Take the bus... to get some new cables: Raspberry Pi 4s are a bit picky about USB-Cs Mere weeks after launch, it appears that the hot (in more than one way) new Raspberry Pi 4 is a bit finnicky when it comes to power cables. The new Pi arrived with all manner of exciting tweaks and modifications including a couple of 4K HDMI outputs, more memory for those who wanted it, and a speedier CPU. To keep things …
Made a mistake for sure, but "an inconvenience for hobbyists, and certainly not ideal for those that bought the things in bulk"? I doubt that. I don't think there'll be an awful lot of people trying to run it on their macbook charger, or if they do, that can't stump up for a PSU, and if they can't, that will really really want to get on their hobby-horse about it.
And if they do, I guess they can just return the pi for a refund, can't they?
Doesn't matter. There is a standard, there is an implementation schematic, and somebody chose to ignore it. Which causes one immediately to wonder about other shortcuts.
The Pi is not a toy but increasing used for serious jobs. This sort of issue doesn't exactly inspire confidence. It's a bit like buying a secondhand car and noticing that one wheel is different from the others.
I was on the verge of buying one till this came up, now I'm waiting for a revision. Not because of the USB issue per se, but in case anything else comes out of the woodwork.
The Pi is not a toy but increasing used for serious jobs
This, for sure! I have 10 Pi units of various vintage in small-but-useful sever roles. I was naturally interested when the Pi 4 came out, but bearing in mind the problems of being an early adopter I decided to wait. Besides, the Pi seems to be needing an increasingly hefty PSU which is kinda problematic when you have more than a couple.
I'll be interested to see what happens next...
The 'official' PSU seems to work fine on mine. For my own amusement I am going to mount the thing on a baseboard along with keyboard and mouse and some mains sockets on a lead and a few attachments so I can put a 21.5 inch monitor on it or fold it all down for transport. I hope they fix the USB problem but its still the best portable I've ever had!
Do they sell it for such uses? I don't think so. They aim to be as cheap as possible.
It's like buying an Holga camera, or even a Fuji Instax and pretending it works as a Canon 1Dx... the former are built to have fun, the latter for serious jobs. At very different price points.
But then again, a £4.99-off-ebay webcam is probably just as useful as a go-pro, if you're going to use it, for example, as a component in a motion-detection system, and that's just an example off the top of my head. Just because you're maiking something 'professionally', rather than as a hobbyist, doesn't mean you'd use more expensive components as long as the cheap ones do the job. If you buy the cheap components, and they don't work as advertised, you'd expect to get your money back.
Would you like to be in front of an autonomous car built with the cheapest components around?
If you build something professionally you vet your supply chain and ensure they met your specifications fully. If you need to build something cheap, you plan for variations in components, and the inevitable bad lot.
I like to have fun with Pis, but would never run anything important on them.
But now I understand how many IoT devices are built...
"Would you like to be in front of an autonomous car built with the cheapest components around?"
Would you like to spend £1000 +on a digital signage rack mounted "server" (read a bog standard PC with a dual port card) or £25 on a Raspberry pi to do EXACTLY the same role, in a much smaller space, drawing much less power and getting into placed the rack version can't.
Like many commentators, you have a VERY blinkered view when it comes to the meaning of professional use.
Fyi, many of the largest digital signage companies actually support and recommend the Pi.
They may not *sell* them for serious purposes, but companies certainly *buy* them for serious purposes. It's been stated that about half of all Pis are purchased by companies. Then there was the time a few years ago when the A+ Pis all disappeared from the sales channels. Turned out that one or two big commercial display companies decided the A+ was just the thing they needed to make "smart" displays and bought up every A+ they could get their hands on.
"If you buy and use something designed for a different use, especially if you do just because it's the cheapest around, and you have issues, it's just your fault."
They're not buying it for a "different use [case]". The pi is meant to be a computer with the specified interfaces, and they're buying it to use the computing with those interfaces. It is failing to properly do one of the things it said it would, namely receive power from a USB C connection.
In addition, you're not at fault if a product fails to do something it said it would. The designers or manufacturers or sometimes plain bad luck are at fault. If you still subscribe to this policy, I'm pleased to inform you that I've just started a business. If you have a task you want to perform, send me a message detailing what you're doing and what tech you would be using for that purpose. Our business will happily sell you technology with the same feature set for twice as much. You don't want to buy our solution for twice as much? Well fine, but just remember that if the one you buy doesn't work for some reason including a design flaw, it's all your fault because you decided to buy a cheaper product.
