A selective introduction to USB cables
Created
Modified
Published
Introduction
USB cables are ubiquitous and standardized, which is a vast improvement over the assorted mess of proprietary charging connectors used for mobile devices in the 1990s and early 2000s. However as the USB standard has grown it has become more complex, especially as USB-C cables have great variety in their capabilities, well beyond the 2.5 W maximum power and data / no data for previous generations of USB cables. Those differences in USB-C cable capabilities are not visible as physical differences in connector shape. If you don’t use the right USB-C cable, your devices will may take many hours longer to charge, and data will be slow to transfer. Conversely, you might have a good cable but a poorly performing charger. If you don’t know what your cable can handle, how can you tell the difference? I recently got a USB cable tester which helped immensely in determining the properties of my USB cables, but it left me with many questions such as:
- Why does the USB-C connector have 24 pins when USB-A connectors only have 4? Does this mean a USB-C cable contains six times as many wires as a micro-USB cable?
- Micro-USB connectors have 5 pins but USB type A connectors have only 4. What’s the deal with the extra pin? Do micro-USB cables have 4 wires or 5 wires?
- What is the shell of a USB cable? Should it be electrically connected to the ground pin or not?
- I have a micro-USB cable that works fine for charging my e-reader, but if I connect it to a computer I can’t get it recognize the e-reader so I can transfer books. But then if I try with a different micro USB cable it works fine. What’s going on?
- I’ve seen those USB cables with a blue USB A port that are faster than regular USB cables. Could I get a micro-USB cable with one those blue ports and get faster data transfer to my e-reader? If no, why not?
- What are eMarkers and why do some USB-C cables have them and others don’t? Why don’t older micro-USB cables need eMarkers?
- I have a laptop dock with a bunch of USB-C ports coming out of it. I want to charge my phone from one of these dock ports. What USB-C cable should I use so that I can charge it as fast as I could with a dedicated wall adapter for a phone or laptop?
- USB-C cables are supposed to be reversible, but some of my USB-C to micro-USB adapters only work in one orientation. Why is this?
- I got a cheap electric razor while I was traveling that uses USB-C. It came with a USB-A to USB-C cable to charge it that works fine. However if I plug the razor into the nice USB-C charger that I use for my phone or laptop it doesn’t charge at all! What gives?
- I have an Anker power bank that supports Qualcomm Quick Charge 3.0. It uses a micro-USB connector to charge. What cable do I need to get the benefits of Quick Charge?
The answers to all these questions and more I have now finally learned and will explain below. However, I will only discuss the basics of USB cables, not the USB protocols or architecture.
USB cable terminology
USB connectors make a physical and electrical connection at an interface; these connectors have exposed electrical contacts called pins. In a USB cable, pins can be connected to wires running through the cable, or they can be left floating. The USB standards distinguish between plugs and receptacles. Plugs and receptacles contain pins that connect as mated pairs of a given type; for example USB type-A plugs connect to USB type-A receptacles. Physical contact between pins of a plug and receptacle forms the electrical connection. Plugs are the connectors at the end of the cable, and receptacles are the connectors in a host, device, or hub. Hosts are computer systems with a CPU and operating system, devices are peripherals that provide a function (e.g. mouse, keyboard, printer, microphone, webcam, flash drive, external hard drive), and hubs are a special kind of device that provide additional USB connections.
USB 1.x connectors
A USB type-A connector has 4 pins:
| Pin | Signal Name | Description |
|---|---|---|
| 1 | VBUS | Voltage bus (+5 V) |
| 2 | D− | Data− |
| 3 | D+ | Data+ |
| 4 | GND | Power ground |
| Shell | Shield | Drain wire |
(In the USB 1.1 spec see “Table 6-1. USB Connector Termination Assignment”.) VBUS stands for voltage bus (+5 V for power) and GND is ground (to complete the circuit). D+ and D- are for data transfer.
