeed, they did quite well.
But be very mindful of the humongous selection bias: these are four successful companies, out of how many thousands that fail for various reasons?
How about another selection: Silk Road. Obviously, facilitating the sales of illicit products is suuuper illegal...
Likely a lot of white-to-gray market stuff was sold there, too. Does that make it wrong?
Well..... yes, as it turns out.
--- Quote ---Now add Segway and all those electric scooter/skateboard/hoverboard companies. All selling products whose use, under the law as written, are illegal.
--- End quote ---
Don't believe I'd ever heard the Segway called illegal, since its introduction. But that was a long time ago, and I didn't follow the news about it.
Here's a current appraisal:
https://www.hg.org/article.asp?id=36294
TLDR it's local only, YMMV.
As for the Hoverboards -- that's under a different rule than present subject.
Or, I'm honestly not really sure if it's a rule or more of a suggestion, but in any case: UL approved products must meet design and inspection standards, and must be tested in a lab (again, for a pretty penny..).
Relevant blurb: http://www.ul.com/hoverboards/
As with other rules, if you don't pass the certification, you're at risk of getting pricked with a nasty action. Getting sued for bodily and property damages, by just one person, would be a rather bad sight worse than getting dinged $5k by the FCC, or having to recall your products. So, UL safety is rather a bigger concern.
What you get from being certified, is a guarantee that, if someone comes after you for xyz reason, you simply point to the test results and say "sorry dude, lab said go". Zero liability.
That's a HUGE deal in the US. Most business operations can be followed, not so much even by the flow of money, but from the minimization of liability.
Again, manufacturers, importers and sellers are at risk here. It would be unwise for them to import anything that's uncertified (or counterfeit).
It looks like a Hoverboard lawsuit is brewing, but as these sorts of things happen, it'll probably be a few years (and however many more clients) before it closes. So until then, the jury's literally still out (well, literally only if it goes to trial).
So... you're saying you'd much rather make a few bucks now, and... somehow dodge or neglect or avoid or abdicate your eventual liability from selling something uncertified..? Because if that's your business plan, you'd better have an aggressive plan to grow your business significantly by the time those liabilities hit.
It's like getting a loan, except the bill comes randomly, at an unknown interest rate...
--- Quote ---Well, if the US / UK wants to remain a competitive player in a global market, in a world were more and more people, like me are able to design and sell electronics online, people who, like me, can't afford to spend over $14K on testing alone (https://www.quora.com/What-is-the-average-cost-to-get-FCC-and-CE-certification-for-a-very-simple-electronic-gadget), and where competition from China and Hong Kong (and soon India), aren't bound by such pithy things as regulations, and who have their shipping costs subsidized by the government on top of that, then regulations are going to have to change. If the FCC wants products to be tested, then the government, not the small business owner, should bear the brunt of those costs. And regulations, which are clearly too strict, because even with all those billions of untested products coming in from China the world has not devolved into chaos, need to be relaxed.
--- End quote ---
Well,
1. You're a small player. You've never mentioned making billions of things. Yes! Scale matters! You're not special, you're a nobody. If you make a billion of something, your product is in the mind of the whole country, and they will collectively bend to make that popular thing, a sure thing.
2. Importers and retailers are liable for this sort of stuff, all the time. They usually have enough (capital or credit) to take a few blows. Sometimes, even if you do everything right, import good stuff, inspect it for all the marks, give it a go yourself; you still end up losing. You have to absorb that cost, and roll it into your bottom line. Retail is expensive, and this is part of the reason why. This is why consumer protection laws exist: to ensure that, okay yeah fine you'll have to pay extra for a variety of reasons, but you're much more likely to get a functioning, safe product. (EU is far more strict about this, of course.)
3. You mention direct China sellers. In these cases, the buyer is the importer. They probably don't realize it, but they are liable for their own safety. There's not much they can do about the seller, except complain with some nasty e-mails or a bad rating. Absolutely no teeth. Indeed, a very large part of why that crap is so cheap, is because it's direct import and buyer-beware -- caveat emptor.
