MIPI DevCon Bangalore 2017: Next generation MIPI Physical Layer Design and Evaluation Challenges
1 like498 views
The document discusses updates and challenges associated with the next-generation physical layer specifications for MIPI C-Phy, D-Phy, and M-Phy in 2017, including changes to eye diagram test methodologies. Key updates include enhancements in transmission rates, voltage levels, and calibration improvements across the different MIPI standards. Additionally, it highlights the importance of direct connections for testing accuracy and the implementation of spread spectrum clocking to reduce signal interference.
MIPI DevCon Bangalore 2017: Next generation MIPI Physical Layer Design and Evaluation Challenges
- 1. SK Choi Keysight Technologies Next Generation MIPI Physical Layer Design and Evaluation Challenges
- 2. © 2017 MIPI Alliance, Inc. Agenda • New specification and CTS changes in 2017 • Eye diagram tests changes and challenges – MIPI C-PHYSM – MIPI D-PHYSM – MIPI M-PHYÒ • SSC test – MIPI D-PHYSM 2
- 3. © 2017 MIPI Alliance, Inc. Specification updates in MIPI C-PHY 3 • TX Pre-Emphasis(TxEQ option) – from Pre- emphasis method to de-emphasis method • RCLK jitter(reference clock jitter) – Annex to chapter 9 • Receiver calibration – removing PVT(Process, Voltage and Temperature variation after Long LP signal) C-PHY 1.1 : Approved Feb 11, 2016 C-PHY 1.2 : Approved Mar 28, 2017 3
- 4. © 2017 MIPI Alliance, Inc. CTS updates in MIPI C-PHY 4 CTS 1.0 : Approved Feb 12, 2016 CTS 1.1 : expected approve Aug, 2017 • Test 1.2.21 – Tx Eye Pattern Test • Test 1.4.1 – HS-TX Differential Voltages Unterminated • Test 1.4.2 – HS-TX Differential Voltage Mismatch Unterminated • Test 1.4.3 – HS-TX Single-Ended Output High Voltages Unterminated • Test 1.4.4 – HS-TX Static Common-Point Voltages Unterminated
- 5. © 2017 MIPI Alliance, Inc. Specification updates in MIPI D-PHY 5 • Up to 6500Mbps with Short reference channel (8K support) • Lower LP voltage level from 1.2V to 1V • HS-Idle (lower latency) • Programmable Preamble(RX PVT calibration due to LP signal) D-PHY 2.0 : Approved Mar 8, 2016 D-PHY 2.1 : Approved Mar 28, 2017
- 6. © 2017 MIPI Alliance, Inc. CTS updates in MIPI D-PHY 6 CTS 1.2 : April 24, 2017 • Test 1.5.7 – HS-TX Eye Diagram • Test 1.4.19 – TX Spread Spectrum Clocking(SSC) Requirement • ZID open case test • Direct connection supporting in HS continuous mode. CTS v2.0/v2.1 CTS 2.0/2.1 : expected finished in October
- 8. © 2017 MIPI Alliance, Inc. CTS updates in MIPI M-PHY 8 CTS 3.1 : On-going(revision 21) CTS 4.0/4.1 : On-going(revision 1) • Test 1.1.7 – HS-TX G3 and G4 Differential AC Eye (TEYE-HS-G3/G4-TX, VDIF-AC-HS-G3/G4- TX)
- 9. © 2017 MIPI Alliance, Inc. Agenda • New specification and CTS changes in 2017 • Eye diagram tests changes and challenges – MIPI C-PHYSM – MIPI D-PHYSM – MIPI M-PHYÒ • SSC test – MIPI D-PHYSM 9
- 10. © 2017 MIPI Alliance, Inc. Eye Diagram Test - General Refence channelEye diagram MIPI C-PHY MIPI D-PHY MIPI M-PHY 10
