Method of Monitoring an Optoelectronic Transceiver with Multiple Flag Values for a Respective Operating Condition

Abstract

An optoelectronic transceiver includes an optoelectronic transmitter, an optoelectronic receiver, memory, and an interface. The memory is configured to store digital values representative of operating conditions of the optoelectronic transceiver. The interface is configured to receive from a host a request for data associated with a particular memory address, and respond to the host with a specific digital value of the digital values. The specific digital value is associated with the particular memory address received from the host. The optoelectronic transceiver may further include comparison logic configured to compare the digital values with limit values to generate flag values, wherein the flag values are stored as digital values in the memory.

Claims

1. A circuit for monitoring operation of an optoelectronic transceiver, said circuit comprising: a housing at least partially enclosing: a laser transmitter and a photodiode receiver: analog to digital conversion circuitry configured to convert a first analog signal corresponding to a first operating condition of said optoelectronic transceiver into a first digital value, and convert a second analog signal corresponding to a second operating condition of said optoelectronic transceiver into a second digital value; memory configured to store the first and second digital values in memory; and an interface configured to enable a host external to the optoelectronic transceiver to access the first and second digital values.

2. The circuit of claim 1, wherein the first digital value is selected from a group consisting of: a supply voltage measurement, a laser temperature measurement, a laser wavelength measurement, and a thermoelectric cooler current measurement.

3. The circuit of claim 1, wherein the first digital value includes a supply voltage measurement.

4. The circuit of claim 1, wherein the first digital value includes a laser temperature measurement.

5. The circuit of claim 1, wherein the first digital value includes a laser wavelength measurement.

6. The circuit of claim 1, wherein the first digital value includes a thermoelectric cooler current measurement.

7. The circuit of claim 1, further comprising a processor configured to store the first digital value in a first memory location and to store the second digital value in a second memory location.

8. The circuit of claim 7, wherein the processor is further configured to access the first digital value in the first memory location and to access the second digital value in the second memory location.

9. The circuit of claim 8, wherein the processor is further configured to access the first digital value in response to receiving a first address from the host and to access the second digital value in response to receiving a second address from the host.

10. The circuit of claim 9, wherein the processor is further configured to provide the digital values accessed by the processor to the host via the interface.

11. The circuit of claim 1, wherein the circuit further comprises control circuitry configured to generate control signals to control operation of the optoelectronic transceiver in accordance with one or more of the first digital value and the second digital value.

12. The circuit of claim 1, wherein the circuit further comprises operation disable circuitry configured to disable operation of at least part of the optoelectronic transceiver in response to a signal, wherein the signal is based on one or more of the first digital value and the second digital value.

13. The circuit of claim 1, wherein the circuit further comprises rate selection circuitry configured to set the photodiode receiver to a bandwidth in response to a signal, wherein the signal is based on one or more of the first digital value and the second digital value.

14. A circuit for monitoring operation of an optoelectronic transceiver, said circuit comprising: a housing at least partially enclosing: a laser transmitter and a photodiode receiver: analog to digital conversion circuitry configured to convert a first analog signal corresponding to a first operating condition of said optoelectronic transceiver into a first digital value, and convert a second analog signal corresponding to a second operating condition of said optoelectronic transceiver into a second digital value; memory configured to store the first and second digital values in memory; comparison logic configured to compare the first digital value with a first limit value to generate a first flag value and to compare the second digital value with a second limit value to generate a second flag value, wherein the first and second flag values i-s are stored in memory; and an interface configured to enable the host to access the first and second digital values and the first and second flag values.

15. The circuit of claim 14, further comprising a processor configured to store the first digital value in a first memory location, to store the second digital value in a second memory location, to store the first flag value in a first flag memory address, and to store the second flag value in a second flag memory address.

16. The circuit of claim 15, wherein the processor is further configured to access the first digital value in the first memory location, to access the second digital value in the second memory location, to access the first flag value in the first flag memory address, and to access the second flag value in the second flag memory address.

17. The circuit of claim 16, wherein the processor is further configured to access the first digital value in response to receiving a first address from the host, to access the second digital value in response to receiving a second address from the host, to access the first flag value in response to receiving a first flag memory address, and to access the second flag value in response to receiving a second flag memory address.

18. The circuit of claim 17, wherein the processor is further configured to provide the digital and flag values accessed by the processor to the host via the interface.

19. The circuit of claim 14, wherein the circuit further comprises control circuitry configured to generate control signals to control operation of the optoelectronic transceiver in accordance with one or more of the first flag value and the second flag value.