Indeed again: if enough people purchase goods in this way, and enough people wind up injured as a result (defective, unsafe, interfering), the ultimate effect will be even worse: an overall chill on direct-China sales. In the extreme (perhaps it wouldn't go *this* far..), trade restrictions could be increased. In short: a few bad apples spoil the bunch. Tragedy of the commons!
4. Back on topic, the FCC will not raise limits, and should not. I am an electrical engineer, and I like my airwaves clean and functional. Just as I want my air clear and breathable, and my water pure and drinkable. The FCC's primary responsibility is to its licensees. There are a great many radio operators, licensed to use their segments of spectrum, who expect that spectrum to be available, and whom are given rights to take action (c/o FCC) against any sources of interference. ...Have you heard of the Tragedy of the Commons?
A possible example, combining both aspects: wireless power.
There's been articles about residential heating, using microwaves. In reality, it's not nearly as unsafe as it sounds, and it literally heats only the squishy meatbags that need it, so it's maximally efficient!
There's been articles about chargers, using various means (ISM band RF wireless; GHz wireless; even ultrasonic power transfer!).
All of these have the same liabilities -- filling the air with a huge amount of energy, that will cause a wide variety of problems (both for bodily safety, and functionality of standard devices). No matter how viral these schemes go, they cannot succeed, because you can't beat physics.
--- Quote ---Do you think Adafruit and Sparkfun got big by rigorously testing every device they sold? Of course not.
--- End quote ---
Most of what they sell, is an incomplete component, and sold as such. These are FCC exempt -- given that the final manufacturer / end user still meets Part 15 requirements.
Likewise, dev kits and application boards. I've seen some truly awful ones. The LTC3810 application board makes pulses <5ns, >50% overshoot, and there's nothing you can do about it. The pulse shows up all over the board. In other words, I had the same experience you did: crap everywhere.
The FAE* was clueless; the best he had to offer was: "eh, toss some ferrite beads on it?"
(*Field Applications Engineer; the guy who comes to your office when you phone up Linear Technologies and ask for some help with a particular product of theirs.)
That particular product ended up in a heavy metal box, and the customer wound up with something like 30% higher unit cost than they wanted. That reflects poorly on us. At least it wasn't so bad they couldn't market the complete system; that would've been a disaster for all parties involved.
--- Quote ---And why should they, when their parts are just going to be stuck in people's art projects or whatever with wires going everywhere and no attempt at shielding being made? Now, legally, they might be in the clear because those items are classified as electronic kits, but speaking from an ethics standpoint, is what they're doing really any different from releasing a finished product that you know is going to emit EMF? I would say not really. The end result is the same, whether someone has to solder the battery holder and neopixels onto their blinky shoe thing or not.
--- End quote ---
Just to be specific -- electronic kits have a specific clause in Part 15. Whereas components don't need to meet any ratings (because... how could they?), kits, assembled according to instructions, must meet regulations.
It's probably a worse deal for the manufacturer, because it's much harder to ensure the end user builds the product to their spec. In theory, they should be off the hook if the user doesn't follow spec, but who knows.
In any case, the offending user is the first target (by the FCC). If it's just some dumb art project (good example ;) ), they can simply turn it off, and probably take it apart and move on to the next cutesy thing to lash together. (Hopefully being a little more cautious next time?)
For much of those electronic components and Arduinos and kits things, that's about the end of it. They're not meant to be operated as a finished product, and operational performance is the user's responsibility (again, whether they realize it or not; it would be prudent of the seller to provide reminders of this..).
I don't know Adafruit's full product line, but I would suppose they sell things meant for office use (e.g., computer peripherals?), or stand-alone (soldering station..?), and those should meet regulations.
Tim
Starlord:
--- Quote from: dmills on July 05, 2016, 03:39:37 pm ---Ok, so re do the sim for a two ohm load, and fiddle to taste, no biggie.
What I meant was a 1uH inductor in series with each (Tied) output from the amp and a 100nF cap to ground from the speaker end of each inductor, very routine for class D (two inductors and caps per combined output channel).
