I have returned a product that did not do C2C charging and sent the company a support request about it. It required quoting the USB spec sheet and requesting they sent it to the appropriate product development department and eventually since there is nothing they can do then I congratulated the product manager on getting their bonus for reducing BOM costs by less than a few cents a unit at the expense of (I) a returned unit which must be sold at a loss, (II) a detailed review of why their product is indicative of a company which does not care about its customer's needs if it makes them a tiny bit more money, and (III) a tarnished brand in my eyes. I hope it was worth it. It was an Altec Lansing bluetooth speaker by the way.
I doubt it was a matter of reducing the BOM cost but rather a junior engineer updating an old product. Good move to push them, however. Far too many usb-c devices don't work properly.
Even better the higher volume you go. On DigiKey right now, a reel of 100,000 5K1 resistors is literally $105.00. At two per USB-C port, that's a unit cost of 0.21 cents per device. Literally a fifth of a penny! Plus maybe a few extra seconds spent by the industrial sized pick-n-place machine on each device
Indeed, at high volume the cost of stuffing/testing is comparable to the cost of the part ~1mil=$/1000. Actually, cost can increase more significantly if the number of different components increases too much (and inventory is more annoying) so it may be cheaper to stuff 2 or more of the same component rather than have more different values. Of course nobody really cares about this, if you're making less than ~100k/mo since setup costs dominate. A single line can do ~250k/mo running 24 hours. If the line runs less than 8h/day, they'll turn it off and/or switch to a different product on that line each month, which can cost multiple $k.
I wonder if the existence of USB-C has significantly increased the number of 5K1 resistors being used. Even on my own projects where I'm manually soldering maybe a couple boards I tend to just pick 5K1s out of part ordering convenience when I just need like an I²C pullup or something, since I'll be using them anyways with the USB charge/debug port
I'd rather buy something with a micro USB at this point, than something that doesn't do proper USB PD. At least you definitely know that you need a different cable.
I mounted it in a little clear case back to back with the microUSB tester board. The battery should effectively last for a decade. There are multiple other similar boards on tindie for comparable prices. I picked based on being able to consolidate shipping price for multiple units; make sure to check where it ships from and how much if you only plan on getting one board.
"This seller is taking a break until Feb. 9, 2023" - any good alternatives in the mean-time? (Tindie has a couple as you mention, just looking for specific recs.)
Searching for "usb c cable tester" on every major search engine pretty much solely returns port testers despite "cable tester" being in quotes. The product category must be pretty stark (and my god do search engines suck now).
I'm also pondering a way to mark my cables with the results of this thing. Color bands (like resitors) with nail polish or something?
As opposed to the one GP linked which just shows which lanes are connected (i.e. if it's only USB2 or not, it has an e-marker or not), this one is fully featured: it reads the actual contents of the E-marker, revealing if the cable supports 100W, Thunderbolt, etc. It even sometimes shows the chipset brand name in the cable!
It can also check Qualcomm QC and USB-PD chargers to see what voltages/amperages they support.
I don't read Chinese, IIRC I found it through another listing with more English and searched for the model number this listing was just cheaper
(I can read Japanese Kanji though, which along with the screenshots which are pretty clear on the feature set gave me the confidence this is the right thing)
Ah, Aliexpress has a tendency to randomly select a language, there should be a language dropdown in the top right (or top left if it has decided on arabic)
Which USB meter is this? I think you might be mistaking a 'cable tester' for 'testing a USB port of some kind through a cable'. The device I linked to just requires two ends of the same cable be plugged into it. You cannot do that with the 'USB power meters' I know of that plug into USB ports.
The latest models of USB power meters have multiple ports and can also do some checking of cables (and chargers) including reading the actual e-marker chip for capabilities and testing the negotiation capabilities of attached USB-PD chargers.
It seems like it should be possible to have a variable rate phone charging cable with just a simple switch and a few resistors.
Instead of hastening the demise of a phone battery by fast charging all the time, select fast, regular or slow charge, depending on your circumstances, with the selector switch.
Just looked around, but couldn't find anything already existing though. Seems like something that would already exist on Aliexpress.
Most (all?) modern phones manage charging speed based on battery health, charge state, and temperature. I'm not sure you could force one to charge faster at e.g. 2% charge in 120 degree heat with any kind of cable. Mabye? But that seems like a software thing.
> Limiting max charge, on the other hand, has known strong effects.
Yes, and you used to be able to do that on a rooted android phone. But, Google required the removal of debugfs in recent android versions, so need to run a custom kernel too now to get back that functionality.
Maximum charge, and percent charge where charging begins should be user accessible settings on any device with a lithium ion battery. Charging rate should also be a user accessible setting. The cable idea is just a hacky work-around for crappy software, that addresses one of the reasons lithium ion batteries prematurely fail, and doesn't require modifying the device to be charged.