USB cables with these connectors have 4 wires, with the data wires arranged as twisted pair.
| Wire | Pin | Name | Description |
|---|---|---|---|
| 1 | 1 | VBUS | Voltage bus (+5 V) |
| 2 | 2 | D− | Data− |
| 3 | 3 | D+ | Data+ |
| 4 | 4 | GND | Power ground |
| Shield | Shell | External copper braid for shielding |
A USB type-B connector has the same pins as type-A, but is mechanically different so cannot be plugged into a USB type-A port.
This way a computer with a type-A port will always act as the host and a printer with a type B port will always act as a peripheral and there is no ambiguity about which end of the cable should go into each device. At this point there is no support for devices which may act as as either hosts or peripherals, but this will show up again in USB 2.0.
USB type-A and type B connectors were standardized in 1996 for USB 1.0. USB 1.1 was standardized in 1998 and added additional specifications for charging behavior but did not add any new connectors. The USB 1.0 specification is 268 pages long. The USB 1.1 specification is 311 pages long (PDF is 327 pages total).
USB 2.0 connectors
This revision added many things, including new connectors. Mini-A, Mini-B, Micro-A, and Micro-B connectors all have the same 5 pins:
| Pin | Name | Description |
|---|---|---|
| 1 | VBUS | Voltage bus (+5 V) |
| 2 | D− | Data− |
| 3 | D+ | Data+ |
| 4 | ID | Host or peripheral |
| 5 | GND | Ground |
This is the same as USB 1.x except for the ID pin. To explain the purpose of this pin, here’s the motivation: devices like GPS navigators, phones, and tablets have a small USB connector like Micro- or Mini-. This is fine if you want to charge your tablet or connect it to your laptop and act like a storage device to transfer files, but what if you want to connect your tablet to a peripheral like a USB flash drive, keyboard or printer, possibly using a hub? In this case your tablet needs to act as a USB host instead of a peripheral, and so would need to have a separate USB-A port in addition to the Micro-B port. But the tablet might not be physically big enough for a regular USB-A port, so instead we use a smaller connector such as Micro-A. Unfortunately then we still need two separate receptacles on the tablet: a Micro-A and a Micro-B. What if instead we could use a single receptacle that does both?
This is the concept behind USB On-the-Go (OTG), which allows devices with Mini- or Micro- receptacles to act as either host or peripheral. There are A and B versions of Mini- and Micro- plug for this purpose: they can fit in the same AB receptacle, but the physical shape of the A and B plug is different. For example, say you have a tablet with a Micro-AB receptacle. You can get a cable with USB Micro-A on one side and Micro-B on the other, and plug the A side into your tablet and the B side into your printer. But how does the tablet know that it’s connected to a Micro-A plug and not a Micro-B plug?
Well, that’s what the ID pin is for: so a device can decide whether to act as host or peripheral. It does this by checking the resistance between the ID pin and ground pin. If the ID pin is shorted to ground (i.e. near-zero resistance), then it knows this is a Micro-A plug and it should behave as a host that supplies power, but if the ID pin is floating (i.e. open circuit or near-infinite resistance between ID and ground) it knows this is a Micro-B plug and it should behave as a peripheral that receives power. (There are also additional behaviors for intermediate resistances that I am ignoring for sake of brevity.) Regardless, this at most requires an additional resistor inside the connector at one side of the cable. This is a purely passive component, and the ID pin is not responsible for any data transfer between devices. This means that USB cables with these connectors still only need 4 wires, which are the same as USB 1.x:
| Wire | Pin | Name | Description |
|---|---|---|---|
| 1 | 1 | VBUS | Voltage bus (+5 V) |
| 2 | 2 | D− | Data− |
| 3 | 3 | D+ | Data+ |
| 4 | 4 | GND | Power ground |
| Shield | Shell | External copper braid for shielding |
In practice not all USB 2.0 cables provide connectivity on the data pins D− and D+. Cables missing these data lines are common and are sometimes called “charge-only USB cables”, although properly speaking these cables do not follow the USB spec and should not have the USB trident logo. Manufacturers make these primarily because they contain half the number of copper wires (two instead of four) and are therefore cheaper to make. On the user side there is effectively no benefit to these cables. There is a theoretical security risk that an untrusted USB port could install malware on a device that connects to it (sometimes called juice jacking), but the risk of this actually happening in the wild is low and there are no documented criminal incidents using this technique. (Furthermore this can be averted with a data blocking USB adapter, sometimes called a USB condom.) By contrast, things like credit card skimmers are well-documented and a much higher risk.
