Techniques for biasing lasers

Abstract

A method of biasing a laser that includes determining a threshold current of a laser and setting a bias current for the laser as a factor of the threshold current.

Description

TECHNICAL FIELD

[0002] This disclosure relates to biasing lasers.

BACKGROUND

[0003] Lasers, such as vertical-cavity surface-emitting lasers (VCSEL), Fabry-Perot (FP) lasers and distributed feedback (DFB) lasers, can have wide variations in their performance. An optical assembly, such as an optical transmitter or an optical transceiver, may require that the laser be tuned by setting the direct-current (DC) bias and alternating-current (AC). Characteristics of the lasers that may vary include the laser threshold current (Ith), slope efficiency, wavelength and over-temperature behavior. The variations in these characteristics may require a different biasing current depending on the desired response of the assembly.

SUMMARY OF THE DISCLOSURE

[0004] In one aspect, a method is disclosed for biasing a laser that includes determining a threshold current of a laser, and setting a bias current for the laser as a factor of the threshold current.

[0005] A second aspect is an article comprising a computer-readable medium storing computer-executable instructions that when applied to a computer system, cause the computer system to perform the method disclosed above.

[0006] Some implementations may include one or more of the following features. For example, the laser bias current may be set as a factor of the laser threshold current and/or the slope efficiency. Other implementations may include setting the laser bias current a predetermined amount above the laser threshold current.

[0007] Other implementations may include one or more of the following advantages. The techniques may provide an increase in yield for lasers installed in laser assemblies. Also, the lasers may be operated away from the performance limits of the laser. Operation of the lasers away from the performance limits may improve the average quality of the eye diagram and may increase the operating life of the laser.

[0008] Various features and advantages will be readily apparent from the following detailed description, the accompanying drawings and the claims.

BRIEF DESCRIPTION OF THE DRAWINGS

[0009] FIG. 1 is a block diagram of a system including an optical assembly;

[0010] FIG. 2 is an illustration of a laser response curve;

[0011] FIG. 3 is an illustration of biasing a family of laser response curves to achieve a target power output;

[0012] FIG. 4 is an illustration of biasing the family of laser response curves of FIG. 3 to achieve a power output within a predetermined range.

[0013] Like reference symbols in the various drawings indicate like elements.

DETAILED DESCRIPTION

[0014] FIG. 1 is a block diagram of a laser assembly 100 such as an optical transceiver or transmitter. A power source 102 is coupled to an optical assembly 104. The optical assembly includes circuitry 106 and a laser 108. Circuitry 106 may be used to bias, drive or otherwise provide current to the laser 108. Laser 108 can be made to emit laser light 110. A photo-detector 112 may be used to measure an output power of the emitted light 110.

[0015] FIG. 2 illustrates a generalized L-I curve 200 for an optical assembly. An L-I curve is a technique for characterizing the performance of a laser. In an L-I curve, the laser output power (L) can be plotted as a function of an electrical current (I) being passed through the optical assembly. For example, a current 212 through a laser characterized by L-I curve 200 produces an output power 214. The first region 202 of the curve is substantially flat. In this region the electrical current supplied to the laser does not result in a significant output of light from the optical assembly.

[0016] As the current through the laser increases, a current threshold (Ith) is reached where the laser begins to produce a significant output light power. The threshold current may be highly variable between lasers of the same type and even lasers from the same manufacturing lot.

[0017] At currents greater than Ith, there is a linear region 206 of the L-I curve where the laser light output is substantially proportional to the current applied to the laser. The proportionality constant may be called the slope efficiency (SE). In the linear region 206 of the L-I curve, the laser speed is proportional to ΔI/Ith where ΔI is the amount of current above the threshold current Ith. As with the threshold current Ith, the slope efficiency may vary greatly from laser to laser.

[0018] The L-I curve “rolls over” in a region 208, at current levels above the linear region 206. In the roll-over region 208 the light output is not directly proportional to the current applied. The roll-over region may also be characterized by undesirable laser characteristics including jitter (small variations in the emitted light waveform), shorter life, unstable power output, slower reaction time and distortion of the eye diagram (the viewing of multiple waveforms superimposed upon each other and may provide a subjective measure of the variations in the waveform, for example, jitter).