20. The circuit of claim 14, wherein the circuit further comprises operation disable circuitry configured to disable operation of at least part of the optoelectronic transceiver in response to a signal, wherein the signal is based on one or more of the first flag value and the second flag value.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Additional objects and features of the invention will be more readily apparent from the following detailed description and appended claims when taken in conjunction with the drawings, in which:

[0036] FIG. 1 is a block diagram of a prior art optoelectronic transceiver;

[0037] FIG. 2 is a block diagram of an optoelectronic transceiver in accordance with the present invention;

[0038] FIG. 3 is a block diagram of modules within the controller of the optoelectronic transceiver of FIG. 2;

[0039] FIG. 4 is a more detailed block diagram of the connections between the controller and the laser driver and post-amplifier;

[0040] FIG. 5A is a block diagram of a high-resolution alarm system and a fast trip alarm system, for monitoring and controlling the operation of the fiber optic transceiver to ensure eye safety;

[0041] FIG. 5B is a block diagram of logic for disabling the operation of the fiber optic transceiver to ensure eye safety;

[0042] FIG. 6 is a flow chart of a method for reducing or preventing potentially unsafe operation of a fiber optic transceiver using the fast trip alarm system of FIG. 5A; and

[0043] FIG. 7 is a flow chart of a method for reducing or preventing potentially unsafe operation of a fiber optic transceiver using the high-resolution alarm system of FIG. 5A.

[0044] Like reference numerals refer to corresponding parts throughout the several views of the drawings.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

[0045] A transceiver 100 based on the present invention is shown in FIGS. 2 and 3. The transceiver 100 contains a Receiver Optical Subassembly (ROSA) 102 and Transmitter Optical Subassembly (TOSA) 103 along with associated post-amplifier 104 and laser driver 105 integrated circuits that communicate the high speed electrical signals to the outside world. In this case, however, all other control and setup functions are implemented with a third single-chip integrated circuit 110 called the controller IC.

[0046] The controller IC 110 handles all low speed communications with the end user. These include the standardized pin functions such as Loss of Signal (LOS) 111, Transmitter Fault Indication (TX FAULT) 14, and the Transmitter Disable Input (TXDIS) 13. The controller IC 110 has a two wire serial interface 121, also called the memory interface, for accessing memory mapped locations in the controller. Memory Map Tables 1, 2, 3 and 4, below, are an exemplary memory map for one embodiment of a transceiver controller, as implemented in one embodiment of the present invention. It is noted that Memory Map Tables 1, 2, 3 and 4, in addition to showing a memory map of values and control features described in this document, also show a number of parameters and control mechanisms that are outside the scope of this document and thus are not part of the present invention.

[0047] The interface 121 is coupled to host device interface input/output lines, typically clock (SCL) and data (SDA) lines, 15 and 16. In the preferred embodiment, the serial interface 121 operates in accordance with the two wire serial interface standard that is also used in the GBIC and SFP standards, however other serial interfaces could equally well be used in alternate embodiments. The two wire serial interface 121 is used for all setup and querying of the controller IC 110, and enables access to the optoelectronic transceiver's control circuitry as a memory mapped device. That is, tables and parameters are set up by writing values to predefined memory locations of one or more nonvolatile memory devices 120, 122, 128 (e.g., EEPROM devices) in the controller, whereas diagnostic and other output and status values are output by reading predetermined memory locations of the same nonvolatile memory devices 120, 122, 128. This technique is consistent with currently defined serial ID functionality of many transceivers where a two wire serial interface is used to read out identification and capability data stored in EEPROM.

[0048] It is noted here that some of the memory locations in the memory devices 120, 122, 128 are dual ported, or even triple ported in some instances. That is, while these memory mapped locations can be read and in some cases written via the serial interface 121, they are also directly accessed by other circuitry in the controller 110. For instance, certain margining values stored in memory 120 are read and used directly by logic 134 to adjust (i.e., scale upwards or downwards) drive level signals being sent to the D/A output devices 123. Similarly, there are flags stored in memory 128 that are (A) written by logic circuit 131, and (B) read directly by logic circuit 133. An example of a memory mapped location not in memory devices but that is effectively dual ported is the output or result register of clock 132. In this case the accumulated time value in the register is readable via the serial interface 121, but is written by circuitry in the clock circuit 132.

[0049] In addition to the result register of the clock 132, other memory mapped locations in the controller may be implemented as registers at the input or output of respective sub-circuits of the controller. For instance, the margining values used to control the operation of logic 134 may be stored in registers in or near logic 134 instead of being stored within memory device 128. In another example, measurement values generated by the ADC 127 may be stored in registers. The memory interface 121 is configured to enable the memory interface to access each of these registers whenever the memory interface receives a command to access the data stored at the corresponding predefined memory mapped location. In such embodiments, locations within the memory include memory mapped registers throughout the controller.

[0050] In an alternate embodiment, the time value in the result register of the clock 132, or a value corresponding to that time value, is periodically stored in a memory location with the memory 128 (e.g., this may be done once per minute, or once per hour of device operation). In this alternate embodiment, the time value read by the host device via interface 121 is the last time value stored into the memory 128, as opposed to the current time value in the result register of the clock 132.

