USB4 LRD Active Cable Test And Certification

The USB4 active cable specification follows the USB type-C connector and cable spec. at present, the latest version is Rev. 2.1. The definition of EPR (extended power range) cable is added, and the ECN (engineer change notice) of previous active cable is integrated into this version. Let's take a look at the USB type-C active cable.

USB 3.2 Short Active Cable

USB4 Short Active Cable

USB 3.2 Optically Isolated Active Cable （OIAC）

Among them, OIAC is an optical fiber cable with a length of 50 meters. At present, it is defined that it can support USB 3.2 Gen2 speed at most (but does not support USB2.0 and VBUS power supply), which is mainly used for industrial end, machine vision, remote sensors, professional video and medical applications. However, the electrical specification of USB4 linear optical active cable has not been defined; next, we will introduce the part of short active cable.

Short active cable active cable

Within 5 meters in length

Full feature type-C and USB PD 3.0 eMarker are required

It is required to support two-way and positive and negative insertion

According to the support capability, it can be divided into USB 3.2 and USB4 active cables

USB 3.2 active cable:

-USB 3.2 Gen 2x2 (10GHz dual channel) must be supported

* active cables supporting only single channel (x1) are not allowed

-Alt mode optional support

USB4 active cable:

-All USB 3.2 and USB4 speeds (dual channel) must be supported

-Tbt3 alt mode must be supported

The wire requirements of VBUS, vconn, CC and USB 2.0 are consistent with the requirements of passive cables

The active cable needs to be powered by vconn

Active cable contains repeater components, such as re timer or re driver, mainly for high-speed signals TX1, TX2, rx1 and rx2. Re timer development is complex and expensive. The linear re driver cable (LRD) dominated by re driver has the characteristics of low complexity, low power consumption and low cost. Although it was only later added to the USB ecosystem and included into the USB type-C specification, the LRD cable was first introduced into the market. For example, the 40Gbps two meter thunderbolt 4 cable is the LRD cable supporting USB4.

The main components of LRD cable include RX equalizer and output driver, which are respectively responsible for compensating cable loss, adjusting DC gain, and adjusting the size of output pre emphasis and signal, as shown in the figure below.

Since LRD cable does not have CDR (clock data recovery), jitter and noise received at the cable input will be transferred to the cable output; At the same time, RX EQ may also amplify high-frequency noise; Due to the addition of active components, the impedance of the pad card will be discontinuous; As well as long cables, it is easy to cause length mismatch and large p-n skew in the production process, resulting in AC common mode exceeding the specification. Based on the above points, the following shall be considered in the design:

High frequency noise caused by cables

Whether the EQ inside the cable is properly balanced, and whether the signal is insufficient or over balanced?

Impedance matching of additional components inside the active cable

P-n skew of cable high speed pair

LRD cable function support and eMarker announcement

It is stated in table 6-3 of USB type-C connector & cable Rev 2.1 specification (as shown in Table 2), USB4 passive cable and USB4 active cable (except OIAC) must support USB4, usb3, USB2 and tbt3.

In particular, USB4 LRD cable is active cable, but in the ID header VDO setting of eMarker, B29... B27 must be declared as passive cable (011b) and declared using passive cable VDO. Because the LRD cable was incorporated into the USB specification at a later stage, it is mainly to be compatible with tbt3 products on the market. It must be declared as passive. Although the product type is declared as passive, tbt3 discovery will continue in the communication process of USB4 discover_ SVID (0x8087), and then judge whether it is "USB4 with tbt3 gen3 active cable".

Usb-if releases new USB4 logo and Icon

Usb-if held the USB devdays 2021 seminar in Seattle (September 30 to October 1), and released the new logo and icon of rated power of USB type-C cable in combination with the EPR (extended power range) specification, as shown in the table below, so that users can quickly identify the speed and power supported by USB products. The cable originally supporting 100W (20V / 5A) is no longer used; Cable supporting 5A must support EPR 240W (48V / 5A).

Table 3: new certified USB type-C cable logo and icon (source: usbdevdays 2021)

The LRD cable active cable adds an active component re driver to make the cable support length longer. In practical use and certification test, the concept is that the performance of LRD cable should be consistent with or even better than that of passive cable. That is, under the same test environment setting, the lrdcable must be equal to or better than the passive cable when compared with the passive cable.