[0019] FIG. 3 illustrates three L-I curves 302, 304, 320 representing three lasers, L1, L2 and L3, respectively, which may have been selected from a lot of lasers received from a supplier. The lasers may be specified to operate so that a laser output power is in a desired range of values 324 between a lower limit and an upper limit. A target laser power within the output range may be determined. Lasers in the lot are biased to achieve the target laser power. Because L1, L2 and L3 may have different L-I curves, the biasing requirements of the lasers may be different to achieve a similar output power. In the illustrated example, laser L1 requires a smaller current 131 in order to achieve the same laser target output power as laser L2 which requires a current 132. Laser L3 may be required to operate in the roll-over region 322 operating at a current 133.

[0020] FIG. 4 illustrates a method of tuning the laser bias. The laser bias may be set after the laser is installed into the optical assembly. It is assumed that the lasers do not need to meet the same target output power. Instead, the bias of each laser is set to achieve a power that is within the optical assembly specification with a ΔI large enough for adequate laser speed and within the roll-off region.

[0021] In the example illustrated in FIG. 4, the laser power is set so that the optical assembly output power is within the desired output range 324 after the laser is installed in the optical assembly. In this example, laser L1 is biased at a current I1 which is ΔI1 above the L1 threshold current (Ith1). That current is large enough for improved laser speed compared to FIG. 3.

[0022] Referring again to FIG. 4, a laser having the L-I curve of L2 may be biased at a current 12 which is ΔI2 above the L2 threshold current (Ith2) to provide an output power within the desired output range. The optical assembly may be designed to source less current than if the bias were set to achieve the target output power illustrated in FIG. 3. The bias current I2 (FIG. 4) is lower than the bias current I2 (FIG. 3) corresponding to the target power output.

[0023] A laser having the L-I curve of L3 may be biased at a current I3 which is ΔI3 above the L3 threshold current (Ith3) to provide an output power within the desired output range. The laser can provide the desired output at an adequate speed without operating in the roll-over region 322 of the L-I curve 320.

[0024] The current ΔIx may be determined from the magnitude of the threshold current (Ithx) of a particular laser used in an optical assembly. Fiber optic transmitter performance may be characterized by key parameters including, but not limited to, jitter mask margin and power output. These parameters may be affected by a DC bias applied to the laser diode.

[0025] A preferred DC bias may be used to select a preferred jitter, mask margin performance and power requirement for a particular laser. The preferred DC bias setting bias setting may differ for each laser type—vertical-cavity surface-emitting laser (VCSEL), Fabry-Perot (FP) laser or a distributed feed-back (DFB) laser—and may vary even within the same laser type. The DC bias setting may also vary for the same laser type supplied by different vendors. The preferred DC bias laser setting for a particular laser may be selected as a derivative of the threshold current (Ith) of the particular laser.

[0026] In one implementation, the current ΔI can be set to a fixed current above the threshold current (Ith). This may be described mathematically as:

I opt =I th +x   (Eq. 1)

[0027] where Iopt is the preferred bias current and Ith is the threshold current for particular laser. Equation 1 describes a relationship where the preferred bias is a constant value above the threshold current of the laser. The value of “x” may be determined experimentally and may be substantially constant for lasers of a particular type from the same vendor.

[0028] In another implementation, the current ΔI can be set to a predetermined percentage above the threshold current (Ith). For example, it may be determined that a VCSEL used in a transceiver should have a current ΔI equal to 200% of the threshold current (Ith). Thus, a laser with a threshold current (Ith) of 2 milliamps (mA) would be biased to 4 mA in a transceiver assembly. A FP laser may require a current ΔI equal to 250% above threshold current (Ith). Thus a FP laser assembly might be biased to 5 mA. This may be described mathematically as:

I opt =I th *k   (Eq. 2)

[0029] Equation 2 describes a relationship where the preferred bias is a fixed multiple of the threshold current of the laser. The value of “k” may be determined experimentally and may be substantially constant for lasers of a particular type from the same lot.

