Maybe they should just improve their product to make it more resiliant, rather than blaming customers for thinking that 148 V is below 150 V? Not everybody buying these has a Ph.D. in physics and if it says 148 V on the label and 150 V on the other label then it's your product that has a problem, not the customer.
And no matter what happens, customer support should help the customer, not blame them.
This is 100% on the manufacturer if they intentionally chose to highlight the "best case" 150V, rather than the 120V lower end. Especially without any additional safety mechanisms.
The article presents it oddly, it's not that the converter maximum input gets lower, it's the solar panel output that gets higher from the nominal quoted value (which is not a maximum, and not really intended to be used as such). Derating your converter is equivalent for the purposes of ensuring margins, but it implies the issue is in the wrong place.
Your converter doesn't need to like the voltage, it just can't break from it.
In fact, if it's the type that shorts the panels for curtailing, it should be happy to just run them at the highest voltage it's comfortable with, loosing a few percent generation during exceptionally cold days.
Even that interpretation would be be wrong because it can and it does and the circumstances are very clearly described. Every inverter worth considering for a home installation has this capability. Only really old ones are not able to disconnect from the HV side when things are about to go pear shaped. If I had an inverter without that capability I'd get rid of it immediately because that's an accident waiting to happen.
The failsafe circuit to protect against this (over voltage condition) is maybe $3 and incredibly routine in anything with power. I would bet everything that it is in there.
What's most likely though is that the fuse is also internal, and externally the unit will appear bricked.
While in general most parts will work beyond their ratings to at least some degree, this isn't really a safety margin. A safety margin is known, rated, and validated, and usually a property of a system as opposed to just one part within it.
I believe the vendor here produces both the device and the panels being plugged into it, and while they also supply other vendors' panels, they seem concerned primarily with customers who buy all components from them and then experience this failure.
I agree the labeling is an issue here, but the solution must come from the wider industry or regulatory bodies; the alternative is for vendors to switch to their own pseudo-units to internalize the math, which would not be good for customers either - think "ACME Generator2000 accepts up to 4 Power Units of input; each ACME SuperEco Panel supplies 1 Power Unit, or 1.5 Power Units if you're in Canada...".
Saving users from having to do a little thinking to not brick their device is a tried-and-true excuse for vendor lock-in in our industry :).
Why not label the panel with the maximum possible voltage that it can produce? Or even have a little table showing the maximum possible voltage at various ambient temperatures. It's not as if there isn't room on the back of the panel for a big enough label.
There's hiding complexity, and then there's creating fake reality for people.
As it is, panels are gonna produce variable power depending on the weather. Putting interoperability with third-party panels aside, to get the simplicity of "max 2 panels in series", they'd have to either cap the max power on the panel/generator link and dump the excess, or set the limit based on the worst case a customer is likely to encounter. I.e. they're either gonna waste power, or gouge their customers for extra hardware. Neither of that makes sense for an ecological product sold to a price-conscious customer base :).
The problem is that you run higher voltages with the same hardware if your in Alaska than if you're in Florida.
Substantially so.
"Wasting" those 5~10% during severe winter conditions isn't worth splurging on the voltage converter.
Though then selling units that suggest to not run a few hundred volt strings before paralleling instead does sound bad, as the string doesn't need separate fuses rated to many volts DC.
Considering the electric code has an 80% rule for loads, anyone assuming they can use 100% of what is on a label probably should not be doing electrical work.
No, you just shouldn't assume that that's the case. In the case of a computer and mains, both of those are nominal voltages and there will be a range of voltages which are expected to function, and if you want to check for sure, you should check those ranges.
The point is, voltages are usually a 1:1 match. That includes when you're working with ranges, you want the supply range to be inside the load range.
And even for amps where you see that 80% rule, that's for keeping the load smaller than the supply. Solar panels aren't a load and don't work that way.
I agree with you, but. The NEC has an 80% rule for continuous loads (over 3 hours) that use a typical circuit breaker as overcurrent protection. If you use fuses or a 100% rated breaker as your over current protection, then you can use all of the available ampacity. Anyways, devices that use a 15A receptacle (for example) will not draw more than 12A continuous if they’re meant to run continuously (3 hours or longer).
