This disclosure relates generally to electronic communication, and more particularly, to signaling over communication links such as Universal Serial Bus (USB) links.
The embedded universal serial bus 2.0 standard, known as eUSB2 or simply eUSB, is a physical layer supplement to the USB 2.0 specification that was developed to address the need for a low voltage USB 2.0 physical layer (PHY) solutions for embedded applications. As technology nodes for system-on-chips (SOCs) in laptops, smartphones, and tables became smaller, it became more difficult to support 3.3 volt input/output cells required for legacy USB signaling. The eUSB 2.0 standard drops signaling levels to 1.2 volts or lower, and provides enhancements to optimize power efficiency. eUSB is capable of supporting USB high speed, full speed, and low speed operation. The standard provides for two main operational modes, native mode and repeater mode.
The repeater mode, which is used for backward compatibility with legacy signaling speeds of USB 2.0, introduces a repeater device as an interface device between legacy and new generation devices. Repeaters are voltage level translators between host/peripheral SOC and legacy USB 2.0 peripheral/host devices.
There are some difficulties in designing such repeaters for high volume, low cost production. The legacy USB host and sink interface employs a differential to differential data transfer. However, with the eUSB standard, the embedded device provides a single-ended data signal for interfacing with legacy devices through a repeater, which converts it to a differential data signal. This differential data signal must meet the signaling specifications of the legacy USB2.0 standard. One particular signaling specification that has been difficult to meet with a low-cost repeater circuit is the crossover point. This crossover point specification should be met at both supported signaling speeds, 1.5 Mbps and 12 Mbps.
One known solution employs multiple output stages connected in parallel to the output terminal of the repeater. By sequentially providing the USB data, this solution activates stages sequentially to control the rise time and fall time of the signal. However, this solution is relatively complicated and expensive. Another known solution is to include a resistor-capacitor (RC) network at the output of the driver. However, this solution has difficulty adjusting to the different speeds required. Furthermore, such solutions will typically need extensive redesign in order to be used with newer, faster standards as the further updates to the USB standard are developed.
The present disclosure may be better understood, and its numerous features and advantages made apparent to those skilled in the art by referencing the accompanying drawings, in which:
The use of the same reference symbols in different drawings indicates similar or identical items. Unless otherwise noted, the word “coupled” and its associated verb forms include both direct connection and indirect electrical connection by means known in the art, and unless otherwise noted any description of direct connection implies alternate embodiments using suitable forms of indirect electrical connection as well.
SOC 102 is the host/peripheral USB controller and signaling chip which is typically coupled into the host system either to a processor or to a data fabric connected to multiple processing clients or other communication clients which may require communication over USB. SOC 102 includes two downstream ports for the eUSB devices labeled “eDSPn”, a downstream legacy port labeled “eDSPr”, and one or more upstream ports for communicating with the host system (not shown).
Each eUSB device controller 104 includes an upstream port labeled “eUSPn”, which is adapted to connect to SOC 102 over a differential eUSB link labeled “eD+/eD−”, which is a 1.8V differential link in this example.
On the right is depicted eUSB2 repeater 106, which includes an upstream port labeled “eUSPh” and a legacy downstream port labeled “uDSP”. The eUSPh port connects to SOC 102 over a link labeled “eD+/eD−”, however this link is not employed in a normal differential signaling mode, but instead each side of the link is used for single-ended signaling depending on the mode of the link, as further discussed below. The uDSP port connects to legacy USB device 108 over a legacy USB differential link labeled “D+/D−”, which are used herein to refer both to the terminals and to the signals on the terminal.
It is noted that while in this example, the repeater is located on the host device, an eUSB repeater may instead be used on the peripheral device to couple an eUSB compliant peripheral device to a legacy USB host controller.
Repeater 106 has a first input/output terminal connected to an “eD+” differential signaling input/output of SOC 102, a second input/output terminal connected to an “eD−” differential signaling input/output of SOC 102, both part of the eUSPh port. On the downstream side, repeater 106 has a third input/output terminal adapted for connecting to a “D+” input/output of USB device 224, and a fourth input/output terminal adapted for connecting to the “D−” input/output of USB device 224, both part of the uDSP port. Repeater 106 includes a pair of receivers 202 and 204, a multiplexer 206, a digital control circuit 208, an inverter 210, a pair of multiplexers 212 and 214, a pair of inverters 216 and 218, and a pair of drivers or output buffers 220 and 222.
Receivers 202 and 204 are connected to the eD+ and eD− inputs of repeater 106. (It is also noted that because USB is uses bidirectional signaling, a pair of drivers, not shown, is also included at each port in the upstream direction, and a pair of receivers in the downstream direction.) Receivers 202 and 204 are supplied with a 1.2V supply as depicted, and the outputs of receivers 202 and 204 are both fed to multiplexer 206 and digital control circuit 208.
Multiplexer 206 has two inputs receiving the outputs of receivers 202 and 204, respectively, and an output connected to the input of inverter 210 and to the input of multiplexer 212. Digital control circuit 208 has two inputs connected to the outputs of receivers 202 and 204, a first output connected to control multiplexer 206, a second output connected to multiplexer 212, and a third output connected to multiplexer 214. Inverter 210 receives the output of multiplexer 206 and provides an inverted version of this signal to an input of multiplexer 214, while multiplexer 212 receives an un-inverted version. The outputs of multiplexers 212 and 214 provide un-inverted and inverted versions of the signal to inverters 216 and 218, respectively, which feed drivers 220 and 222. Drivers 220 and 222, in this implementation, are inverting drivers and operate on the higher voltage, typically 3.3V, for driving a differential signal to legacy USB device 224.