Not a knee jerk at all here, just a realist. That whole cheapest component shtick works in anywhere but in The Real World, the one where that's what I'm used to dealing with at all levels of systems design and implementation. Go back to that autonomous car example and really look at where the components were designed and sourced from and then get back to me on how great they are in use. Now the stuff I design and build for my own use spares no expense when it comes to systems that are critical but no one else gets between the purse and my control-freakary.
BTW, I do have exactly one RasPi, a ZeroW used for exactly nothing. Powered on once to make sure it works and tossed in a drawer for when I find that roundtoit for a project here. Other shit keeps getting in the way. I like them, as an embedded engineer. I'm also waiting for v1.1 for all the kinks to be worked out. That RasPi 4 would become my new Internet connected device. Crack it or even melt it down, meh!
"Go back to that autonomous car example and really look at where the components were designed and sourced from and then get back to me on how great they are in use."
I'll get right on that. As soon as a RealWorld autonomous car actually exists.
Hint: Don't hold your breath.
> This sort of issue doesn't exactly inspire confidence.
Especially as it is not their first fuck up. The power over ethernet board had to be recalled from market and they had shorted power rails on some earlier boards which caused things to over heat. After the power over ethernet fiasco they promised better testing but problems have been identified with the latest model which should have been found in testing.
I doubt it will stop them patting themselves on the backs and declaring how bloody brilliant they are or having a quality metric based on units sold and money made. They have proved themselves less competent than they like to think they are and like people to believe.
Especially as it is not their first fuck up
I guess people will go and use companies with unblemished history of no fuck ups then. Good luck finding one.
Being a high profile outfit, like some others I could mention, their fuck ups get a lot of attention. Other lesser outfits fuck up but it slips by under the radar.
In this case, USB C isn't implemented perfectly. I would never have known about it if this was lemon pi or one of the other souped up clones that barely work.
In this case, there is no need to recall or stop sales like the PoE thing. Just tidy up next revision, which won't be far away.
I would never defend someone for not doing enough testing, but I find it hard to hate these guys for this issue. It constantly amazes me just how much the Pi provides in such a small package, at such a low price. Speaking strictly a hobbyist, personally I wouldn't have a problem paying for an official power supply on top of the Pi unit itself - still seems like good value for money.
That's just my personal opinion based on my personal circumstances though. Like other have noted, if I was someone buying in bulk / using these in a large scale non-hobbyist context then my opinion may well be different.
"But they decided not to follow the specs"
It was a sodding mistake.
How many hundreds of pages of specs did they have to troll through to implement this design?
It's great to see all the people on here getting stuck in over a mistake, I hope all your software / hardware works first compile and no bugs ever get past testing, because that's what you're complaining about.
It was a sodding mistake.
And that's why code review became a thing. And that also applies to hardware design. So either there was no hardware design review or more than one person thought it was okay. It was obviously never passed by anyone who could see that it was wrong and not compliant with the specification.
Mistakes will always happen. The real issue is they shouldn't make it into the wild.
How many hundreds of pages of specs did they have to troll through to implement this design?
One. One.
In the USB-C reference documentation is a reference design for a USB-C port implementation. A 1-page circuit diagram that if followed exactly will give you a fully functioning, to-spec USB-C port. In fact, it's not just 'a' reference design, it is the USB-C port design full stop.
Just to remind everyone... it's not a PSU issue, it's a cable issue (caused by the RPi design error).
The irony is that a cheap non-e-marked cable is going to work (regardless of PSU) while an expensive Apple e-marked cable won't work. This is really is not a big deal, just get a cheap non-e-marked cable and you're done.
The cable "shouldn't" matter if both ends implement the spec, but with the SNAFU on the implementation it leads to confusion at the complaint end.
Of course this brings on a separate debate of the evils of integrated circuitry in cables, have some very annoying micro usb leads which look identical to normal micro usb's but are in fact integrated serial ports and dont work as usb... Would much prefer dongles added inline to normal cables rather than invisible tech (same is true with active HDMI cables, if i had £5 for every time i have had to swap ends over because someone has connected out to out and in to in, did midi teach us nothing????)
What about SCART and composit video plus audio. It could handle multiple formats and was really good picture quality for the time.
At least the spec was readable even to me with my limited electronics experience. I still remember rewiring / soldering a SCART connector to connect a console to a 12ft projector with separate output for audio for their speaker system. Looked awesome in the pub hall and tennis was really good as you had a lot of space to run around.
USB C umm I'm crap at soldering.
If the inventor of USB-C lived in the middle ages, he would would have been put besides the IPv6 guy in a stock so passers by could toss rotten fruit at them.
Have to buy all these new cables, and for what... Oh yeah I know.. share holder Value.
Millions of people have to buy new cables, but hey lets order some Chinese solar panels, that is good for carbon dioxide and the environment.
