Shared publicly  - 
 
My eighth #USB #TypeC analysis is a charger on loan from Benson, and highly noncompliant: the AUKEY 60W with QC 3.0. [Model PA-Y6]
tl;dr: Breaks many rules. Vbus HOT. Bridged CCs. Burns energy at idle. I got this for free -- and I want my money back!!

https://www.amazon.com/dp/B01EXWXVWO/

[PDF of Compliance Checklist for AUKEY PA-Y6]
https://drive.google.com/open?id=0B2OJRSgNnm4GZTBFVmdTdGc3TlU

This is my second review, first failure, in the new format: less data, more to the point. This charger is 110% bad. There's no way I'd recommend this (And I'm not just talking about the QC-on-USB-C part.)

I actually had to create a stamp to show the individual errors they did. (And highlight exactly what +AUKEY Official needs to do so they can potentially fix it.) In order of severity:

* First and most deadly, this charger is Vbus HOT. That means it cranks out amps with no regard for what's down the line. See the photo. I'm drawing 6 amps of current without ANY notice that a device is connected. If the receptacle gets shorted out, or you plug an A-to-C cable into it, kiss your motherboard goodbye.
[4.8.1.2 USB-based Chargers with USB Type-C Captive Cables]

* Second, AUKEY bridged the CC pins. With Vbus HOT the CC pins were already not useful, only used to signal 3A current with a pullup. But AUKEY also bridged them internally. This means ANY active or eMarked cable (read: good ones, USB3.1Gen2, Thunderbolt 3) will cause major problems!
[4.5.1.2.1 Detecting a Valid Source-to-Sink Connection]

For example, Benson's Pixel Chromebook interprets the cascade of electrical failures that occurs with eMarked cable+bridged CC pins as a "device being plugged in". So the Pixel starts trying to deliver current out. Combine this with a Vbus HOT charger fighting it backwards with 6A over Type-C, and you might just fry your laptop!

See my post here explaining why Vbus HOT is so darn dangerous (using Thevenin Equivalent Circuits):
https://plus.google.com/102612254593917101378/posts/GEvCNc5hARR

* Third, it has QC on USB-C connectors. This is bad because the only way you should change voltage on Type-C is over USB-PD. So it was an automatic fail -- but the rest was icing on top.
[4.8.2 Non-USB Charging Methods]

* Finally, even with a 60W charger with QC circuitry, AUKEY somehow messed up sleep settings in the microcontroller so it draws around .5W at idle. That may not seem like much, but given that whole "Energy Star" thing, and the fact the vast majority of other chargers nowadays consume 0 watts at idle, this is again indicative of major design oversights. (Also, 80% efficiency when most other chargers get 90%+.)

Data is attached as a PDF this time. Look if you want to see some pretty red ink.

Avoid this charger at all costs. It gets a rating of 1 star for the combination of failures. The only thing it has going for it is the internal circuitry seems capable of 6A output (at least), so with a MAJOR redesign to be compliant it might be fixable.

BONUS: This is the EXACT combination of failures that can cause the situation described here: plug in a LeEco or Moto Z with a USB-C headphone jack, and use an active cable, and you're driving straight into bat country.
https://plus.google.com/102612254593917101378/posts/7Cjq1yS2CHz

(Edit 7/26): I would like to point out minor gripes like poor efficiency are just indicative of larger systemic problems. You can read more about the DOE/Energy Star requirements for energy here:

http://electronics.stackexchange.com/questions/6123/how-much-power-is-really-wasted-by-a-wall-wart
http://www.cui.com/efficiency-standards

#USBC
24
7
Gianluca Bertoncelli's profile photoCa Cycleworks's profile photoKai S. (GokieKS)'s profile photoNathan K.'s profile photo
13 comments
 
I asked a question on the amazon site for this product. You might like it...
 
+Ca Cycleworks I'll keep an eye posted! Your question haven't been made live on the site yet, so I can't see it just yet.
 