USB-OTG and mini-USB was standardized in 2000, and micro-USB was standardized in 2007. The USB 2.0 specification from April 2000 is 622 pages (PDF 650 pages total). USB-OTG is rare nowadays, as tablets and phones have USB-C receptacles instead of micro-AB receptacles.
USB 3.0 Micro-B
Micro-B 3.0 connectors (a.k.a Micro-B SuperSpeed) have have 10 pins instead of 5 and a different physical connector shape than Micro-B 2.0. Micro-B 3.0 is shaped like a Micro-B 2.0 with an extra “sidecar” attached on the side with those extra 5 pins.
This is designed so that you can insert a Micro-B 2.0 plug into a Micro-B 3.0 receptacle but you can’t insert a Micro-B 3.0 plug into a Micro-B 2.0 receptacle. (Personally I find this connector terminology confusing, I would prefer to to call it USB 2.0 Micro-B and USB 3.0 ExtendedMicro-B or something rather than USB 2.0 USB Micro and USB 3.0 Micro-B. This would also make it much easier to do a web search for these kind of cables.)
Here’s the list of pins for Micro-B 3.0 (see Table 5-5. USB 3.0 Micro-B Connector Pin Assignments):
| Pin | Signal Name | Description |
|---|---|---|
| 1 | VBUS | Power (voltage bus, 5 V) |
| 2 | D− | Data− |
| 3 | D+ | Data+ |
| 4 | ID | OTG identification, left open |
| 5 | GND | Ground for power return |
| 6 | SSTx− | SuperSpeed transmit− |
| 7 | SSTx+ | SuperSpeed transmit+ |
| 8 | GND_DRAIN | Shielded ground for SuperSpeed |
| 9 | SSRx− | SuperSpeed receive− |
| 10 | SSRx+ | SuperSpeed receive+ |
| Shell | Shield | Connector metal shell |
This is the same familiar 5 pins as USB 2.0 Micro-B with 5 more pins added. The purpose of USB 3.0 was faster data transfer for things like external hard drives (hence the SuperSpeed branding), as well as bi-directional data transfer. .. TODO: why was bi-directionality important? Don’t we normally just copy files to an external hard drive? The 5 new pins (SSTx-, SSTx+, GND_DRAIN, SSRx-, SSRX+) are to enable this high-speed data transfer, which requires shielded cabling to reduce noise, SSTx stands for SuperSpeed transmit and SSRx stands for SuperSpeed receive.
These new pins also require new wires, and each shielded differential pair (SDP) needs its own ground wire, so USB 3.0 cables need ten wires in total and are typically thicker than USB 2.0 cables. See Table 5-7. Cable Wire Assignments:
| Wire Number | Signal Name | Description |
|---|---|---|
| 1 | PWR | Power |
| 2 | UTP_D- | Unshielded twist pair, negative |
| 3 | UTP_D+ | Unshielded twist pair, positive |
| 4 | GND_PWRrt | Ground for power return |
| 5 | SDP1- | Shielded differential pair 1, negative |
| 6 | SDP1+ | Shielded differential pair 1, positive |
| 7 | SDP1_Drain | Drain wire for SDP1 |
| 8 | SDP2- | Shielded differential pair 2, negative |
| 9 | SDP2+ | Shielded differential pair 2, positive |
| 10 | SDP2_Drain | Drain wire for SDP2 |
| Braid | Shield | External copper braid for shielding |
Wire 7 and 10 (SDP1_Drain and SDP2_Drain) are both connected to GND_DRAIN on the connectors, which accounts for the discrepancy between 10 wires but only 9 connected pins.