[0030] In another implementation, the current ΔI may be determined from a combination of the threshold current (Ith) and slope efficiency (SE) of the laser. For example, the current ΔI can be set to a percentage of the threshold current (Ith) plus a percentage of the slope efficiency. Other combinations of the threshold current (Ith) and slope efficiency (SE) may be used as well.

I opt =(Ith*k)+(SE*y)  (Eq. 3)

I opt =(Ith*k)/(SE*y)  (Eq. 3b)

[0031] Equations 3a and 3b describe relationships where the preferred bias is related to the slope efficiency (SE) as well as to Ith. The value of where “k” and “y” may be determined experimentally and may be substantially constant for lasers of a particular type from the same vendor.

[0032] Other formulae may be determined that describe a preferred DC laser bias derived from the laser threshold current. As one example, the current ΔI may be determined from a combination of the threshold current (Ith), slope efficiency (SE) and the roll-over current of the L-I curve.

EXAMPLE #1

[0033] In one example, Eq. 1 was applied to Honeywell 850 nanometer (nm) implant VCSELs used in the manufacturing of Finisar Corporation's (Sunnyvale, Calif.) FTRJ-85XX-7D-XX product family of transceivers. In this example a value of “x” was determined as follows:

[0034] (1) A group of transceiver samples was selected from a production lot population. The samples were chosen to represent the entire population lot;

[0035] (2) Ith of the laser used in a first transceiver sample was determined;

[0036] (3) A DC bias current 3 mA above Ith was used to setup the laser in the first transceiver sample. For example, if Ith was 4 mA, the total current applied to the laser was 7 mA;

[0037] (4) The extinction ratio was adjusted to meet the specification of the transceiver;

[0038] (5) The jitter, mask margin, and power were measured and recorded;

[0039] (6) Items (3) through (5) were repeated using successively higher bias current levels above Ith (e.g., 4 mA, 5 mA, 6 mA, and so on) until a preferred laser bias current setting had been exceeded. An excessive bias current was determined as an increase in jitter beyond a predetermined acceptable limit, a decrease in mask margin below a predetermined acceptable value or a movement of output power from within to without a predetermined acceptable band;

[0040] (7) Items (3) through (6) were repeated with the remaining previously selected samples from the population lot, and a preferred current setting above Ith for each sample was measured and recorded; and

[0041] (8) The value of “x” in Eq. 1 to be used for the entire population of the transceivers was determined as an average of the preferred bias currents above Ith for all of the samples selected from the lot.

[0042] The data collected for the Honeywell 850 nm implant VCSEL is shown in Table 1. In this implementation, the laser bias was set using Eq. 1 with x=6.5 mA.

TABLE 1
Current above Ith		Mask	Power
(m)A	Jitter (ps)	Margin (%)	(dBm)
3 mA	53.3	38	−7.81
5 mA	35.6	47	−5.91
7 mA	37.8	45	−4.81
9 mA	51.1	32	−3.64

[0043] In a second implementation, the laser bias was set to 7.5 mA above Ith for the Mitsubishi 1310 FP laser used in Finisar Corporation's FTRJ-13XX-7D-XX transceiver products.

EXAMPLE #2

[0044] In this example, Eq. 2 may be applied for determining the bias current levels for Honeywell 850 nm oxide VCSELs.

[0045] (1) A group of transceiver samples was selected from a production lot population. The samples were chosen to represent the entire population lot;

[0046] (2) Ith of the laser used in a first transceiver sample was determined;

[0047] (3) Ith was multiplied by a “k” value of 2.0. The result was the bias current for setting up the laser. For example, if Ith was 3 mA, the total current applied to the laser was 6 mA;

[0048] (4) The extinction ratio was adjusted to meet specification;

[0049] (5) The jitter, mask margin, and power were measured and recorded;

[0050] (6) Items (3) through (5) were repeated using successively higher “k” values (e.g., 1.3, 1.4, 1.5, and so on) until a preferred laser bias current setting has been exceeded;

[0051] (7) Items (3) through (6) were repeated with the remaining previously selected samples from the population and the preferred “k” value for each such sample was measured and recorded.

[0052] (8) The value of “k” in Eq. 2 was determined as an average of the optimal “k” from the samples.