I don’t know about the components they’re selling, but with electronic components, it’s on the buyer to properly read the data sheet and understand what the quoted nominal specs mean.
Unless it’s safety critical, you usually don’t want a system with a bunch of active electronics to prevent someone wiring it up wrong, because those components will interfere with whatever you’re hooking it up to, such as the MPPT, the battery, or whatever else.
This is like how AA batteries have a nominal voltage of 1.5V but the actual open circuit voltage is 0.9V~1.65V depending on charge level, temperature, etc. If you connect an AA to something that’ll explode at a voltage of 1.55V, that’s on you.
Similarly if you buy a 470 ohm resistor, you will find in on the data sheet that’s usually at 20°C. To know what it’ll be at any other temperature, you’ll need to use the temperature coefficient to calculate it.
In your AA batteries analogy, this is like saying that you do not need to state that the device exploding at 1.55V would do so, not about battery declarations (panels in this case).
This isn’t electronic components, this is sold as a consumer level gadget that anyone can use. No one expects a standard consumer to understand data sheets like that.
You started off by saying you don’t know about the components they’re selling - but that turns out to be absolutely critical to understanding the context here.
Either way, I can’t believe they just let it fry the main board instead of having a sacrificial fuse or equivalent go first in these scenarios, whether it was a product aimed at professionals or not. It’s just dumb.
This feels a bit like the logic behind "do not put hamster in microwave" warning labels.
I think if you work with electrical or electronic systems in practice, you learn pretty quickly to respect tolerances and that data sheets are a map, not the territory.
Also, electrical installations are usually seen as a field that should be done by trained personnel, not arbitrary laymen home owners. So I think the appropriate reaction would be to remind people that they should hire an electrician to do the installation, if they don't have the necessary specialized knowledge themselves.
So on the one hand we have a product which isn't even remotely designed for the use case (hamsters), and during normal use shows obvious behaviour (cooking) that should imply risk to said hamsters. On the other side, we have a product designed to be installed in an electrical system, and shows no signs during normal use that it's installed unsafely, and where the advertised specs are not actually safe for normal usage.
Whether or not the company in this case shares some or most of the blame with novice users - the analogy is not a great one.
I have the impression these are consumer products so I would them to be designed to be installed by people who do not normally work with electrical systems. If they are only sold to tradesmen that would be different.
Yeah, it would be important to understand who the company is intending to sell to or do the installation.
Though the lines are often blurred, because I guess most companies would like to sell directly to end customers, even if their product requires a professional to install.
Even IKEA does this. You can go in and buy an electric stove and oven, grab them from the warehouse and take them home with you right there. But it's a bit of an illusion: You're still supposed to call an electrician to actually connect the things.
> Yeah, it would be important to understand who the company is intending to sell to or do the installation
You can do that by just reading the product page[1]. The delta pro, the equipment in question, looks like a plug-in appliance. It visually communicates that it is portable (by having a wheel and a handle) and by virtue of having a power lead connect to it it communicates that you can just plug it in. They further reinforce this by writing this: "Plug & Play home backup solution". "Easy installation with completely pre-wired Plug & play home backup solution" "The solution provides a convenient home battery system without rewiring or running dangerous extension cables through your home." "Plug directly into an AC wall outlet and make sure that the wall output current is more than 15A."
And on top of that the manual[2] makes no mention of needing an electrician.
In contrast an IKEA electric oven's product page[3] states this: "No plug is included. Installation to be carried out by a qualified installer." and then the manual[4] states "Installation, including water supply (if any) and electrical connections, and repairs must be carried out by a qualified technician."
But of course nobody reads the manual. The big difference is that one comes with a plug while the other doesn't.
Actually if "you work with electrical or electronic systems in practice" you get pissed off at everything for how dumb it all is: 12V DC batteries are more like 14 nominal? AC wall plugs dip voltage when printers turn on, random equipment you arent even sure in the building can trigger UPSes based on unknown settings in the device? International standards and communication protocols mean nothing as "a standard" because each company has their own entire list of bugs/implementation mistakes. All the international enforcement certifications care way too much about inconsequential bullshit and miss all the true showstopping problems in most industries?