Meanwhile, USB 3.0 Type-A connectors have no ID pin, so 9 pins in total (see Table 5-2. USB 3.0 Standard-A Connector Pin Assignments):
| Pin | Signal Name | Description |
|---|---|---|
| 1 | VBUS | Power (voltage bus, 5 V) |
| 2 | D− | Data− |
| 3 | D+ | Data+ |
| 4 | GND | Ground for power return |
| 5 | SSRx− | SuperSpeed receive− |
| 6 | SSRx+ | SuperSpeed receive+ |
| 7 | GND_DRAIN | Shielded ground for SuperSpeed |
| 8 | SSTx− | SuperSpeed transmit− |
| 9 | SSTx+ | SuperSpeed transmit+ |
| Shell | Shield | Connector metal shell |
Note that this time the receive SDP comes first and the transmit comes second. Since the physical type-A connector for USB 3.0 has 9 pins, they are colored blue to distinguish them from USB 2.0 connectors. However everything is still backward compatible with USB 2.0 and even USB 1.0 and USB 1.1.
The USB 3.0 specification was published in 2008; the PDF is 482 pages long.
USB-C connectors
USB-C connectors have 24 pins. However the connector is symmetric and the pins are laid out 12 on each side, with each side labelled A and B. The A pins and B pins are mostly mirror images of each other. The full table is (see Table 3-4 USB Type-C Receptacle Interface Pin Assignments):
| Pin | Name | Description |
|---|---|---|
| A1 | GND | Ground return |
| A2 | SSTXp1 (“TX1+”) | SuperSpeed differential pair #1, transmit, positive |
| A3 | SSTXn1 (“TX1−”) | SuperSpeed differential pair #1, transmit, negative |
| A4 | VBUS | Bus power |
| A5 | CC1 | Configuration channel |
| A6 | D+ | USB 2.0 differential pair, position 1, positive |
| A7 | D− | USB 2.0 differential pair, position 1, negative |
| A8 | SBU1 | Sideband use (SBU) |
| A9 | VBUS | Bus power |
| A10 | SSRXn2 (“RX2−”) | SuperSpeed differential pair #4, receive, negative |
| A11 | SSRXp2 (“RX2+”) | SuperSpeed differential pair #4, receive, positive |
| A12 | GND | Ground return |
| Pin | Name | Description |
|---|---|---|
| B12 | GND | Ground return |
| B11 | SSRXp1 (“RX1+”) | SuperSpeed differential pair #2, receive, positive |
| B10 | SSRXn1 (“RX1−”) | SuperSpeed differential pair #2, receive, negative |
| B9 | VBUS | Bus power |
| B8 | SBU2 | Sideband use (SBU) |
| B7 | D− | USB 2.0 differential pair, position 2, negative |
| B6 | D+ | USB 2.0 differential pair, position 2, positive |
| B5 | CC2 | Configuration channel |
| B4 | VBUS | Bus power |
| B3 | SSTXn2 (“TX2−”) | SuperSpeed differential pair #3, transmit, negative |
| B2 | SSTXp2 (“TX2+”) | SuperSpeed differential pair #3, transmit, positive |
| B1 | GND | Ground return |
Now that we have seen the progression up to USB 3.0 (SuperSpeed), this should look mostly familiar. We have the same ground, voltage bus, and data pairs as USB 1.0, just with two VBUS and two GND, accounting for six of the twelve on each side. We also have the four pins for SuperSpeed transmit and receive, accounting for another four of the twelve on each side. The ID pin is gone, instead we have configuration channel pins CC1 and CC2. These take on the role of the ID pin, with resistance to ground specifying its role as sink or source. They are also necessary for higher-power delivery methods specified by USB-PD.