LRD certification test items include:

Usb-c functional test

USB PD e-marker test

Active cable power: IR drop and power consumption

Thermal overheat protection

LRD electrical characteristic test

LRD cable compatibility test (still under discussion)

The detailed LRD cable electrical test is as follows:

1. Usb-c functional test

According to the usb-c function test specification, the cable test is as follows:

TD 4.1.3 Unpowered Cable Test

TD 4.13.5 Cable EnterUSB and Data Reset Test

TD 4.14.x

TD 4.14.1 Cable Vconn Swap Test

TD 4.14.2 Cable Reset Test

TD 4.14.3 Cable Alternate Mode Test

TD 4.14.4 Cable USB 3.2 Test

TD 4.14.5 Cable USB4 Test

2. USB PD: e-marker test

According to USB PD CTS, test the following three items related to cable:

Common Procedures and Checks

Physical Layer Specific Tests

Protocol Specific Tests

LRD cable needs to support tbt3 and confirm whether the tbt3 announcement and SOP response are correct

Cable SOP’ Discover Identity response

[ID Header VDO] B26 (model operation) set to 1b(alt mode)

[ID Header VDO] B29..27 (product type)set to 11b(passive cable)

[Cable VDO] B2..0 (USB highest speed) set to 010b(USB3.2/USB4 Gen2)

Cable SOP’ TBT Discover Mode VDO response

B20..19 (rounded/rounded & none) set to 01b(both)

B21 (Optical/none) set to 0 (none)

B22 (Re-timer/Re-driver) set to 0b (Re-driver)

B23 (Uni/Bi-directional) set to 1b (Uni)

B25 (Active/Passive) to 1b (Active)

3. Active Cable Power Requirements

3.1. The IR drop specifications of VBUS and ground cables are the same as those of passive cables

VBUS IR Drop：≤500 mV

Ground IR Drop：≤250 mV

3.2. The power supply of the active cable is mainly through the vconn, and the maximum power consumption is limited

Power consumed by vconn ≤ 1.5W

4. Thermal Test

For safety reasons, temperature sensing must be set inside the active cable. When the surface temperature of the plastic shell of the active cable reaches 80 ℃ ˚ C or metal surface temperature reaches 55 ˚ C. The USB 3.2/usb4 data transfer must be stopped.

In addition, the surface temperature of the active cable plug and the maximum working temperature of the shell surface shall not exceed the ambient temperature by 30 ℃ ˚ C. Or metal shell surface temperature 15 ˚ C。

The surface temperature (TS) of the plastic shell of the active cable mainly includes the working temperature (TMB) of the connected host and device motherboard, the active components in the cable, and the current ambient temperature (TA). The actual certification test is mainly divided into two parts: surface temperature (TS) and thermal shutdown. The test environment is shown in the figure below:

4.1 Surface Temperature (Ts)

surface temperature

As shown in Figure 3, test the connection, simulate the host / device motherboard through the thermal test fixture heater board at room temperature, and let the TMB temperature rise to (TA + 25) ˚ C) Then connect the active cable and set the full load from the host to the device (including simultaneous high-speed data transmission and PD 100W load). At this time, use the infrared camera to find out the highest temperature area of the cable plug (Fig. 4), and stick the "thermal couple" patch at this high temperature for temperature test (Fig. 5). Detect the surface temperature of the plastic shell of the cable plug, and judge whether the test has passed: TS < TA + 30 ˚ C。

Figure 4: infrared camera to find the highest temperature area

Figure 5: "ThermalCouple" patch adhered to the hottest area.

4.2 thermal shutdown overheating protection

The test environment for overheating protection is the same as above. In addition, the heating patch is coated on the plastic shell of cable plug (Fig. 6). Start heating the heating patch at room temperature. When the temperature reaches 85 ℃ ˚ C, the test results are determined: active cable needs to stop USB 3.2/usb4 data transmission.

Figure 6: flexible heater heating patch

Note: for active cable thermal test equipment and related accessories, please contact OD Liao@luxshare -ict.com

Electrical Test

For LRD cable, passive cable is still used in USB 2.0, SBU and CC cable configuration. The test method and specification are the same as that of passive cable. The high-speed signal pair TX1 / rx1 / TX2 / rx2 is equipped with re driver active components. The test specifications are in accordance with LRD active cable CTS version 0.8. The test items are mainly divided into the following three items:

Frequency domain test

Time domain – cable stand alone cable test

Time domain – cable output eye test

5.1 frequency domain test

Integrated Return-Loss (IRL)

Integrated Multi-Reflection (IMR)

Channel Operation Margin (COM)

Figure 7: connection diagram of S parameters captured by VNA of vector network analyzer

For frequency domain test items, the test method is the same as that of passive cables. The S parameters are retrieved by vector network analyzer VNA. There are 8 s4p files (TX1 / rx1 / TX2 / rx2, bidirectional) for high-speed differential pairs, and through get_ iPar_ V0p91a software for analysis.

