“The introduction of USB Power Delivery (PD) has greatly improved the power capability of the USB interface protocol, with the potential to support systems up to 100W. With the transfer of data, it is now possible to supply more power to Electronic/electrical devices using only one cable. Therefore, both the space used and the installation cost can be reduced. USB PD means that the power direction is no longer fixed.
By Arun Pappan and Martin Tan, engineers at FTDI chips
The introduction of USB Power Delivery (PD) has greatly improved the power capability of the USB interface protocol, with the potential to support systems up to 100W. With the transfer of data, it is now possible to supply more power to electronic/electrical devices using only one cable. Therefore, both the space used and the installation cost can be reduced. USB PD means that the power direction is no longer fixed. So, depending on the situation, the device can act as a sink (drawing power from VBUS) or a power supply (supplying power through VBUS). Devices can utilize the PD specification for high-power applications to enumerate connected hardware. After that, they can provide the necessary power. Charge-through units are one example.
There are many potential approaches to designing charge-through devices. The principle is that power from the charger is transferred through port 2 to port 1 to charge connected hardware such as a laptop. However, this can be achieved in a number of different ways, each with its own specific advantages and disadvantages.
In this section, we describe two different design approaches that can be used for vSafe5V implementation. In one design, vSafe5V also passes through port 2, while in another, vSafe5V is generated internally.
Figure 1: Schematic depicting vSafe5V charging pass-through implementation via port 2
The vSafe5V through port 2 is depicted in Figure 1. When connected hardware requests a higher power profile, port 2 will negotiate the same profile. Once port 2 completes the negotiation process, port 1 will notify the laptop that power is ready.
The main advantage of this approach is that it keeps the design simple and does not require the use of an internal 5V generator. However, this route also has some drawbacks. Since vSafe5V is pass-through, the voltage level and current capacity will be directly dependent on the port 2 power supply. If the input voltage is close to the lower limit of the PD specification (4.75V to 5.5V), the vSafe5V may fall below the specification on the Port 1 side. This is due to the voltage drop across the circuit. Another disadvantage to be aware of is that the high voltage negotiation process has to go through an extra negotiation step on the port 2 side.This can be a bit time consuming, so there are
May exceed PD timing specifications.
Figure 2: Schematic diagram of onboard vSafe5V pass-through implementation
Another option available is to generate vSafe5V on board, as shown in Figure 2. In this case, vSafe5V can be generated in full compliance with the usbpd specification. The example shown here uses the laptop’s receiver capabilities and determines the best charging mode for the connected device. This profile is then negotiated over port 2 and the voltage level is set before the role switches to port 1. Once the switch to port 1 is complete, it can set the higher profile requested by the connected hardware, since the charging profile is already available on port 2.
This design layout is advantageous because the performance of the vSafe5V can be predicted. The reason is that vSafe5V is generated internally. Also, the negotiated voltage does not have to be renegotiated on port 2, which will speed up the turnaround time experienced. However, this approach has one glaring disadvantage that needs to be acknowledged. This type of design cannot be used for some hosts, i.e. hosts that switch between multiple profiles. Based on the various hosts and hubs tested by the FTDI engineering team, differences in the implementation of power negotiation were observed between different manufacturers.
In addition to the vSafe5V detailed above, there are other factors to consider when designing a charge pass-through unit. These include:
1. VBUS discharge – Since the charge pass-through will initially function as a receiver, engineers must be aware of the need for VBUS discharge. When a device is disconnected as a source, it needs to revert to its original role as a sink. At the same time, VBUS must discharge across the entire pass-through path for a specified period of time.
2. Voltage Drop – The voltage drop across the entire pass path must be minimized.
In order to ensure that the power paths in the partial discharge system operate in an efficient manner, load switches controlled by the pass-through partial discharge devices are used. This load switch will consist of a pass transistor (usually a MOSFET with an on/off control block). Since the charge current for partial discharge can be as high as 5A, it must be remembered that the drain-source on-resistance (RdsON) in the load switch must be low so that the power loss involved is not too great and the voltage at port 1 remains at the partial discharge within the specification.
3. Inrush Current Limiting – When a charger is connected to port 2 and a charging device is connected to port 1, an inrush current may occur if a capacitive load is switched to the power rail. The magnitude of the inrush current will depend on the rise time of the voltage rise and the load capacitance. Steep voltage ramps will increase inrush current and cause a momentary dip in VBUS. This can cause connected hardware to reset itself, which needs to be avoided. In addition to affecting the functionality of the connected hardware, this condition may damage or reduce the operational life of the load switch components. Although a large load capacitance will reduce the transient voltage drop, it will increase the magnetizing inrush current. Therefore, the load switch must be slew rate controlled to extend the rise time of the voltage ramp, and the load capacitance must be optimized to limit inrush current and droop in VBUS. Another benefit of having slew rate control on the load switch is that as an overcurrent protection measure (due to high inrush currents) it prevents the charger from shutting down the pass-through, which would obviously cause partial discharge charge interruptions.
4. Internal Power Consumption – For the purpose of internal power consumption, obviously a certain amount of power needs to be deducted from the charging mode. This needs to be taken into account when determining the operating parameters of the system.
The overall design goal of a pass-through capable PD device is to ensure a smooth connection and power supply between a wide variety of USB PD hosts or hubs. Engineers will be able to achieve this if they ensure they fully understand all the items discussed in this article. FTDI provides the advanced dual-port IC technology needed to support the development of USB power delivery systems, as described above. The company’s power delivery ICs allow hardware to switch from a sink to a source without any interruption to the data flow.