ALPHA A little bit on type-C power distribution (PD)
anonymous 2018-11-29 20:23:26 34301 Views10 Replies After trying two USB type-c meter/diagnosis tools (one is a popular UM25C/UM34c) without much a success,
now that a new tool (WEB-U2) just arrived today, I'd like to share my finding...
1. first I have to upgrade WEB-U2's firmware to latest (now ver 4.7, very easy to do) to get a stable tester and probably more features
2.if I cascade other testers or odd ball cables, most cases the type-C protocol negotiation (over CC1 / CC2 leads < -- not D+ D-
leads) will fail. In such case, the power adapter only supplies 5V to Alpha and the Alpha will be as cold as rock. See #5, most likely stuck in BC1.2 DCP 5V 1.5A mode.
3. LPA supplied 45W power adapter with type-C plug, if direct connects to LPA, it will work, certainly. I also tried Apple 61W (seems to be a clone, since in one of WEB-U2's menu, it couldn't detect its S/N), this one has a type-C receptacle. I can use a type-C cable (plugs on both ends) to power LPA no problem.
4. With WEB-U2 added in the loop above, I saw the protocol used is PD 2.0 negotiated to 15V 3A in either Apple 61W or LPA 45W. Its screen show it's in "MTK PE" mode (should be equivalent to PD 2.0 mode.). Very important to note that both D+ and D- are left floating, most likely not connected to each other, and read as 0V. < -- this is critical (see later)
5. Important to note that LPA's 45W unit not only supports PD 2.0 (this is what this power supply arrangement should be), it also supports several others (such as QC2.0 up to 12V, QC3, BC 1.2 DCP 5V 1.5A.) The Apple 61W supports PD 2.0 as well as BC1.2 DCP 5V 1.5A.
6. Now the sequence is important... if I used #2 to trip LPA into "only getting 5V supply" feed, by using #4 method will not recover the LPA into (asking PD2.0 mode as default), i.e., even when using #4 method at this point, it is still getting 5V power feed, not PD 2.0's 15V feed.
Then if I use method #3, LPA is back to asking PD2.0 as the negotiation method and the system is now back to normal.
7. I am guessing somewhere in LPA side, type-C receptacle's D+ and D- still has some voltage and/or D+ D- is in a non-PD source or PD sink mode. This is the challenge a universal type-C (PD source, PD sink, accessory mode/OTG, and video mode) all-in-one design gave to us.
All my guesstimate and few things learned.
I am leaning toward a simpler power adapter (e.g., only support PD2.0 mode, no QC2, QC3, BC1.2 stuff) and probably taking out D+ and D- leads ( not sure, since USB PD is a complex standard) cto the power adapter end may help.
Anyway, I am back to 12V direct power DC plug solution. Easy, clean, and can also supply external devices are all big pluses.
now that a new tool (WEB-U2) just arrived today, I'd like to share my finding...
1. first I have to upgrade WEB-U2's firmware to latest (now ver 4.7, very easy to do) to get a stable tester and probably more features
2.if I cascade other testers or odd ball cables, most cases the type-C protocol negotiation (over CC1 / CC2 leads < -- not D+ D-
leads) will fail. In such case, the power adapter only supplies 5V to Alpha and the Alpha will be as cold as rock. See #5, most likely stuck in BC1.2 DCP 5V 1.5A mode.
3. LPA supplied 45W power adapter with type-C plug, if direct connects to LPA, it will work, certainly. I also tried Apple 61W (seems to be a clone, since in one of WEB-U2's menu, it couldn't detect its S/N), this one has a type-C receptacle. I can use a type-C cable (plugs on both ends) to power LPA no problem.
4. With WEB-U2 added in the loop above, I saw the protocol used is PD 2.0 negotiated to 15V 3A in either Apple 61W or LPA 45W. Its screen show it's in "MTK PE" mode (should be equivalent to PD 2.0 mode.). Very important to note that both D+ and D- are left floating, most likely not connected to each other, and read as 0V. < -- this is critical (see later)
5. Important to note that LPA's 45W unit not only supports PD 2.0 (this is what this power supply arrangement should be), it also supports several others (such as QC2.0 up to 12V, QC3, BC 1.2 DCP 5V 1.5A.) The Apple 61W supports PD 2.0 as well as BC1.2 DCP 5V 1.5A.
6. Now the sequence is important... if I used #2 to trip LPA into "only getting 5V supply" feed, by using #4 method will not recover the LPA into (asking PD2.0 mode as default), i.e., even when using #4 method at this point, it is still getting 5V power feed, not PD 2.0's 15V feed.
Then if I use method #3, LPA is back to asking PD2.0 as the negotiation method and the system is now back to normal.
7. I am guessing somewhere in LPA side, type-C receptacle's D+ and D- still has some voltage and/or D+ D- is in a non-PD source or PD sink mode. This is the challenge a universal type-C (PD source, PD sink, accessory mode/OTG, and video mode) all-in-one design gave to us.
All my guesstimate and few things learned.
I am leaning toward a simpler power adapter (e.g., only support PD2.0 mode, no QC2, QC3, BC1.2 stuff) and probably taking out D+ and D- leads ( not sure, since USB PD is a complex standard) cto the power adapter end may help.
Anyway, I am back to 12V direct power DC plug solution. Easy, clean, and can also supply external devices are all big pluses.