Now you are going for 2 ohms, so maybe a little more cap would not hurt, possibly 220n instead of 100n, whatever, basically you have fixed L & R, so you need to set C to avoid excessive current draw or excessive peaking (tends to be the same thing), and take the corner being what it is.
--- End quote ---
I've been trying to simulate this in Partsim because I couldn't figure out how to get LTSpice to work, but I'm not getting results that seem to make sense. Perhaps my model is wrong.
The graph above is the probe of the wire between L2 and R1.
By the way, I found this application note from TI detailing the design of filters for class D amplifiers. It confirms what you said about excessive peaking tripping the overcurrent protection on pg 6:
http://www.ti.com/lit/an/sloa119b/sloa119b.pdf
If I'm not mistaken, Wo is the cutoff frequency in radians? How do I convert from Hz to that?
CJay:
'wah wah wah, the law isn't letting me do what I want'
I want lots of money, it's only the law that prevents me wandering into a bank with a shotgun and taking as much as they'll hand over, so that should be fine, I mean, it's not as there aren't successful bank robbers who went on to be very rich right?
Seriously, if you're betting everything on this one product then you *REALLY* need to be sure it's legal to sell and works properly or you're in for a world of pain, arguing and ranting against people who have years of experience in product design, compliance testing etc. doesn't remove the fact that your design is poor (being generous).
Oh, and BTW, http://lmgtfy.com/?q=radians+to+hz
Starlord:
--- Quote from: CJay on July 06, 2016, 10:04:37 am ---I want lots of money, it's only the law that prevents me wandering into a bank with a shotgun and taking as much as they'll hand over, so that should be fine, I mean, it's not as there aren't successful bank robbers who went on to be very rich right?
--- End quote ---
Yes, let's pretend the real world is black and white, and that there is no difference between designing an electronic product you expect to sell a hundred of a year that might emit enough EMF to interfere with someone's CB radio occasionally, and bank robbery. That's one hell of a strawman you've constructed there. Did you get the right permits for that?
Also, let's pretend that EVERYONE doesn't break the law. Have you ever exceeded the speed limit? You're putting people's lives in danger so you can get to work a little faster. I'd say morally, that's less justifiable than what I'm doing.
--- Quote ---Seriously, if you're betting everything on this one product then you *REALLY* need to be sure it's legal to sell and works properly or you're in for a world of pain, arguing and ranting against people who have years of experience in product design, compliance testing etc. doesn't remove the fact that your design is poor (being generous).
--- End quote ---
I didn't intend for this project to take over a year. It was supposed to take 3-4 months. And I've taken pre-orders from customers who bought the previous version to pay for the R&D on the new version. So you want to talk about bank robbery? How do you feel about taking people's money for a product and then delivering nothing? Because that's what you're suggesting I do.
--- Quote ---Oh, and BTW, http://lmgtfy.com/?q=radians+to+hz
--- End quote ---
Very funny. You think I didn't google it before I asked? That conversion is for frequency to RADIANS PER SECOND. The datasheet specifies Wo is frequency in RADIANS. Are these the same thing? Are meters per second the same thing as meters? No. That's why I asked people who know better. I assume the datasheet omitted the per second part, but I don't know that for certain. And some references I found talked about the conversion requiring a sample rate, and I had no idea what to use for that.
Also, I have to ask, but did you happen to miss the part where I am here asking questions specifically because I want to design a better product and get rid of these emissions? What I'm taking issue with is that the idea that I should design nothing because I don't know everything and don't have an unlimited budget to make it absolutely perfect. I feel like if I invented the microwave oven today, people like you would be giving me shit because it kills WiFi in the vicinity and in order to design something that is affordable for consumers, sacrifices need to be made.
And speaking of microwaves, you know what's ironic is that WiFi only exists because of the microwave. If the microwave didn't make those frequencies nonviable for commercial use, then the businesses that wanted to design wireless routers wouldn't have had an unregulated frequency spectrum to utilize. That band would have been auctioned off to the highest bidder. One could make the argument then that interference drove innovation. And speaking of innovation, I believe that frequency hopping routers were also invented to get around the issue of potential interference. One wonders what other advances in signal processing will be made because businesses were forced to work around a too common household appliance?
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version