5.2 time domain – cable stand alone test

Cable ilfit mask (DC / F1 / NQ / F2 / F3 / WB): insertion loss

OUTPUT_ Noise (𝝈): standard deviation of output noise (excluding nonlinear noise)

SIGMA_ E (𝝈): standard deviation of output nonlinear noise

Cable CM_ Noise: AC common mode

The cable body test (excluding system ISI and jitter) mainly tests the insertion loss, output noise, nonlinear noise and AC common mode of the cable itself. The test connection is shown in the figure below. Under the TP2 output (as shown in Table 4) pattern, swing, no jitter, no SSC, no TX EQ settings, such as "pattern: prbs15, swing 800mv, SSC off, jitter off, preset 0", the signal generator first connects the worst case passive cable for test, the oscilloscope captures the waveform *. Bin, then changes to the LRD cable for test, and carries out the following parameter analysis through the software, Then compare the LRD cable. The test result must be equal to or better than the passive cable. The test shall cover three speeds: USB4 Gen2 / gen3 and USB 3.2 Gen2.

Note: "worst case passive cable" refers to the passive cable with the maximum insertion loss within the cable specification, such as 1m USB 3.2 Gen2 passive cable, 2m USB4 Gen2 passive cable and 0.8m USB4 gen3 passive cable.

Fig. 8: connection diagram for cable body testing

5.3 time domain - cable output eye diagram test

(Cable output Eye Test）

5.3.1. USB4 Gen2 / gen3 test:

The connection of cable output eye diagram test (including system ISI and jitter) is shown in the figure below. The test environment is the same as the USB4 host / device RX receiver authentication test environment. The USB4 RX test environment needs to be corrected first. You can directly control the Anritsu mp1900 pattern generator through the grl-usb4-rx test app, as shown in Figure 9, Calibrate the USB4 RX test environment with keysight or Tektronix oscilloscope.

After calibration, connect the "worse case passive cable" first during the test. The test condition setting includes outputting prbs31 in pattern Gen and setting USB4 preset (16 groups in total). After the signal passes through the cable, capture five waveforms in the oscilloscope, and 80 waveforms need to be captured in each group at high speed; Then, under the same test conditions, remove the passive cable, replace the LRD cable, and capture the waveform on the oscilloscope; Then the eye diagram, eye width and eye area were tested and analyzed with USB4 sigtest software.

Figure 10: USB4 gen3 / Gen2 cable output eye diagram test connection

USB4 determination of Gen2 / gen3 test results (average of 5 captures):

LRD cable best eye area ≥ passive cable best eye area

And the eye width of LRD cable ≥ 0.9 * passive cable

1.3.2. USB Gen2 test:

The test environment is the same as the usb3.2 RX receiver authentication test environment. The USB 3.2 RX test environment needs to be corrected first (as shown in Figure 11). The Anritsu mp1900 pattern generator can be automatically controlled directly through the grl-usb3-rxtest app and calibrated with keysight or Tektronix oscilloscope

After calibration, under the USB 3.2 Gen2 RX test conditions, select Rx to calibrate the PJ @ 100MHz environment. After passing through the fixture and LRD cable, the oscilloscope captures the waveform five times, and then analyzes it with usb3sigtest software and 7 CTLE templates. (usb3 sigtest has only one ctle_5db template initially, and it is necessary to manually set and supplement ctle_0db ~ ctle_6 DB template)

Judgment of USB 3.2 Gen2 test results, (average value of 5 captures):

LRD cable best eye area ≥ passive cable best eye area

And the eye width of LRD cable ≥ 0.9 * passive cable

summary

USB type-C is widely used in computers and related peripheral devices, as well as passive cables and active cables. Some only support USB 2.0 and charging, and some can support USB 3.2 and USB4; They are all USB type-C connectors, which support different capabilities and speeds, which is easy to confuse users. USB association also focuses on user experience and is committed to a type-C cable, which can meet all applications. In the active cable part, the specification must support two-way transmission, positive and negative plug-in, dual channel (x2), etc. Taking the USB4 LRD active cable as an example, it can support USB4, USB 3.2, USB 2.0, thunderbolt 3, PD charging, etc. to meet the USB type-C application with one line.

The biggest difference between USB type-C active cable and passive cable is whether there are active components, which also leads to different methods for testing high-speed differential signals. The active cable test adopts the existing high-frequency test method of USB4 host and device, and uses high-frequency signal generator and high-frequency oscilloscope for testing. The test environment and methods are relatively complex, GRL provides automated testing solutions for USB4 testing, which can reduce testing complexity. GRL has rich experience in adjusting EQ, gain and other parameters and assisting customers in debugging. GRL can also provide USB4 host and device, USB4 passive cable, USB4 active cable and other testing and certification services.

Reference

USB Type-C Cable and Connector Specification, Release 2.1, May 2021

USB Type-C Connectors and Cable Assemblies CTS, Revision 2.1b, June 2021

USB4 ™ Thunderbolt3 ™ Compatibility Requirements Spec, Version 1.0, January 2021.

USB4 ™ Thunderbolt3 ™ Compatibility CTS, Revision 1.0, January 2021.

USB Power Delivery CTS, Revision: 1.2, Ver 2, June 20, 2021

USB Type-C Functional Test Specification, Chapter 4 and 5, May 23, 2021, Rev 0.88