[0053] Various features of the system can be implemented in hardware, software, or a combination of hardware and software. For example, some aspects of the system can be implemented in computer programs executing on programmable computers. Each program can be implemented in a high level procedural or object-oriented programming language to communicate with a computer system.

[0054] Furthermore, each such computer program can be stored on a storage medium, such as read-only-memory (ROM), readable by a general or special purpose programmable computer, for configuring and operating the computer when the storage medium is read by the computer to perform the functions described above. Other implementations are within the scope of the following claims.

Claims

1. A method comprising: determining a threshold current of a laser; and setting a bias current for the laser as a factor of the threshold current.

2. The method of claim 1 comprising installing the laser in a laser assembly.

3. The method of claim 1 comprising installing the laser in an optical receiver or optical transceiver.

4. The method of claim 2 wherein setting the laser bias current further comprises: predetermining an output power range for the laser assembly; and setting the bias current to achieve an output power within the output power range.

5. The method of claim 1 wherein the factor is dependent on the type of laser installed in the laser assembly.

6. A method comprising: determining a threshold current of a laser; and setting a bias current for the laser to a predetermined value above the threshold current.

7. The method of claim 6 comprising installing the laser in a laser assembly.

8. The method of claim 6 comprising the laser in an optical receiver or transceiver.

9. The method of claim 7 wherein setting the laser bias current further comprises: predetermining an output power range for the laser assembly; and setting the bias current to achieve an output power within the output power range.

10. The method of claim 6 wherein the predetermined value is dependent on the type of laser installed in the laser assembly.

11. A method comprising: determining a threshold current of a laser; determining a slope efficiency for the laser; and setting a bias current for the laser as a factor of the threshold current and the slope efficiency.

12. The method of claim 11 comprising installing the laser in a laser assembly.

13. The method of claim 11 comprising installing the laser in an optical receiver or transceiver.

14. A method comprising: determining a threshold current of a laser; determining a slope efficiency of the laser; determining a bias current for the laser to cause the laser output power into a roll-off region; and setting a bias current for the laser based on the threshold current, the slope efficiency and the roll-off bias current.

15. The method of claim 14 comprising installing the laser in a laser assembly.

16. The method of claim 14 installing the laser in an optical receiver or transceiver.

17. An article comprising a computer-readable medium storing computer-executable instructions that, when applied to a computer system, cause the computer system to: determine a threshold current of a laser; and set a bias current for the laser as a factor of the threshold current for a laser installed in a laser assembly.

18. The article of claim 17 wherein the laser assembly is an optical receiver or transceiver.

19. The article of claim 17 wherein the instructions to set the laser bias current further comprise instructions to: predetermine an output power range for the laser assembly; and set the bias current to achieve an output power within the output power range.

20. The article of claim 17 wherein the factor is dependent on the type of laser installed in the laser assembly.

21. An article comprising a computer-readable medium storing computer-executable instructions that, when applied to a computer system, cause the computer system to: determine a threshold current of a laser; and set a bias current for the laser to a predetermined value above the threshold current.

22. The article of claim 21 wherein the instructions to set the laser bias current further comprise instructions to: predetermine an output power range for the laser assembly; and set the bias current to achieve an output power within the output power range.

23. The article of claim 21 wherein the predetermined value is dependent on the type of laser installed in the laser assembly.

24. An article comprising a computer-readable medium storing computer-executable instructions that, when applied a computer system, cause the computer system to: determine a threshold current of a laser; determine a slope efficiency for the laser; and set a bias current for the laser as a factor of the threshold current and the slope efficiency for a laser installed in a laser assembly.

25. The article of claim 24 wherein the laser assembly is an optical receiver or transceiver.

26. An article comprising a computer-readable medium storing computer-executable instructions that, when applied to a computer system, cause the computer system to: determine a threshold current of a laser; determine a slope efficiency of the laser; determine a bias current for the laser to cause the laser output power into a roll-off region; and set a bias current for the laser based on the threshold current, the slope efficiency and the roll-off bias current for a laser installed in a laser assembly.

27. The article of claim 26 wherein the laser assembly is an optical receiver or transceiver.