DisplayPort over Type-C – Overcoming cable assembly test challenges
06 January 2017
Immediately after the release of the USB Type-C connector specification, VESA (Video Electronics Standards Association) readied the DisplayPort technology to operate as an Alternate Mode (Alt Mode) over this new connector.
(Click here to view article in digital issue)
As a result, DisplayPort now appears to be the go-to choice for video transport in designs that target the USB Type-C connector, which portends a marked shift in preferred display interfacing. This article will review the cable assembly test challenges for the DisplayPort standard and its implementation over the USB Type-C connector.
The “VESA DisplayPort Alt Mode on USB Type-C Standard” was released on September 22, 2014. The standard defines four scenarios in which the USB Type-C connector can be used to support DisplayPort, in a way that is fully interoperable with existing DisplayPort products.
• USB Type-C Cable (Scenario 1): use of DisplayPort as an Alternate Mode on the USB Type-C cable assembly
• USB Type-C to DisplayPort Cable (Scenario 2): cable assemblies that adapt the USB Type-C connector to the DisplayPort connectors
• USB Type-C to Protocol Converter (Scenario 3): cable assemblies for DisplayPort on USB Type-C to other video protocols
• USB Type-C to Multi-function Dock (Scenario 4): simultaneous use of USB Enhanced SuperSpeed and DisplayPort for docking applications
The electrical specification of the cable assembly is currently only defined for USB Type-C to DisplayPort cable (Scenario 2) and is based on the specification methodology for USB Type-C to USB Type-C passive cable assemblies. Since the DisplayPort AltMode compliance test requirements are not final, the material presented are the opinions of the author and do not necessarily reflect the opinions of VESA.
DisplayPort cable assembly test challenges
There are a number of challenges when working to integrate DisplayPort over USB Type-C into your products, while ensuring interoperability and achieving test compliance. First, to ensure correct measurements, the different impedances of the USB Type-C (85 ohm) and DisplayPort (100 ohm) need to be managed. In addition, a new signaling rate was added in the DisplayPort 1.3 specification, which increased the maximum bit rate to 8.1 Gbps/lane. This is a 50% increase in data rate and there is a lot more impact of loss, reflection, and crosstalk in your measurements. To ensure that the measurement is not affected by the test environment (such as test fixtures), a more rigorous approach to removing fixture effects is required. Finally, improvements in the pass/fail judgment method are also required. The traditional method does not allow for the trade-offs between each of the test items. Due to tighter margins, a new methodology is required to ensure sufficient yield.
1. Managing the different impedance environments
The first challenge is how to manage the different impedance environments, 85 ohm for the USB Type-C and 100 ohms for DisplayPort.
Measurement and calibration in a non-50 ohm environment may be difficult or untraceable. Therefore, measurement and calibration should be done in a 50 ohm environment (using conventional calibration kits and techniques) and the results renormalised to the desired impedance. The port reference impedance conversion function can be used to renormalise the USB Type-C port to 85 ohms.
Reference impedance is an important concept to understand when using S-parameters. Typically, S-parameters are stated as xxdB. However, to be exact, it should be stated as xxdB when yyO is used as the reference impedance. The reference impedance is typically 50O and is omitted, but in principle, it may be an arbitrary value. It is important to keep in mind that the S-parameter is a relative (normalised) value. In other words, the S matrix values change depending on the reference impedance. 50 ohm is only a reference value, so it can be changed.
A return loss measurement example using Keysight Technologies’ ENA Option TDR is shown in Figure 1. The port reference impedance conversion feature has been used to convert the measurements from 100 ohm on ports one and two (light trace) to 85 ohm on port 1 and 100 ohm on port 2 (dark trace). It is essential to manage the different impedances of the Type-C and DisplayPort environment in order to obtain correct results.
2. Removing fixture effects from measurement
Test fixtures are required to connect from the test equipment to the cable assembly. At the 8.1Gbps data rate, it is essential to remove the effects of the fixture to ensure sufficient yield. A similar approach to USB 3.1 is recommended. In the USB 3.1 Cable and Connector Compliance Specification, the “2x thru de-embedding” and “in-fixture TRL calibration” methods are introduced.
For 2x thru de-embedding, full calibration is performed with an electronic calibration (Ecal) unit, to establish the calibration reference plane at the end of the test cables. Then, the S-parameters of the fixture traces are de-embedded to extend the reference plane to the edge of the USB connectors, effectively removing the effects of the fixture from measurement. The key to the de-embedding method is the quality of the S-parameters of the fixture traces. The Automatic Fixture Removal (AFR) feature is recommended to obtain these S-parameters. The AFR function is available on Keysight’s N1930B Physical Layer Test System (PLTS) software. It provides a simple three step procedure to obtain highly accurate S-parameters of the fixtures. An alternative method is in-fixture TRL calibration, but compared to the 2x thru AFR method, additional standards are required. After completing the calibration, the reference plane is extended to the edge of the USB connectors.
A differential insertion loss measurement example of an USB 3.1 cable assembly, using both techniques to remove the effects of the fixture, is provided in Figure 2. The limit shown is for the USB Type-C cable assembly. As can be seen, not removing the test fixture effects (red trace) can be the difference between passing and failing a test item. The AFR (orange trace) and TRL (blue trace) results overlap each other, providing excellent correlation between the two methods.
3. New compliance methodology
Interconnects have traditionally been characterised by measuring parametric parameters, such as impedance in the time domain and insertion loss in the frequency domain. The limitation of the traditional parametric measurement is that it does not allow tradeoffs among the various test parameters. For instance, a channel with less loss could tolerate more crosstalk or reflection, and vice versa. As the parametric specification needs to guarantee interoperability for cables which marginally passes all the parametric test items, the limits are set conservatively. There will be cables which marginally fail in one or two parametric test items and pass other items with sufficient margin. Therefore, there is a possibility of rejecting a functioning cable assembly.
The DisplayPort Alternate Mode electrical specification is based on the Channel Metrics. There are three signal integrity impairments that impact the end-to-end link performance: attenuation, reflection, and crosstalk. The compliance specification is all about managing these three signal integrity impairments. To represent these three impairments, three parameters are defined as the Channel Metrics: insertion loss fit at Nyquist frequency (ILfitatNq), integrated multi-reflection (IMR), and integrated crosstalk (IXT). The ILfitatNq is calculated by applying curve fitting to the insertion loss measurement result. The difference between the insertion loss measurement and the smoothing curve, or the ripple of the insertion loss, represents the IMR. The IXT is defined as the integrated sum of all crosstalk sources between the DisplayPort lanes. The pass/fail judgment is performed based on these Channel Metrics. Since the IMR has a dependency on ILfitatNq, the limits are specified as a function of ILfitatNq.
Figure 3 shows several pass and fail examples from the specification. The x-axis is ILfitatNq and the y-axis is IMR. The green area represents the passing region, while the red area is the failing region. The passing area increases as ILfitatNq decreases. This pass/fail criterion allows tradeoffs between ILfitatNq and IMR. For example, a cable assembly may have more IMR if loss is smaller.
Several key topics aimed to ensure interoperability of DisplayPort Alternate Mode cable assemblies have been introduced. The port reference impedance conversion feature can be used to manage the different impedance environments of USB Type-C and DisplayPort. In addition, the 2x thru de-embedding using AFR removes test fixture effects, enabling you to measure the true performance of your device. Finally, there is a paradigm shift in the measurement methodology from the traditional parametric testing to the Channel Metrics method.
Contact Details and Archive...