Finally, we have a genuinely new pin type, the sideband use pins SBU1 and SBU2. These are for things like DisplayPort and audio accessories, like those USB-C to headphone jack adapters.
Fortunately we don’t need 24 wires for USB-C cables, since e.g. the four GND pins and four VBUS pins can be tied together. In total we only need 15 wires for a USB-C cable, or up to 18 wires since there are 3 optional wires. (See Table 3-10 USB Full-Featured Type-C Standard Cable Assembly Wiring)
| Wire | Pin(s) | Wire name | Pin name | Description |
|---|---|---|---|---|
| Braid | Shell | Shield | Shield | Cable external braid for shielding |
| 1 | A1, A12, B1, B12 | GND_PWRrt1 | GND | Ground for power return |
| 2 | A4, A9, B4, B9 | PWR_VBUS1 | VBUS | Bus power |
| 3 | A5 | CC | CC | Configuration channel |
| 4 | A6 | UTP_Dp | D+ | Unshielded twisted pair, positive |
| 5 | A7 | UTP_Dn | D− | Unshielded twisted pair, negative |
| 6 | Plug 1 A2, Plug 2 B11 | SDPp1 | Shielded differential pair #1, positive | |
| 7 | Plug 1 A3, Plug 2 B10 | SDPn1 | Shielded differential pair #1, negative | |
| 8 | Plug 1 B11, Plug 2 A2 | SDPp2 | Shielded differential pair #2, positive | |
| 9 | Plug 1 B10, Plug 2 A3 | SDPn2 | Shielded differential pair #2, negative | |
| 10 | Plug 1 B2, Plug 2 A11 | SDPp3 | Shielded differential pair #3, positive | |
| 11 | Plug 1 B3, Plug 2 A10 | SDPn3 | Shielded differential pair #3, negative | |
| 12 | Plug 1 A11, Plug 2 B2 | SDPp4 | Shielded differential pair #4, positive | |
| 13 | Plug 1 A10, Plug 2 B3 | SDPn4 | Shielded differential pair #4, negative | |
| 14 | Plug 1 A8, Plug 2 B8 | SBU_A | Sideband use A | |
| 15 | Plug 1 B8, Plug 2 A8 | SBU_B | Sideband use B | |
| 16 | A1, A12, B1, B12 | GND_PWRrt2 | Optional extra ground wire | |
| 17 | A4, A9, B4, B9 | PWR_VBUS2 | VBUS | Optional extra bus power wire |
| 18 | B5 | PWR_VCONN | VCONN | Optional, for powered cables |
This accounts for all the pins except B6 and B7, which are not populated:
Contacts B6 and B7 should not be present in the USB Type-C plug.
This allows room for other things like the 5.1 kOhm pull-down resistors. (The UTP_Dp and UTP_Dn are called unshielded but in practice they are shielded under the cable braid just like everything else.)
One thing to note is that VBUS can be higher than +5 V for USB-C. This allows for high power, all the way up to 240 W for cables that can handle it. How do power supplies know if the cable can handle it? For cables that go above 5 V and 3 A, they cannot just use passive components on CC to signal, they must have an active component called an e-Marker that signals the cable’s capabilities. This special e-Marker chip is powered by the VCONN wire. Similarly, for data transfer rates above 5 Gbps (the USB 3.0 rating), there must be a special chip for handling that.
While the 24 pins for USB-C make sense in context, it should be evident at this point that USB-C cables are considerably more complex than prior generations of USB cables, to say nothing of the protocol level complexity of USB4. I hope this additional complexity was worth it, because USB-C has been around for more than 10 years at this point and we are stuck with it for the foreseeable future.
USB-C connectors were standardized in 2014; revision 2.0 of the spec was published in 2019. USB4 was also standardized in 2019.
Answering the questions
Now that we know more about how these cables function, let’s tackle those questions from the beginning.