+Ca Cycleworks It showed up! Oi. I actually had a question for you, actually. Maybe the public can benefit from this discourse too.

I know from recent events my analyses can be a bit pointed. But only few manufacturers have acknowledged they were planning to improve their product. [Tronsmart, and iOrange.] The rest haven't responded at all to my messages offering help fixing their designs. (Ideally a qualification lab should be doing this testing, not a member of the public.)

In your experience as a company, how willing have most manufacturing partners been to correct flaws? I also ask this since I'm speaking with an American manufacturer, Qualcomm, later today. (Their reasons for not changing are more business-oriented.)

When people play the "blame game", it's hard to encourage progress. But without an initial poke, it's hard to encourage action. Do you have any advice as a successful and well-respected organization doing your own QC?
 
Nathan, this situation is all about money. A company might have $100,000 to multiple millions of dollars tied up in those products. Further, if assembled "correctly", the cases won't open up to allow replacement of the circuit board. Far easier to build something than to take existing inventory and correct mistakes. So the companies choose to go on with the assumption that most people won't have a problem. The root cause of the situation is a too rapid rush to market without understanding the application. And if they do the right thing, beyond not selling the bad parts, they need to ensure recyclers won't resell the bad units on eBay, etc.

Our own FWIRES is a similar situation: I wasn't able to test them to quantify quality to my satisfaction -- but I'm not willing to knowingly let customers "have a problem". I just completed a pressure vessel to simulate a fuel tank, only with air pressure. I'm hoping some of the parts we bought end up testing good... Lots of money tied up in those. And beyond money, the FWIRES are desperately needed on the market.

I expect failures among the products we have manufactured for us overseas and at the very least, we perform a simple functional test. if need be, I'll build a test apparatus to evaluate the product more thoroughly. Sometimes the issue is not where the product is manufactured but instead "industry practice". If every factory that makes throttle position sensors (Europe or otherwise) tests them with 4 points, why do better? That's why we test ours with over 100 data points -- and why the reject pile of those range from 10% to 30%. We factor in the added cost of demanding quality better than industry standards when we set pricing for our customers.
 
Shameless plug, but this is what +Ca Cycleworks is talking about: Chris introduced me to MCU programming. (Sorry,  I'm a nerd, I can't help it, this stuff is awesome. :p)

[Timing belt test bench]:
https://www.youtube.com/watch?v=NwH9ZBMGF8Y
[TPS test bench]:
https://www.youtube.com/watch?v=cZK9fVvNaC4

Ducatis are high-performance motorcycles with a unique desmodromic valve train that uses rubber belts instead of metal chains in an ACTIVE INTERFERENCE engine. Meaning the valves will slam into the pistons if it fails.
https://en.wikipedia.org/wiki/Desmodromic_valve
https://en.wikipedia.org/wiki/Interference_engine

TPS stands for "throttle position sensor", or essentially the "gas pedal sensor". Both are super-critical-important, and his custom-built rigs are freaking impressive!
 
+Gianluca Bertoncelli Without testing it myself and seeing the data, I cannot responsibly say. But I have noted that with other manufacturers (Tronsmart) if one product is non-compliant all their others usually are in the same manner.

I'll add it to my Amazon Wishlist of USB-C stuff I need to take a look at later, time and budget willing.
 
+Gianluca Bertoncelli It is not permitted per the USB-C Spec 1.2. Furthermore, I am inclined to believe AUKEY made the same design mistakes as they did here, since they are the same company.
 
My question on Amazon is updated to say it conforms to the spec...
 
+Ca Cycleworks Oh boy. Well, I e-mailed the USB-IF directly about people like this. They seem more concerned about the fact that I don't misrepresent myself as being related to them. So I'll say it again, I'm a member of the public doing this tests on my own dime, and they should go to a USB-IF compliance workshop for proper help.
 
Hi. Have you tested the Anker PowerPort 5 USB-C or any other multi-port wall chargers with USB-C port(s)? There aren't many available, with the AUKEY and Anker being the more popular brands.
Add a comment...