This to the power of 100. Colleague had issues with his Mac in the office, and to rule in / out hardware, I suggested target disk mode on another Mac and booting up from the affected Mac (to rule out SW). Turns out you can't use the charging cable for data.
Yes, now fully aware that I need to buy a THUNDERBOLT cable, which looks identical to do this.
For once, this doesn't appear to be an Apple decision, but the USB-C specs
So, I has a little sympathy for the Pi Guys
Shame, but not that big a deal to be honest, just means buying a decent but not 5A rated usb C lead. In simple terms if it came with your fancy laptop to charge it dont use that cable, otherwise just use any decent (> 50p) usb c phone charging cable.
Something like this would be fine (as i have been recommending them to friends and family since USB C emerged and yet to have any cable related problems (apart from teenagers hatred of strain reliefs but no cable survives that...)) https://www.amazon.co.uk/LUORIZ-Braided-Charger-Charging-Devices-3-Pack-2m-White-Black/dp/B07425QG2T
Actually, the best USB A to USB C cables I've had in recent years are the ones that were on sale in Lidl a month or two back. Most adaptors fail where the cable meets the plugs. The hard plastic "strain relief" on most Amazon-sourced cables doesn't work for that purpose. The Lidl ones have a rubber strain relief with some give and hence are much kinder to the cable.
Depends on what you mean by 'good'.
Personally, I don't recommend any USB charger not capable of delivering the maximum permitted current concurrently to all ports - lesson learnt from having multiple iPads & iPhones.
So the Ikea Koppla 1-port fails as it is limited to 5Vdc 2A and the Koppla 3-port fails as whilst it can deliver 2.4A to a single port, the maximum load across all three ports is 3.4A. Mind you at least the Koppla has the ports are right angled to the mains plug - the early ones had them in line, resulting in the the total length of the adaptor sticking out of the wall socket coming in at around 4-inches.
So, the new cable introduced because we needed a new harmonised standard and the one that every single sane device was using already wasn't good enough just because you usually had to try twice to plug it in, isn't actually harmonised at all, and far from reducing cable clutter (which it never did, thanks to the above, even by design) is actually now confusing the waters still further. Have I missed anything?
Power may be the real reason for the new connector, but how many USB chargers are actually fully making use of this new capacity? Certainly Samsung are not yet shipping high power chargers with all their USB-C equipped devices, so to the end user it certainly seems like change for change's sake, and justified with hypocritical nonsense about saving polar bears and shizz.
Most of the micro USB connectors I have have some kind of tactile keying for orientation (either an indent or ridges), and those that don't (thankfully all white) I've sharpied. High tech solution huh? As for three attempts to plug in? Absolutely. I considered 2 to be a fair average...
"Most of the micro USB connectors I have have some kind of tactile keying for orientation..."
That's nice, but I have two e-Readers with the sockets mounted opposite ways up. People keep saying that I just need to make the logo face up (or forward), but my experience over 20 years is that this rule has only been true about 50% of the time. Reversible cables/sockets is the only sensible solution.
Once you have decided to change the spec, it makes sense to make the new one reversible.
Also worth noting that cheap micro-USB cables would oftne be made from a folded piece of metal, attached to the plug only on one side, making them extra flimsy and prone to bending/breaking. USB-C certainly seems to be more sturdy in the design. However, having the pins on the inside means that the sockets have a spindly little bit in the middle with the contacts, which not only makes them a magent for pocket-fluff, but makes them easy to damage if you try to remove said fluff. I think, on balance, I would have specified making this 0.5mm wider to make it more sturdy.
Of course it makes sense to do that, but here are a few other things it makes sense to do:
1. Have one, and only one, type of cable that can connect to the connector so you can't have, say, a thunderbolt cable which will connect to the port but won't actually work unless both devices are thunderbolt, or a cable that is designed only for visual data so can't carry power or connect two general purpose devices.
2. Have one, and only one, power style so you can't accidentally connect a standard USB 5V system to the newer high-power system and fry it.
3. Make the spec as rudimentary as possible so people who don't read it all the way through (random startups, the raspberry pi people this time unfortunately), are likely to have it work.
And yes, I would have to ask if USB C is really that big an advantage over normal USB. There were a lot of downsides about it, but the flipping cables over to try the other way was a minor annoyance and the connectors proved their longevity.
"wasn't good enough just because you usually had to try twice to plug it in"
Not just that. Having got it plugged in you found it had the wrong micro-connector on the other end. So did all the others within sight. When you found the right cable you then found it didn't work anyway. It was a faulty cable.
Palm used micro-USB cables which had a big silver disc on the side of the plug you should be looking at when looking at the phone. It was also recessed so you could insert it correctly by feel.