Why does the USB-C connector have 24 pins when USB-A connectors only have 4? Does this mean a USB-C cable contains six times as many wires as a micro-USB cable?
The 24-pin connector is two pairs of 12 pins. The 4 pins from USB type A are still there, and the additional 8 pins are necessary for high-speed data transfer, power level configuration, and functionality like DisplayPort and headphone jack peripherals. USB-C cables don’t have 24 wires, they have between 15 and 18 wires, which is significantly more than 4 but only a few more than USB 3.0 cables that require 10 wires.
Micro-USB connectors have 5 pins but USB type A connectors have only 4. What’s the deal with the extra pin? Do micro-USB cables have 4 wires or 5 wires?
By specification, USB 2.0 cables have four wires: two for power and ground and two for data transmission. Mini-USB and Micro-USB connectors have 5 pins; the additional pin is called the ID pin. USB Mini- and Micro- plugs come in A and B variants, and the resistance between the ID pin and ground pin distinguishes between A and B. This allows e.g. a device with a micro-AB receptacle to tell whether a micro-A plug or a micro-B plug has been inserted, and switch between host and peripheral roles accordingly as specified in the USB On-the-Go (OTG) standard. However, this functionality is only relevant for devices with a mini-AB or micro-AB receptacle.
What is the shell of a USB cable? Should it be electrically connected to the ground pin or not?
The metal shell surrounds the pins and is the visible metal outer part of the connector. It is electrically connected to the metal braid or foil that surrounds the insulated wires of the cable; this metal braid may also be connected to an additional drain wire to help dissipate current for safety reasons. For cables with a USB-C connector, the USB specification requires that the shell be electrically connected to the ground pin (“Shield and GND grounds shall be connected”). For older cables, such as USB-A to micro-B, I was unable to determine if this is required by the standard. See details below.
I have a micro-USB cable that works fine for charging my e-reader, but if I connect it to a computer I can’t get it recognize the e-reader so I can transfer books. But then if I try with a different micro USB cable it works fine. What’s going on?
Some cheaper USB cables do not have the data wires that the spec-compliant USB cables do, so instead of four wires they just have two. These “charge-only” cables have no real benefit over proper data-enabled USB cables and should generally be avoided.
I’ve seen those USB cables with a blue USB A port that are faster than regular USB cables. Could I get a micro-USB cable with one those blue ports and get faster data transfer to my e-reader? If no, why not?
The blue USB A port cables are USB 3.0 cables. Your e-reader has a USB 2.0 micro-B receptacle. You can get a USB 3.0 micro-USB cable but it is not compatible with the USB 2.0 micro-B receptacle in your e-reader. This is because your e-reader does not have the physical connector or software support for the high-speed duplex data transfer that USB 3.0 provides. The same is true for USB 2.0 Mini-B cables.
What are eMarkers and why do some USB-C cables have them and others don’t? Why don’t micro-USB cables need eMarkers?
eMarkers are for allowing higher power transfer in USB-C to USB-C cables. USB-C already has full USB 3.0 wiring for 5 Gbps data transfer without needing an eMarker. Micro-USB does not support these higher power ratings and so does not have an eMarker.
I have a laptop dock with a bunch of USB-C ports coming out of it. I want to charge my phone from one of these dock ports. What USB-C cable should I use so that I can charge it as fast as I could with a dedicated wall adapter for a phone or laptop?
The most important thing here is to check the specs of your dock. Many of them do not support USB PD and will be limited to 15 W or less and the choice of USB-C cable will not matter. For example, here are the specs for the ports on a Dell WD19TB from 2019:
- Front USB 3.1 Gen1 x1: Dell PowerShare BC1.2; 2 A @ 5 V (max 10 W)
- Front USB 3.1 Gen1/Gen2 Type-C: 1.5 A @ 5 V (max 7.5 W)
- Rear USB 3.1 Gen1 x2: 0.9 A @ 5 V (max 4.5 W)
- Rear USB 3.1 Gen1/Gen2 Type-C with DisplayPort 1.4 x1: 1.5 A @ 5 V (max 7.5 W) @ 5 V (max 15 W)
- Rear Thunderbolt Type-C Port: 3 A @ 5 V (max 15 W)
This makes some sense because the entire dock is running off its own adapter that is either 130 W or 180 W. The priority of the dock is getting power to the laptop, not to the other peripherals connected to the dock. In this case you’re better off with a dedicated wall adapter that implements USB PD 3.0, which goes up to 100 W. USB PD 3.1 goes up to 240 W but this is overkill for a phone and most devices and cables do not support it.
