27
What is the OnePlus 3 DASH charging/sync cable doing internally?? Let's probe it!

  1. sbooth
    US Brand Ambassador Jul 19, 2016

    sbooth , Jul 19, 2016 :
    Short of getting a razorblade out and physically dissecting the stock OP3 charging/sync cable to see how it plays its role in the DASH ecosystem, I decided to grab some USB development boards from my lab and see what is [electronically] going on inside this cable. Although I cannot guarantee there is or isn’t any internal logic within the USB-A or C connector side, most of the connectivity can be seen by probing each pin and seeing if any interaction is present.

    As I mentioned in my other thread regarding a USB3.0 extension cable working to lengthen the OP3 DASH cable, the USB3.0 TX/RX pins are missing except for the rear GND_Drain pin, which I thought may have something to do with the DASH feature. With this in mind, I noted the following observations:
    1. Yes, there is a 56K resistor tied between the +5VDC and the PDC [Power Data Communication] pin next to it.
    2. There is an interesting low resistance (4.6M ohm) tie-in between the GND_Drain pin and the other primary GND pins; like something is in-circuit facilitating a feature trigger (DASH??)
    Now, maybe someone needs to order an extra DASH cable or two from OnePlus and do a proper physical dissection of the cable. Any takers?? :eek::cool:

    Scott-

    op3_dash_usb1.png
     
    Last edited: Jul 19, 2016

    #1
  2. Joyyy
    Jelly Bean Jul 19, 2016


    #2
  3. OPFanBoyTillEnd
    Honeycomb Jul 20, 2016


    #3
  4. G_Max_Dobmann_Mlba
    Donut Jul 21, 2016

    G_Max_Dobmann_Mlba , Jul 21, 2016 :
    This might sound dumb, but did you test between the PDC Pins and the GND-Drain ? Are they directly connected ?
    And are the two PDC Pins connected together or independent from each other ?
     

    #4
    patrice58 likes this.
  5. IvanMuse
    Ice Cream Sandwich Jul 21, 2016


    #5
  6. sbooth
    US Brand Ambassador Jul 23, 2016

    sbooth , Jul 23, 2016 :
    Yes, everything was probed to confirm pin to pin continuity. The PDC pins are tied together but are independent of the GND_Drain and all other GND pins. When powered, A4/B4 receive the primary +5.20VDC from the DASH wall adapter while the 56K resistor linked PDC pins end up with approx. +5.15VDC. The other primary +5.20VDC is fed directly to A9/B9. At any rate, the more I think about, the more I believe the GND_Drain is utilized in conjunction with (but independent of) the PDC by the suspected logic inside the DASH wall adapter to enable the feature.

    Scott-
     

    #6
  7. mrbearbear
    Ice Cream Sandwich Jul 23, 2016

    mrbearbear , Jul 23, 2016 :
    I'll order it, u cut it open?
     

    #7
    Alvie, sbooth and sanmansan like this.
  8. sbooth
    US Brand Ambassador Jul 23, 2016

    sbooth , Jul 23, 2016 :
    Sure... will take photos too. :D
     

    #8
    mrbearbear likes this.
  9. G_Max_Dobmann_Mlba
    Donut Jul 23, 2016

    G_Max_Dobmann_Mlba , Jul 23, 2016 :
    I read on another post, that they successfully extended the Dash-Charge Feature through a USB 3.0 extension cable.
    With this information i did some testing on my own with a cheap usb 3.0 extension cable i cut apart and plugged into the carger.
    The "GND_Drain" Pin is fed 3.3 V from the charger and one of the USB-Datalines is at 3.3 V too (thought it dropped a little from the meters internal resistance, alluring to a weak pullup). The other line didn't carry any significant voltage.

    I didn't dare to hook my shiny new OP3 up to it though (i would really hate to damage it).

    I did some testing with the Dash-Cable too: The GND_Drain pin shows ~1400 mV in one direction on my diode-test the other direction is beyond the scale. So they have at least a silicon diode embedded inside the Cable.