This world is amazing anything runs at all. The slightest addition of complexity is causing everything to fail now.
Modern computers are less reliable than ever, some companies have decided to REMOVE the pinhole bios reset (that has been around for 30 years) at the same time as things are buggier now and dont boot again until you physically unplug the bios battery deep inside and hard to get to.
This is what makes software engineering so seductive, everything works exactly as it was designed to (whether intended or not). Imagine trying to program for computers with memory that drifts values progressively more as it wears down.
Maybe it would be a good idea to at least add a pair of MOSFETs (one for each rail, + and -) and a voltage meter? Like, the voltage doesn't rise to maximum instantaneously, there should be ample of time to detect voltage rising to a critical amount.
So say, the input is rated for 150V, spec the components to sustain 180V, and trigger the MOSFETs to disconnect the panels at > 160V.
And maybe also add a big ass buffer capacitor, that can be used to soak up a bit more energy in the case of an inrush spike before the MOSFETs actually disconnect.
MosFET's or IGBTs are likely what failed. And, capacitance is something you do not want on a string of PV.
DC starts getting really nasty to deal with somewhere between 36-52v, with 150v of panels not being something joe-blow should be able to buy on amazon. Designing these systems to be safe is difficult.
> Not everybody buying these has a Ph.D. in physics and if it says 148 V on the label and 150 V on the other label then it's your product that has a problem, not the customer.
Idk. I don't have a PHD, but 220V sounds like 240V to me. I wouldn't do this.
I feel like getting advice about how to wire up electronics should not be so hard.
> Maybe they should just improve their product to make it more resiliant
Adding "resilience" usually adds to the per-unit cost. I think making a web page adds some cost too, but at least that can be amortised.
> And no matter what happens, customer support should help the customer, not blame them.
I think that's happening here: Making a web page to educate future customers seems like a really good idea. I wouldn't have thought that necessary until I saw it, but I'm always excited to learn something new.
The existing customers who did dumb should consider this a relatively cheap education in electronics; cheaper than a PHD at least!
Also, whether the company also gave them rebates or credits we don't know here, but telling "customer support" they "should help the customer" is also telling them they're not helping the customer, and you don't know that.
I have all sorts of electronics that say everything from 208V - 240V that all go in the wall, so I think all those numbers are probably close to each other in whatever a volt is.
I think if I'm worried about a limit of some kind, being within 5% of that limit seems like I might as well be over-limit if the commonly seen distribution I see in my house is greater than 15%
I have a lot of electronics that accept 100-240v doesn’t mean I think they’re close together, just that they have compensating mechanisms to handle such voltages.
The margins are generally lower for higher power devices because the electronics are more expensive. Thankfully these electronics in general are becoming cheaper which is one reason why they’re ending up in the hands of people inexperienced with them.
Also try to tell someone they need an extra 30% in margin and often they'll think they’re being upsold.
> I have a lot of electronics that accept 100-240v doesn’t mean I think they’re close together, just that they have compensating mechanisms to handle such voltages.
I bet you also have a lot of electronics that don't though, and those that do probably say so.
My kitchen mixers, dishwasher, washing machine, driers, rice cooker, refrigerators, sauna, and pool pump aren't that tolerant by a long shot. I've got a few computers with a switch on the back to choose between 220v and 110v.
Plugging something that takes 110v into my house breaks the thing, so I've learned to check.
But I don't have anything that's 208v that can't go into the house. So I think whatever the situation is with volts, within 15% is "basically the same", so coming within 15% of the rated limit, is probably just like exceeding the limit by 15%.
And so this is why I would not expect something at 146V to be under the safety limit of 150V.
> Also try to tell someone they need an extra 30% in margin and often they'll think they’re being upsold.
You’re listing things that are high powered and or old. Modern PSUs don’t have that switch. While it’s in the early stages I expect more white goods to switch to BLDC motors which will likely use voltage transformers that’ll support the 110-240v.
People generally are not lugging white goods internationally, the average persons experience with different voltages is for laptop and phone chargers when they travel.
But for the matter at hand, the margin mentioned is needed on the solar systems, this is where the inverters can get expensive, which is why it can look like an unnecessary upsell to people who’ve never blown a device before.