USB-C cables are supposed to be reversible, but some micro-USB to USB-C adapters only work in one orientation. Why is this?
Short answer: one side of the VBUS pins are not connected.
Long answer: Most new devices use USB-C but some devices still need micro-USB, such as my e-reader and older power banks. It’s convenient to be able to plug a little adapter onto the end of the USB-C cable and then just charge that way. Unfortunately, all of the adapters I’ve found only connect half of the power pins. For example, they connect these pins:
- GND: A12, A1, B12, B1
- VBUS: A9, B4
- D+: A6, B6
- D-: A7, B7
However, they do not connect the A4 and B9 VBUS pins. This means that the USB-C connector is no longer reversible, it only connects to power in one orientation. As far as I know there’s no reason to omit these pins, maybe they just ran out of space to route these connections?
I got a cheap electric razor while I was traveling that uses USB-C. It came with a USB-A to USB-C cable to charge it that works fine. However if I plug the razor into the nice USB-C charger that I use for my phone or laptop it doesn’t charge at all! What gives?
TODO: discuss this and the USB-C pulldown resistors
I have an Anker power bank that supports Qualcomm Quick Charge 3.0. It uses a micro-USB connector to charge. What cable do I need to get the benefits of Quick Charge?
Short answer: an ordinary micro USB cable will work as long as the data pins are connected.
Long answer: Unlike USB PD, Qualcomm Quick Charge 3.0 is a proprietary protocol, i.e. not part of the USB standard. It works by applying a series of voltages along D+ and D-. One of the patents is here:
https://patents.google.com/patent/US20140122909A1/
This is just one of many non-standard proprietary protocols, there’s also VOOC (from OPPO), PumpExpress+ (from MediaTek), TurboPower (from Motorola), Smartpower (from HUAWEI) and Adaptive Fast Charging (from Samsung), just to name a few. These kinds of proprietary protocols are generally bad for the same reason proprietary connectors are bad: unilateral control by a single vendor, incompatibility between protocols means Qualcomm chargers and Motorola devices don’t work together, the specification isn’t publicly documented and requires signing an NDA, manufacturers have to pay a fee to to the patent holder just to make a compatible device, etc. On the bright side, at least in this case Qualcomm realized that requiring a non-standard USB cable was a non-starter, so only the charger and device need to implement the proprietary protocol to be compatible. They confirm this in the FAQ:
Q: Does it matter what type of charging cable is used with a Quick Charge adapter?
A: Quick Charge is designed to be connector- and current-independent. Quick Charge is designed to be compatible with a variety of formats, including:
- USB Type-A
- USB micro
- USB Type-C
- Proprietary connectors
Quick Charge high-voltage operation is designed to minimize charging issues associated with long or thin cables, allowing for a superior charging experience, independent of cable type or cable current capability.
Notes on grounding
The USB 1.0 standard has this to say about grounding:
6.6 Grounding
The shield must be terminated to the connector plug for completed assemblies. At the host end, the shield, DC power, and chassis ground should be bonded together. The complete bus should have only one DC ground point at the host end. All other devices should not connect the shield or DC return to chassis ground. This prevents circulating low frequency currents. However, AC coupling is permitted for EMI compliance. The coupling impedance must be less than 250 kΩ at 60 Hz and not greater than 15 Ω between 3 and 30 MHz. The dielectric voltage rating of the capacitor must be 250 Vac (rms).
