HWZ Forums

Login Register FAQ Mark Forums Read

Xiaomi 45W USB-C power adapter review: I still can't believe this thing only costs S$20

Reply
 
LinkBack Thread Tools
Old 20-12-2017, 12:42 AM   #1
Member
 
Join Date: Nov 2010
Posts: 106
Xiaomi 45W USB-C power adapter review: I still can't believe this thing only costs S$20

http://www.hardwarezone.com.sg/revie...only-costs-s20

A USB-C power adapter can only support 1 type of quick-charging: USB-C Power Delivery or Qualcomm’s Quick Charge 1.0, 2.0 or 3.0. Both methods cannot co-exist. (QC 4.0 can apparently co-exist).

Sources:
https://www.extremetech.com/mobile/2...y-incompatible

https://plus.google.com/+BensonLeung/posts/cEvVQLXhyRX

Appreciate it if you could revisit your tests to confirm which quick charging the Xiaomi adapter conforms to, and edit the article accordingly.
ryusho is offline   Reply With Quote
Old 21-12-2017, 08:31 AM   #2
Master Member
 
chucks's Avatar
 
Join Date: Jan 2000
Posts: 2,975
Hi ryusho, thanks for your feedback.

Yes, per USB spec, USB-C doing PD cannot do other proprietary charging methods at the same time. That is, until QC4+ and PD 3 / PPS arrive. I've briefly talked about that in my other Nintendo Switch article: http://www.hardwarezone.com.sg/featu...power-delivery

That said, that hasn't prevented charger makers to do things their own way to support both PD and QC in the same unit. The Xiaomi 45W charger is one of them. I've known at least one other brand that is doing it the "by left" way.

Are these makers breaking spec? Well, you can draw your own conclusions.
chucks is offline   Reply With Quote
Old 23-12-2017, 01:06 AM   #3
Member
 
Join Date: Nov 2010
Posts: 106
Hi chucks,

Thanks for your reply!

I don't have this particular charger nor am I an electrical engineer, but based on what I've read thus far, the only possible way a power adapter can support both at the moment is to adhere to all USB PD "Power Rules", and then subsequently using the D+/D- (usb data) lines for the Qualcomm QC solution, such that devices which don't support USB PD can revert to QC 3.0 after the initial handshake (or is it vice versa?). And because there's no use for the usb data lines here since it's a power adapter, it still works as intended. I presume devices that don't support any quick charge methods or their own proprietary quick charge like Oppo's or OnePlus' would then revert to USB BC Spec 1.2.

I guess the biggest concern for consumers besides price would be whether an incompatible charger or one that supports multiple quick charge methods would cause any issues to the battery of the device, and in the process voiding any warranties on the device. Would the theoretical ability for 2 types of quick charging methods to co-exist give rise to a potential over/under supply of power? Based on USB PD, a compatible PD device would signal the amount they are able to draw, and the PD manager within the charger regulates the power drawn. But with QC together, it would allow QC to change the voltage delivered on VBUS, and create mixed signals on how much power to provide. Of course, quite a few devices right now have inserted limitations on the maximum charge they can take in (Google Pixel 2 XL, and the latest iPhones as you mentioned in your update) that may limit any potential damage.

Are you able to share details of your test measurements? Would be very interested to see the various changes in power output over time (from initial handshake to full charge), especially if you are using a performance analyzer such as the one from Granite River Labs, which would provide a lot more details than a normal USB C power meter.
ryusho is offline   Reply With Quote
Old 23-12-2017, 03:25 PM   #4
Master Member
 
chucks's Avatar
 
Join Date: Jan 2000
Posts: 2,975
Hi ryusho,

For a non-EE, your understanding of the matter is very good.

Yes, the meddling of the VBUS voltages is the reason why USB-IF “forbids” anyone implementing the C and PD spec to also support other charging methods. But as you know, the different parties sort of got together, which is why we now have PD 3 / PPS as well as QC4+ that plays nice with PD.

The Xiaomi charger is primarily a PD charger, and supports 5 PD power rules: 5 / 9 / 12 / 15 / 20V. The charger (or more specifically, the 3rd party PD controller in the charger) also has a secondary logic for HVDCP detection for enabling QC. Regardless whether a PD or QC device is connected, it starts off with DCP 5V (BC1.2 compatible). The way I read it from my measurements is, like other QC chargers, it still monitors D+/D- for initial negotiation. Upon successful handshake it'll remove the D+/D- short and enter HVDCP mode, and continue to monitor the D+/D- lines. In QC/HVDCP mode, it doesn't go beyond 12V, which means it adheres to QC3 Class A protocol. When a non-QC or PD device is detected, QC and HVDCP are disabled and it'll fall back to legacy 5V/1.5A. On the other hand, PD uses CC lines to discover and configure VBUS.

Once the "handshake" is successful and the appropriate charging mode engaged - be it for QC or PD - the charging behavior becomes predictable, and a lot of it now depends on the PMIC of the connected device. For example, on the iPhone 8, at 0%, the charger doesn't support 2.7V on the D+/D-, so it'll only engage 5V/1A or 5W charging (instead of 2.1A (10W) or 2.4A (12W) like Apple's iPad chargers) until after power up and upon entering PD mode. The iPhone's sink PMIC seems to also have a very strict temperature or charge level threshold, and I say that because regardless of PD charger/powerbank, the iPhone's maximum draw from the power source always tapers off very quickly. For instance, the next charging stage for the iPhone 8 (as well as 8 Plus and X) is at 9V with up to ~15W power draw, followed by ~9-10W (still at 9V), and then 5V/1A, and lastly, 5V with gradually dropping current till the phone is fully charged and enters trickle-charging stage. In short, every device's PMIC is different.

Regarding regulation of power, that's on both power source and sink (connected device). A good controller will also have VBUS, DP/DN short circuit protections. If you're wondering if the charger will push out a PD 20V to a QC device that can't accept it and hence damage the device, I'd say that's very unlikely. That's the whole point for the initial negotiation. In cases when handshake is broken, the charger will fall back to BC1.2, which means while the device isn't being charged at the fastest possible speed, it's still being charged at a safe voltage and current.

Last edited by chucks; 26-12-2017 at 05:34 PM..
chucks is offline   Reply With Quote
Reply
Important Forum Advisory Note
This forum is moderated by volunteer moderators who will react only to members' feedback on posts. Moderators are not employees or representatives of HWZ. Forum members and moderators are responsible for their own posts.

Please refer to our Terms of Service for more information.


Thread Tools

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are On