Apotheker was paid lots of money to be an idiot, I wish I knew how to do that.
Since these things are aimed at a hobbyist market; is it possible for the user to add the missing resistors, if provided the locations of the resistor end points???
A long time ago, when the Radio Shack TRS-80 was new, I recall people double stacking the RAM chips, with the exception of one pin which was (IIRC) the address line decoder for a 16K block of memory. Easy way to get 32K of RAM without having to spring for the Expansion Interface.
"Tiny SMT resistors need proper skills to mount even if there were unused pads for them.'
Maybe not as much as you think ...
(Note: I cut to the chase on that video. Some folks might have issues viewing the first couple minutes when at work. Some might find the rest NSFW, too. Don't say I didn't warn you. Poor bastards.)
User fixable? Why bother, just source a cheap $2 USB-C non-e-marked cable - way cheaper, and no risk of destroying your RPi4 (or whatever it is connected to).
No doubt a hyped up Dave Jones on EEVBlog will soon be extolling us with his explanation of how the RPi Foundation "came a gutser" by falling into such an obvious "trap for young players"... Hindsight is such a wonderful thing...
Although I did laugh when Dave complained about the "electrical isolation slot" in the PoE Hat (it was the slot for the camera cable). Muppet... if you're gonna shoot people down for their mistakes, try not to make your own mistakes and look a fool, Dave.
One of the things I never got around to buying were the 'octopus' style power strips, which instead of a bar with outlets all lined up, had a central box with short extension whips hanging off it, which was expressly to allow hooking up a bunch of 'wall-wart' power adapters. Haven't seen them in a while, so I'm wondering if they just sucked in terms of quality, or if they just weren't all that popular...
I have about 15 Raspberry Pi and hope the brand continues to make progress but its releases have long been made to fit poetic (e.g. anniversary) or other deadlines. In this case, the MagPi magazine (July) had already been prepared to announce the new model at the end of June, so it was going out, ready or not.
It's taking a long time for the supplier to learn the basics of engineering. Requests (for an earlier model) to know of what current the power supply should be capable were met with "Well, we find that..." instead of the conventional formal specification. Crashing when exposed to bright light (not what you'd want if it controls your dialysis machine) was described officially as an "adorable bug". Complaints that the Wi-Fi driver crashed the entire system were met with "you don't have to use it if you don't want".
I made the mistake of buying the PoE card as soon as it was announced: it failed to supply power when a keyboard dongle was plugged in. It seems that my decision to defer buying the new model has been justified quite soon.
I made the mistake of buying the PoE card as soon as it was announced: it failed to supply power when a keyboard dongle was plugged in. It seems that my decision to defer buying the new model has been justified quite soon.
A full refund was available to you, and once the updated PoE HAT became available you also had the option of a free replacement: Blog
I wonder if you would have been treated so fairly with a cheap Chinesium SBC or similar device? No, you'd have been left up sh*t creek without a paddle - _then_ you'd have cause to whine about "making a mistake", but not with how the RPF have dealt with their quality issues to-date.
I don't work for the Raspberry Pi Foundation (or am associated with it in any way) but fairs fair... some people just like to complain!
Reading the description: A MacBook charger doesn’t try to charge audio output devices - because it could cause damage. It doesn’t charge a Pi4 because it advertises itself as an audio output device. The Raspberry charger however charges it. Does that mean the Raspberry charger will try to charge and possibly destroy audio output devices?
So, this was, basically a simple mistake, a misunderstanding when trying to reduce component count AIUI - important when you sell a lot of kit, and especially on Pi's where the profit margins are slim anyway. So, it'll be fixed in a board revision in the next few months. It is indeed odd, as the boss said, that this didn't come up in testing. We had an internal test program, and lots of alpha/beta tester, and no-one reported it. We did use a number of different power supplies prior to building our own, they all worked!
In the meantime, not really a big deal. Use a cheaper cable, use an adapter, use the very reasonably priced official power supply, use someone else's reasonably priced power supply. Some of the reports 'out there' seem to be promoting alternative devices - and by some coincidence, the links they provide to these (not so good) products seem to be amazon associate or similar links, so they make money on the sales....how odd.
My own opinion - if you buy a dedicated power supply, you can use your Macbook charger for charging your Macbook without turning the Pi off....
- How large is the design 'team'.
- Who back-checked the PCB?
- Did they do a pre-production test run?
- How many PCBs was their first run?
Another case of British "we'll muddle through". Pi designs always had compromises from Version 1 (aka Mark 1).
We use a Chinese outfit that critiques proposed designs and suggests pre-production improvement. Only $100-300, depending on complexity.
Pathetic. Even Chinese knock-off artists test their products before production.
This post has been deleted by its author
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