Well if you do it every time at like 2-3C, you’re talking maybe 500 cycles compared to 1000 cycles at a slower charge rate. Depends on other stuff, too. Depth of discharge and max charge voltage are both even larger effects, but it still matters. (Knowing the exact chemistry of your phone’s battery can make a difference here, of course.) https://www.mpoweruk.com/life.htm
Fast charging from, say, 10% to 50% probably is a fairly small effect compared to trying to fast charge to 100%.
Even that much is aching for more numbers. How about .5C versus 1.5C? What if my phone is one of those fancy ones rated for 6C or even 10C? Unless someone gives some precise scenarios it's so hard to guess how my phone might be affected.
I don't recall if this is "standard" or not, but some devices have mode where it slowly ramps up the current until the voltage starts to drop.
IIRC, this is a fallback mode on some TI charge controllers. If it can't detect a capacity signal on the D+/- lines, it reverts to this mode of testing the charger. This was intended to support sketchy Chinese trash chargers that leave the data lines disconnected, and whose real power capacity is a mystery. This strategy lets you pull as much power as the charger will suffer
> This strategy lets you pull as much power as the charger will suffer
Some chargers will allow you to pull enough power to melt the thing or catch it on fire if they don't have sufficient protections, but it's likely those would have data lines (+ protections) anyway
Well, that's kind of the point of the check. Except in the most extraordinary conditions, the voltage will drop as you approach the supply's current limit.
It takes significant effort to design a circuit that can sustain large amounts of current while keeping voltage constant. There's basically no overlap between that sort of design and cheap Chinese fire hazards.
I've been all-in on usb-c for a long time, and have made a hobby of updating old devices to usb-c instead of microusb and in some cases adding li-ion cells. My favorite things so far is updating all of my retro consoles to usb-c power input and adding usb-c charging and a li-ion cell to my favorite keyboard, the 4th gen Apple wireless keyboard[1]. It has a bit more travel than the non-butterfly macbook keyboards. At one point I bought all US layout versions I could find in my country. I have about 10, some new in the box, which should be a lifetime supply.
Unfortunately, like another poster said, many usb-c devices are not compliant and only charge with an "a-c" cable. This is because usb-a defaults to 5V power, but usb-c defaults to no power output without negotiation or CC resistors. If you are in a pinch, a usb c-a adapter plus a usb a-c cable does work. In my experience these use the cheapest possible usb-c connectors that don't expose the CC pins, so to fix them, you need to remove the connector and install a new one. See [3] for power only and [4] for power+data.
Anyway, I always like to share 3 cheap products for anyone interested in upgrading devices to usb-c charging or connectivity. Since these are available from many aliexpress sellers but the actual listings come and go I will share search terms which hopefully work 1+ years from now.
- usb-c charging PCB for a lithium ion cell[2]
- usb-c connector with integrated 5.1kohm resistors, to easily use usb-c for 5V power without data.[3]
- usb-c connector with just enough pins for adding CC resistors and connecting to power and usb 2 data lines in an existing product, but without all the additional pins for usb3 and thunderbolt (because this makes it very difficult for a hobbyist to solder it in place without bridging pins). Unfortunately I don't know of a product with CC resistors + power pins + data pins so you do have to add CC resistors yourself.[4]
- small PCB with a chip to negotiate more than 5V (9V, 12V, 15V, 20V). Some of these offer fixed voltage and some are configurable.[5]
small PCB with a chip to negotiate more than 5V (9V, 12V, 15V, 20V)
The 20V versions of these make excelent replacements for broken rectangular thinkpad power connectors, as they fit within the footprint. Add a magnetic Usb-c adapter and it's much better than the factory options. Sugru works well for mounting it.
Something I'd love to find is a multi-port (say 10+) PD power supply. My use case is synths and pedals that randomly take 5/9/12 and a smattering of other voltages, replacing a bunch of DC adapters with USB-C cables and a decoy module would be ideal (or one of the options above where the device already has USB Micro ports).
One thing to watch out for (check the reviews) is most multi-port chargers have a total power maximum limit which is less than the sum of each port's maximum. So they renegotiate all ports when something is first connected. In other words, if you plug in a second synth, the first synth will probably be power cycled.
Another tip: if something takes a voltage not available in PD, for example 7.5V, you can often open the device and near the power connector find a voltage regulator. In the regulator's datasheet you will often see that it will accept a higher voltage (like 9V in this example). Be aware that a linear regulator will dump the extra power as heat.
Good points - it's likely that I won't be opening these up anytime soon. Most of the synths in question take 12V @ <800mA @ 12V (so ~10W) and usually come with a 12V 1A supply. Or they take 9V (current ranges). And there are a few that take USB power already. I'd imagine everything being either on or off and not having to hit the first problem too often.
The second problem is something I'd definitely avoid. I think I only have one or two odd values (5.7V etc.)
Definitely something that's more of a pipedream - I hope someday synth manufacturers will just settle on USB-C as the common power standard instead of having something different for each device.
That is a great link list! I've been curious about converting over a few micro-USB devices I have around so I might pick up some of these boards.
Do you have any recommendations for search terms for boards that would take in DC power (12V+) and make that available for output over a USB-C connection?