At least in parent's case, probably because the vast majority of their comments are black-and-white hot takes that are always downvoted regardless. And they basically never respond when asked for any sources.
They're just being cheap. If you're going to let customers plug panels directly into your box you should have overvoltage protection. It's that simple.
Besides the ratio between the peak and effective value, you must also account for the standard tolerance of the nominal value.
The actual maximum peak voltage for the European mains, is 230 V * 1.15 * sqrt(2), because of a 15% tolerance. That is about 375 V. With a small safety margin, the minimum voltage rating for components connected to 230 V a.k.a. 220 V is of 400 volt.
Makes me wonder if that was how EV DCFC quickly settled with 400VDC. The voltage makes sense if it was somehow known that engineers has intuitions with safe designs for 400Vp-p systems.
So the issue is that 220V is nominal in China, 230V nominal in UE and 240V is UK/part of Australia. So if anyone is preparing product for global market (as most are doing now) more likely then not will support all of this voltages. Thus is kind of normal (but wrong) to assume 220V sounds like 240V.
When the voltage was unified in UE, the nominal value was set to the median of 230 V, but its tolerance was raised from 10% to 15%, so that the new maximum peak value of 230V + 15% will match the old value of 240 V + 10%.
So now for all 220/230/240 V standards you have the same maximum voltage value that is used for electrical designs (about 265 V effective), so they are equivalent, regardless of the name.
Many charguers are now 100-240V, 50-60Hz, that is close to pluggable anywhere on Earth. (I burned one or two a long time ago, when I forgot to check and used a 120V transformed here with 220V)
Same thing happened to PC PSUs. I don't think there is a recent unit that still has the self-destruct voltage selector switch which pops them if you are in 230V land (and the switch is set to the smaller setting).
> Making a web page to educate future customers seems like a really good idea
I don't think this is an official website of ecoflow.
Other than that I agree. I don't think asking for a bit of knowledge from the customers is a bad thing. A warning in the manual about safety factors should be enough.
Misleading title. I was expecting to read about why devs actually embrace type hints. What the article is about is why you should and how you can use type hints. That's valuable, but different from what the title suggests.
Besides, the lack of static typic is what makes a lot of the appeal for beginners. It's much harder to convince a non-CS beginner why they should bother with the extra burden of type hits. They are optional anyway and just slow folks down (so they might think). Careful with generally demanding that everybody use them.
But they probably help coding assistants to make fewer mistakes, so maybe that will soon be an argument if it isn't already. (That's an angle I expected in the article.)
Already the title of your submission does not check out. Do you know how many clock cycles a 1 GHz CPU realizes in one nanosecond? One. Just reading the input argument of a function takes a "nanosecond-scale" amount of time.
> I'm a self-taught developer and researcher who left school at 16, and I've spent some time exploring a first-principles approach to system design for various frontier problems.
As much as I appreciate new ways of thinking, whenever I read "first-principles approach", my alarm bells go off. More often than not it just means "I chose to ignore (or am too impatient to learn about) all insights that generations of research in this field have made". The "left school at 16" and "self-taught" parts also indicate that. This may explain the hyperbole of the title as well, as it does not pass the smell test.
If you are looking for advice, here is mine: try to not ignore those that came before you. Giants' shoulders are very wide, very high up and pretty solid. There is no shame in standing on them, but it takes effort to climb up.
Kids hardly know what a multivariate equation is. Unless you use "kid" to denote 20-year old college students enrolled in a math program which some people do.
The other claim is doubtful too:
> while it require an experienced expert hours of work in Lean 4.
No, it doesn't. If you have an actual expert, it only takes a few minutes.
And besides, isn't this exactly what an artificial intelligence would solve? Take some complex system and derive something from it. That's exactly what intelligence is about. But LLMs can't deal with the complex but very logical (by definition) and unambiguous system like Lean so we need to dumb it down.
Turns out, LLMs are not actually intelligent! We should stop calling them that. Unfortunately, there are too many folks in our industry following this hyped-up terminology. It's delusional.
Note that I'm not saying LLMs are useless. They are very useful for many applications. But they are not intelligent.