The arrows on
With the correct input selected at multiplexer 206, the output of multiplexer 206 is used to create inverted and un-inverted versions of the signal for transmission to legacy USB device 224, as depicted by the signals along the driver output path to the D+ and D− connections.
For interfacing with legacy USB2.0 devices, the legacy USB1.1 data transfer rates for low speed (1.5 Mbps) and full speed (12 Mbps) are supported by using a repeater such as repeater 106. If either the USB host or sink is an eUSB device then the data transfer is single ended to differential as depicted for the case of an eUSB host and a legacy USB sink. Repeater 106 is used between eUSB and USB2.0 as an interface to convert the signals accordingly while meeting USB2.0 specs at the interface, including higher signaling voltages for communicating over external ports through USB cables.
Importantly, meeting the crossover specification of 1.3V to 2V at the USB2.0 interface is required for both speed modes. This is challenging to do with a single, low-cost driver circuit because there is a large difference in the specifications for rise time and fall time for LS mode (75 nS) and FS mode (4 nS).
Multiplexer 506 includes two inputs connected to eD+ and eD− terminals of the repeater for receiving signals in eUSB format, and a control input labelled “LS/FS” for receiving a control signal from a digital control circuit such as circuit 208 (
Delay capacitors 534 and 536 are selectively connected to the inputs of output buffers 520 and 522, respectively under control of delay capacitor control circuits 530 and 532. Delay capacitor control circuits 530 and 532 each have an input receiving a control signal labelled “LS/FS or I2C”, and an output connected to a switch for activating the delay capacitors 534 and 536, respectively, to connect them between the output terminal and the negative power rail. The signal LS/FS or I2C, in various embodiments, can be implemented as a binary signal indicating whether LS or FS mode is active, or an I2C (Inter-Integrated Circuit) bus connection or another type of control signal connection able to control delay capacitor control circuit 530 and 532 to provide desired values. An I2C bus may be employed, for example, to communicate capacitor values for programmable capacitors.
In operation, delay capacitors 534 and 536 act to tune the crossover by delaying only the rising edge, or only the falling edge of the signal in the D+ or D− path will ensure meeting the cross over specification across wide range of load conditions (50 pF to 600 pF).
In some implementations, for LS and FS modes, different delay settings are selected automatically to meet the crossover spec across different load conditions (50 pF to 600 pF). Preferably, no adjustment is required to the operation of output buffers 520 and 522 to achieve the desired crossover point, which is instead achieved with a change in capacitance provided by delay capacitors 534 and 536.
A delay, for example one applied by delay capacitors 534 and 546, is labeled “Delay during Tr”. In diagram 620, points 624 depict the crossover point in LS mode if no delay is applied. Points 624, under certain load conditions on the D+ and D− terminals, is above the required crossover point of 2V in the USB 2.0 specification, because of the speed of output buffers designed to meet the faster FS mode specifications. Points 626 depict the crossover point when a delay is applied to only the rising edge of the D+ and D− signals. The effect, in this example, is to lower the crossover point back to within the specified range, and more preferably to near the center of the specified range to at or around the typical crossover point of 1.65V (for a 3.3V signaling voltage).
At block 702, the process receives a single ended signal, in this example from an eUSB SOC. At block 704, the signal is routed from a multiplexer to a driver path of the repeater. In the implementation of
At block 706, the process selects a delay amount based on the speed mode detected. At block 708, a differential signal is created at the repeater. At block 710, the selected delay is applied to one of, but not both, of the rising edges or the falling edges of the differential signal, as depicted in
In some embodiments, the delay is adjusted at block 712 to account for a capacitive load mismatch on the board or cabling signal path for the differential signal. The adjustment at block 712 is also able to adjust for mismatches between the D+ and D− sides of the signal path, which is a common problem in USB. For example, a programmable delay capacitor may be used, or a parallel combination of multiple delay capacitors, which are individually controlled to adjust the total delay capacitance individually for each side of the differential signal, producing a different delay on one side the differential signal than that on the other.
Thus, various embodiments of a USB repeater circuit, an apparatus including such a circuit, and a corresponding method have been described. The various embodiments provide for a low-cost yet reliable repeater design for using common output buffers to provide for different signaling rates and still achieve a desired crossover point.
The above-disclosed subject matter is to be considered illustrative, and not restrictive, and the appended claims are intended to cover all such modifications, enhancements, and other embodiments that fall within the true scope of the claims.
Thus, to the maximum extent allowed by law, the scope of the present invention is to be determined by the broadest permissible interpretation of the following claims and their equivalents, and shall not be restricted by the forgoing detailed description.
Number | Name | Date | Kind |
---|---|---|---|
7173453 | Prather | Feb 2007 | B2 |
20070064778 | Lacy et al. | Mar 2007 | A1 |
20200350899 | Huang | Nov 2020 | A1 |
Number | Date | Country | |
---|---|---|---|
20230315663 A1 | Oct 2023 | US |