Optical Receiver Sensitivity Rainer Schwemmer LHCb Electronics Meeting
Optical Receiver Sensitivity Rainer Schwemmer, LHCb Electronics Meeting Dec. 2014
Goal • Determine the spread of sensitivity of different MINIPod receivers – For the Detector DAQ direction • Sample of 20 x 12 receivers from Marseille – 240 receivers in total – ~ 2 -3% of planned installation
Methodology • Unfortunately no x 12 tester for sensitivity measurements available • The receivers need to go onto the new AMC 40 prototypes limited time frame • Fixed transmitter with stable optical power output, change receivers and try to map resulting BER to sensitivity
Methodology 1. 00 E-02 1. 00 E-03 1. 00 E-04 1. 00 E-05 BER 1. 00 E-06 1. 00 E-07 1. 00 E-08 1. 00 E-09 1. 00 E-10 Channel 1 Channel 2 • Assumptions: 1. 00 E-11 1. 00 E-12 Optical Receive Power – Curve for different receivers has same shape but is moved to the left or to the right depending on sensitivity – Transmitter output power is stable between on/off cycles
Methodology Fixed Optical Power at receiver input 1. 00 E-02 1. 00 E-03 Measured BER spread 1. 00 E-04 1. 00 E-05 BER 1. 00 E-06 1. 00 E-07 1. 00 E-08 1. 00 E-09 1. 00 E-10 Channel 1 Channel 2 1. 00 E-11 1. 00 E-12 Optical Receive Power Sens. difference • Inject optical signal at fixed power • Measure BER of different receivers at this power
Methodology Fixed Optical Power at receiver input 1. 00 E-02 1. 00 E-03 Measured BER spread 1. 00 E-04 1. 00 E-05 BER 1. 00 E-06 1. 00 E-07 1. 00 E-08 1. 00 E-09 1. 00 E-10 Channel 1 Channel 2 1. 00 E-11 1. 00 E-12 Optical Receive Power Sens. difference • If the two curves have the same shape then the two triangles are identical • Use the BER difference measured on a reference receiver to get the power/sensitivity value of the device under test
Setup • One x 12 Transmitter of one prototype AMC 40 as source – Attenuated to intensity which actually produces errors • Swap out receivers on the board and measure BER Prizm Connector Attenuator Source Receiver AMC 40 Frontplate Connectors
Setup • One x 12 Transmitter of one prototype AMC 40 as source – Attenuated to intensity which actually produces errors • Swap out receivers on the board and measure BER • Have to power cycle the board for swapping out receivers though Prizm Connector Attenuator Source AMC 40 Frontplate Connectors Receiver 1 Receiver 2
Setup Transmitter Receiver Attenuator
Results • Temperature dependency 10 min
Results • Bit error rate vs Channel ID • Individual channels form clusters – Injected optical power is different on different channels – Attenuator is not same on every channel
Results • Every now and then: put the same receiver back into the socket • Variation is quite large • Prism connector seems to have quite a large spread on re-connects
Results • After conversion of BER into optical power • Distribution is centered around average • No absolute values yet, because we have not measured absolute value of optical output power
Conclusions • Spread of sensitivity is quite large (+- 1 d. B) • A large portion of it is probably due to losses in connector rather than receiver • Need to do calibration of transmitter to get absolute sensitivity values (2 -3 days of measurements in optics lab)
- Slides: 14