    Maybe i can get ahold of the car charger unit when it is available again, i don't trust 240V enough to tear open my mains unit (plus i don't want to kill my only charger).
     

    #9
    kylesweet, NaterTots and sbooth like this.
  10. notcyf
    Honeycomb Jul 23, 2016

    notcyf , Jul 23, 2016 :
    240V is nothing man :rolleyes: i know out of experience :rolleyes::rolleyes::rolleyes:
     

    #10
  11. G_Max_Dobmann_Mlba
    Donut Jul 23, 2016


    #11
  12. G_Max_Dobmann_Mlba
    Donut Jul 23, 2016

    G_Max_Dobmann_Mlba , Jul 23, 2016 :
    Hooked my LA up to the Charger and found out that it tries to communicate with something when it senses a modest amount of power-drain. Seems to me like the third pin is used to query the cable for it's dash-capabilities via some one-wire protocol. When my spare USB-C charging cables have arrived i will try and capture the response from the cable, currently i don't want to fry my only USB-C cable... Screenshot_20160723_190858.png Screenshot_20160723_190939.png Screenshot_20160723_190959.png

    It seems like the communication is controlled by the carger placing a short low-pulse on the line and the transmitting end pulling the line low for either a long or short period. In the first 8 frames the charger seems to put information onto the bus by itself. Then it seems to provide "clocks" for more bits (presumably listening for the cable to answer something). I am currently looking for any one-wire protocoll that matches this scheme (assuming they used off-the-shelf parts).

    Edit: Did some more testing and the short high-pulse appears to be from noise and not part of the protocol. I think this might be just maxim one wire....
     
    Last edited: Jul 23, 2016

    #12
    kylesweet, MarkusRanz and sbooth like this.
  13. k3lvar
    Cupcake Jul 24, 2016

  14. G_Max_Dobmann_Mlba
    Donut Jul 24, 2016

    G_Max_Dobmann_Mlba , Jul 24, 2016 :
    The BYCI marked component appears to be bq2022 from ti. It seems like ti just copies maxims one wire protocol. That is why it looked like a 1w interface on the LA.
     

    #14
    kylesweet likes this.
  15. nufc1991
    Froyo Jul 24, 2016


    #15
    Andrei R., tjy724, mihaitaus and 5 others like this.
  16. mrbearbear
    Ice Cream Sandwich Jul 24, 2016

    mrbearbear , Jul 24, 2016 :
    Alright, I believe I have one for u. Pmd.
     

    #16
    sbooth likes this.
  17. sbooth
    US Brand Ambassador Jul 24, 2016

    sbooth , Jul 24, 2016 :
    Good info...
    That confirms a few things. So, there is a transistor on the GND_Drain pin so I that is likely triggering the DASH functionality. The 46M-Ohm readings to GND (A01/A12/B01/B12) from there are probably the transistor gate leaks. That little 603 sized SMT resistor next to the transistor is likely a zero ohm.

    On the USB-C side, the little resistor there looks to be the 56K to 5VDC PDC pin.

    So, in summary, not much going on in the cable itself, which leave us the wall adapter (or car adaptor) doing most of the DASH charging work. If @mrbearbear still wished for me take closer photos and confirm the transistor/resistor specs, I will do it. At the very least, we will have all the info to make our own (custom length) cables. In theory, we could also reverse engineer the wall adapter and/or car adapter, but that may be "more involved" and questionably cost-effective.

    Scott-
     
    Last edited: Jul 24, 2016

    #17
    kylesweet, mrbearbear and Vector like this.
  18. sbooth
    US Brand Ambassador Jul 24, 2016

    sbooth , Jul 24, 2016 :
    Likely... these little devices are common these days and there are multiple generics out there.

    bq2022a.png
     

    #18
  19. mrbearbear
    Ice Cream Sandwich Jul 24, 2016

    mrbearbear , Jul 24, 2016 :
    Yea, let's continue this teardown. I personally have no need for a longer cable, but I'm curious to see if people will make one.
     

    #19
  20. mrbearbear
    Ice Cream Sandwich Jul 24, 2016


    #20