I studied 2x2 linear equation system in high school at 14 (13?) y.o. It was a technical school with more math and physics, and later specialization (like chemistry, electronics, building, ...). I think in a normal school they study that at 16 y.o.
We also teach 2x2 systems to 18 y.o. in the fists year of the university for architects, medics and other degree that don't need a huge amount on math. (Other degrees like engineering or physics get 4x4 or bigger systems that definitively need the Gauss method.)
Unfortunately, whenever you try to exclude LLMs from the holy church of intelligence based on what they can't do, you end up excluding a whole lot of humans too.
Even low-IQ humans can in principle learn how to use Lean to represent a multivariate system. It might take a while, but in principle their brain is capable of that feat. In contrast, no matter how long I sit down with ChatGPT or Gemini or whatnot, it won't be able to. Because they are not intelligent.
It's a great achievement of the AI hype that the burden of proof has been reversed. Here I am, having to defend my claim that they are not intelligent. But the burden of proof should be on those claiming intelligence. The claim that earth is a sphere was extraordinary and needed convincing evidence. The claim that species have evolved through evolution was. But the claim that LLMs are intelligent is so self-evident that rejecting the idea needs evidence? That's upside-down!
> Here I am, having to defend my claim that they are not intelligent.
It's because all you've done is made a claim without any evidence. Someone pointed out a challenge that most claims about them not being intelligent can't submit any evidence that can't also be met by an LLM.
But instead of submitting any evidence to support your claim, you descended into hyperbole about how hard done by you are being expected to support your claims.
In science, it's okay to say we don't know. The amount of disagreement - even amongst smart people - about if LLMs are intelligent or not, suggest to me that we just don't have universally accepted research and definitions that are tight enough to decisively say.
But you're talking not only like you _know_ the answer for sure, so much that you don't need to support it with evidence or credentials, because those who disagree are obviously just poor victims of the AI hype machine.
Please make sure you pass your knowledge of your LLM discoveries onto the scientific community, you could change the world!
Did you read my post? The claim is "LLMs are intelligent". And instead if requiring evidence for that, apparently most folks (including you) are fine with just accepting that claim and require evidence if somebody questions this. That's what I'm doing.
It's like a religion. "God exists" is the claim. Nobody needs to provide evidence that this is not the case. LLMs are intelligent" is the claim. Nobody needs to provide evidence against that. In either case, the burden of proof is with the one making the claim.
> In science, it's okay to say we don't know
But that's not what's happening. LLMs are called "AI". You know what the I stands for, right? It's not "artificial we-don't-know-if-intelligent".
No you're not questioning it, you're making statements against it, without evidence, which is just as useless as the statement without evidence.
Like I suggested, like the person you responded to suggested: when science tries to prove or disprove LLM intelligence it generally descends into disagreements about definition or evidence (neither of which you provided).
The reason why no evidence was provided for the original LLMs are intelligent claim is because - if you read through this thread - you were the first to make that claim, and the counter claim.
> But that's not what's happening. LLMs are called "AI". You know what the I stands for, right? It's not "artificial we-don't-know-if-intelligent".
I don't know what you want me to say here - if you want to continue acting like there's some widely accepted and agreed on definition of intelligence that everyone who isn't you is an idiot for not knowing, then carry on.
I don't have a reliable definition for intelligence. Like I said, if you do, please share your finding with science, you could settle some fairly big debates and change the world in a meaningful way.
1. LLMs are widely called "intelligent". Evidence for my claim: The term "artificial intelligence" that is used everywhere. It has its own TLD.
2. There is no evidence that this terminology is applicable. Questioning it faces some variant of "well do you have evidence to the contrary?". Evidence for my claim: This thread.
You are welcome to disprove my claims, as in the scientific spirit that you say you uphold.
the way it goes is that a device/program of some sort displays a broad number of behaviours that previously was believed to be a future sign of artificial intelligence (e.g. somewhat coherent text, Turing test, giving the impression of following instructions and arriving at the correct results, etc).
some people claim this is artificial intelligence of a lower quality than humans, and these people expect that such mechanisms will eventually match and then potentially surpass humans.
then there's another crowd coming along and claiming no, this isn't intelligence at all, for example it can't tie its shoelaces.
my point was that every time you try to say that no this can't be what intelligence means, it needs to do X, I can find a human who can't do X, no matter how many years you might try to coach them. (for example, I will never be a musician/composer. I simply lack the gene.)