The USB 2.0 standard is more brief:
6.8 USB Grounding
The shield must be terminated to the connector plug for completed assemblies. The shield and chassis are bonded together. The user selected grounding scheme for USB devices, and cables must be consistent with accepted industry practices and regulatory agency standards for safety and EMI/ESD/RFI.
The USB specification for USB-C cables requires electrically connecting the ground pin to the shell:
Shield and GND grounds shall be connected within the USB Type-C plug on both ends of the cable assembly.
This is also true for USB-A to USB-C cables:
Shield and GND grounds shall be connected within the USB Type -C and USB 2.0 Standard-A plugs on both ends of the cable assembly.
https://www.usb.org/sites/default/files/USB%20Type-C%20Spec%20R2.0%20-%20August%202019.pdf
However, in practice many USB cables leave the connection between the ground pin and the shell open. This is definitely out of spec for USB-C cables. For other cables such as a USB-A to micro-B cable I wasn’t able to determine if this is required, since in principle this could be provided by the host and device at either end. As best I can tell connecting shell/shield and ground pin is optional for USB cables without a USB-C connector.
The shield one is separate and should go though the RC filters. Although there are different opinions on that at least for USB2 devices.
https://www.eevblog.com/forum/beginners/drain-wire-in-usb-3-0-(gnd_drain)-not-a-signal-return-path/
This is a surprisingly contentious topic, although most of the debate appears to be on the device side, not the cable side.
- https://electronics.stackexchange.com/questions/116392/understand-the-shielding-of-a-usb-device
- https://electronics.stackexchange.com/questions/389972/usb-shield-to-ground-or-not-to-ground?noredirect=1&lq=1
- https://electronics.stackexchange.com/questions/404536/why-is-the-usb-shield-always-tied-to-ground-gnd-pin
- https://electronics.stackexchange.com/questions/677134/usb-peripheral-connecting-usb-shield-to-gnd
- https://forum.arduino.cc/t/grounding-or-not-grounding-an-usb-shield-that-is-the-question/1430891
- https://forums.adafruit.com/viewtopic.php?t=199596
- https://hackaday.com/2024/09/13/where-do-you-connect-the-shield/
- https://sierraconnect.protoexpress.com/t/best-practices-for-grounding-usb-high-speed-cable-shields/3579
- https://www.eevblog.com/forum/beginners/on-devices-side-should-i-connect-usb-cable-shielding-to-the-black-wire/
- https://www.eevblog.com/forum/projects/tying-usb-shield-to-ground-%28vbus-0v%29-always-a-complete-fail/
- https://www.eevblog.com/forum/projects/usb-shield-ground-connection/
Additional sources
- https://acroname.com/blog/mastering-usb-c-how-usb-c-establishes-connection
- https://electronics.stackexchange.com/questions/291413/usb-c-to-usb-a-pinout
- https://electronics.stackexchange.com/questions/35462/why-does-micro-usb-2-0-have-5-pins-when-the-a-type-only-has-4
- https://nassaunationalcable.com/blogs/infographics/usb-c-cable-color-code
- https://somanytech.com/usb-c-wiring-color-code-pinout-wiring-diagram-of-charger/
- https://www.digikey.com/en/articles/decoding-the-usb-standards-from-1-to-4
- https://www.digikey.com/en/articles/how-to-simplify-connectivity-using-usb-c
- https://www.electronicproducts.com/usb-type-c-connector-pins-and-signal-plan/
- https://www.pshinecable.com/article/why-are-there-no-b6-and-b7-pins-on-the-pcb-of-the-type-c-male-connector.html
- https://www.theverge.com/2023/11/6/23948486/usb-c-cables-marking-speed-power-delivery-elgato
- https://www.ti.com/lit/wp/slly017/slly017.pdf
- https://www.usb.org/sites/default/files/D1T1-2%20-%20USB%20Type-C%20System%20Overview.pdf