- 11. © 2017 MIPI Alliance, Inc. Eye Diagram Test Challenges for MIPI C-PHY/D-PHY • RTB(Reference Termination Board) can’t support new specifications 500Msps 1Gsps 1.5Gsps 2.0Gsps 2.5Gsps Same data, +/-250mV HS swing, 82ps R/F time, without reference channel However, eye diagram is distorted MIPI C-PHY RTB MIPI D-PHY RTB Click to add text sps : Symbol Per Second 11
- 12. © 2017 MIPI Alliance, Inc. MIPI C-PHY/D-PHY Eye Diagram Test Setup 2 Direct connection to scope(HS only) DUT Oscilloscope § Now, PHY WG defines direct connection to oscilloscope § However, RTB is still required for LP to HS timing test. MIPI D-PHY MIPI C-PHY TX HS Test Setup 2 Direct connection to oscilloscope (HS, 100ohm termination only) 12
- 13. © 2017 MIPI Alliance, Inc. Resolving Issue with Direct Connection • Direct connection provide more accurate result on tests 500Msps 1Gsps 1.5Gsps 2.0Gsps 2.5Gsps Same data, +/-250mV HS swing, 82ps R/F time, without reference channel sps : Symbol Per Second 13
- 17. © 2017 MIPI Alliance, Inc. MIPI M-PHY Eye Diagram Test • Test setup is same on both HSG3 and HSG4 but testing points has changed. DUT CH1/CH2 Oscilloscope CH1/CH2 PKG CTLE PLL PLL DFE Gear3 Gear4 MIPI M-PHY HS G3/G4 test setup MIPI M-PHY HSG3/G4 Test point 17
- 18. © 2017 MIPI Alliance, Inc. MIPI M-PHY Embedding Channel + Package Model DUT CH1/CH2 Oscilloscope CH1/CH2 PKG CTLE PLL PLL DFE Gear3 Gear4 -3dB at 5.83GHz 18
- 20. © 2017 MIPI Alliance, Inc. MIPI M-PHY RX Equalizer-CTLE • Not likely another application, M-PHY CTLE has vary wide range of zero pole value and Adc value What if A company think 2.5dB Adc + 400MHz Fz is optimal CTLE value B company think 0dB Adc + 400MHz Fz is optimal CTLE value? Does it correlated between? 20
- 21. © 2017 MIPI Alliance, Inc. MIPI M-PHY RX Equalizer-CTLE Same waveform but only change Adc value from 2.5dB to 0dB 2.5dB Adc case Fz = 400MHz 0dB Adc case Fz = 400MHz Adc : CTLE DC gain Fz : CTLE zero frequency 21
- 25. © 2017 MIPI Alliance, Inc. Test Setup Tip 4 Use SigTest tool Commonly used for High speed digital interface • USB • PCIe Provide similar result between oscilloscopes A company B company 25
- 26. © 2017 MIPI Alliance, Inc. Agenda • New specification and CTS changes in 2017 • Eye diagram tests changes and challenges – MIPI C-PHYSM – MIPI D-PHYSM – MIPI M-PHYÒ • SSC test – MIPI D-PHYSM 26
- 27. © 2017 MIPI Alliance, Inc. SSC (Spread Spectrum Clocking) - General SSC onSSC off Peak power : -6dBm = 112mV Peak power : -22dBm = 17.8mV § SSC makes distribution of RF power on signal so that it can reduce interruption to another signals like wireless signal. § Now MIPI D-PHY fundamental frequency is over 2GHz bandwidth, where lots of wireless signals have used. § Because of Clock line in MIPI D-PHY, SSC feature is required. Up to 4.5Gbps Up to 6.5Gbps Mobile Wireless bandwidth 27
- 31. © 2017 MIPI Alliance, Inc. Test Setup Tip 5 Source : Wikipedia 2nd order Butterworth filter = 40dB/decade w0 = cutoff frequency Use 2nd order Butterworth filter to meet SSC df/dt test condition § Some of Digital oscilloscope are possible to use Matlab code in the oscilloscope itself so that Matlab can apply complicated filter function. 31