The retort is always "oh but in principle a human could do this". well, maybe next year's LLM will do it in practice, not just in principle, for all I know.
As they say, person who says it can't be done should not stop person doing it.
Heavier than air flight was once thought to be impossible. As long as you don't have a solid mathematical theorem that says only carbon replicators born from sexual intercourse can be intelligent, I expect some day silicon devices will do everything carbon creatures can do and more.
> my point was that every time you try to say that no this can't be what intelligence means, it needs to do X, I can find a human who can't do X,
Indeed, the point you are making is reasonable. But I'm trying to say that the premise is wrong. Nobody should be expected to come up with a reason why it is not intelligence. We should expect to be presented with evidence that it is intelligence. Absent that, the null hypothesis is that it isn't, just like any other computer program before isn't, uncontroversially.
I'm sure you already got my point, apologies for repeating it, but some clarification to clearly carve out our points may not hurt.
>Nobody should be expected to come up with a reason why it is not intelligence.
I'm not asking anybody to come up with a reason why it's not intelligence. I'm telling people they're wrong when they do try to justify calling it not intelligent. if you want to gatekeep a word, you should at least try to define it and then stick to the definition.
It is intelligence in the sense that if somebody had described it in 2010, we would have said yes, that's intelligence and it's hundreds of years away. It isn't intelligence in the sense that it's now here and we've found holes in the story.
Intelligence is so poorly defined that it's an ever receding finish line that somehow we're supposed to cross before we can call the device intelligent.
As Dennett said, it's like magic. Magic that is possible is just tricks. Real magic is that which is impossible.
> Objective fact and "more likely" do not match well.
Huh? If I take a die and paint a 6 on the sides which previously had 4 and 5 then it is an objective fact that you will be more likely to roll a 6 than a 1 with that die.
Indeed. But they typically are contingent on a certain number of views. If adblockers cause that stat to go down, then you get the opposite of what you are aiming to achieve: the user will see the "message from our sponsor" but their view does bot contribute to providing that sponsor with the data that the youtuber held up their end of the deal. Ends up bring an unpaid ad.
Views might be important to get the attention of a potential advertising partner, but once the relationship has started then keeping it going will likely be dependent on much more relevant metrics for the advertiser. And those metrics will usually be tracked on their end, rather than via YouTube. I'm referring to metrics like click-through rate, propensity to order, revenue on advertising spend, etc. Personalized referral URLs and discount codes are what allow the advertisers to connect their tracking and reporting to the originating YouTuber.
And you my friend are demonstrating why this keeps being used. It's so common that now generations of devs and designers are so used to it that they don't see anything wrong. And if on the phone with grandma, instructing her to go to "my files" and her asking where to find my files (instead of hers), that's shrugged off as stupid user rather than an UX fail.
If you're talking to someone who is mostly computer illiterate, you'd say something like "do you see a folder icon on the screen that says My Cases? Double click on that." and not "go to My Cases"
Yeah, if somebody is really that computer-illiterate, you'll also need to tell them where on the screen to look since they're likely overwhelmed by all the other things. These tend to be the same people who, unfortunately, haven't installed ad blockers, and are constantly tempted to click on an ad, thinking it's the "right" place to click.
Ironically, that response runs into the standard problem that many "limit" arguments have.
Generally speaking just because something looks like it's converging from some angle, it doesn't mean that it actually has a well-defined limit, and if it does then it also does not mean that the limit shares the properties of the items in the sequence of which it is the limit.
Examples: 1/n is strictly positive for all n. Its limit for n going to infinity, while well-defined, is not strictly positive. Another example: You can define pi as the limit of a sequence of rational numbers. But it's not rational itself.
So, no, your argument does not prove that pi is a number.
(I'm not arguing that pi is not a number. It definitely is. It's just that the argument is a different one.)
No time. Busy writing blog posts blaming customers.
reply