[0051] As shown in FIGS. 2 and 3, the controller IC 110 has connections to the laser driver 105 and receiver components. These connections serve multiple functions. The controller IC has a multiplicity of D/A converters 123. In the preferred embodiment the D/A converters are implemented as current sources, but in other embodiments the D/A converters may be implemented using voltage sources, and in yet other embodiments the D/A converters may be implemented using digital potentiometers. In the preferred embodiment, the output signals of the D/A converters are used to control key parameters of the laser driver circuit 105. In one embodiment, outputs of the D/A converters 123 are use to directly control the laser bias current as well as to control the level of AC modulation to the laser (constant bias operation). In another embodiment, the outputs of the D/A converters 123 of the controller 110 control the level of average output power of the laser driver 105 in addition to the AC modulation level (constant power operation).

[0052] In a preferred embodiment, the controller 110 includes mechanisms to compensate for temperature dependent characteristics of the laser. This is implemented in the controller 110 through the use of temperature lookup tables 122 that are used to assign values to the control outputs as a function of the temperature measured by a temperature sensor 125 within the controller IC 110. In alternate embodiments, the controller 110 may use D/A converters with voltage source outputs or may even replace one or more of the D/A converters 123 with digital potentiometers to control the characteristics of the laser driver 105. It should also be noted that while FIG. 2 refers to a system where the laser driver 105 is specifically designed to accept inputs from the controller 110, it is possible to use the controller IC 110 with many other laser driver ICs to control their output characteristics.

[0053] In addition to temperature dependent analog output controls, the controller IC may be equipped with a multiplicity of temperature independent (one memory set value) analog outputs. These temperature independent outputs serve numerous functions, but one particularly interesting application is as a fine adjustment to other settings of the laser driver 105 or postamp 104 in order to compensate for process induced variations in the characteristics of those devices. One example of this might be the output swing of the receiver postamp 104. Normally such a parameter would be fixed at design time to a desired value through the use of a set resistor. It often turns out, however, that normal process variations associated with the fabrication of the postamp integrated circuit 104 induce undesirable variations in the resulting output swing with a fixed set resistor. Using the present invention, an analog output of the controller IC 110, produced by an additional D/A converter 123, is used to adjust or compensate the output swing setting at manufacturing setup time on a part-by-part basis.

[0054] In addition to the connection from the controller to the laser driver 105, FIG. 2 shows a number of connections from the laser driver 105 to the controller IC 110, as well as similar connections from the ROSA 106 and Postamp 104 to the controller IC 110. These are analog monitoring connections that the controller IC 110 uses to provide diagnostic feedback to the host device via memory mapped locations in the controller IC. The controller IC 110 in the preferred embodiment has a multiplicity of analog inputs. The analog input signals indicate operating conditions of the transceiver and/or receiver circuitry. These analog signals are scanned by a multiplexer 124 and converted using an analog to digital converter (ADC) 127. The ADC 127 has 12 bit resolution in the preferred embodiment, although ADC's with other resolution levels may be used in other embodiments. The converted values are stored in predefined memory locations, for instance in the diagnostic value and flag storage device 128 shown in FIG. 3, and are accessible to the host device via memory reads. These values are calibrated to standard units (such as millivolts or microwatts) as part of a factory calibration procedure.

[0055] The digitized quantities stored in memory mapped locations within the controller IC include, but are not limited to, the laser bias current, transmitted laser power, and received power (as measured by the photodiode detector in the ROSA 102). In the memory map tables (e.g., Table 1), the measured laser bias current is denoted as parameter B.sub.in, the measured transmitted laser power is denoted as P.sub.in, and the measured received power is denoted as R.sub.in. The memory map tables indicate the memory locations where, in an exemplary implementation, these measured values are stored, and also show where the corresponding limit values, flag values, and configuration values (e.g., for indicating the polarity of the flags) are stored.

[0056] As shown in FIG. 3, the controller 110 includes a voltage supply sensor 126. An analog voltage level signal generated by this sensor is converted to a digital voltage level signal by the ADC 127, and the digital voltage level signal is stored in memory 128. In a preferred embodiment, the A/D input multiplexer (mux) 124 and ADC 127 are controlled by a clock signal so as to automatically, periodically convert the monitored signals into digital signals, and to store those digital values in memory 128.

[0057] Furthermore, as the digital values are generated, the value comparison logic 131 of the controller compares these values to predefined limit values. The limit values are preferably stored in memory 128 at the factory, but the host device may overwrite the originally programmed limit values with new limit values. Each monitored signal is automatically compared with both a lower limit and upper limit value, resulting in the generation of two limit flag values that are then stored in the diagnostic value and flag storage device 128. For any monitored signals where there is no meaningful upper or lower limit, the corresponding limit value can be set to a value that will never cause the corresponding flag to be set.

[0058] The limit flags are also sometimes call alarm and warning flags. The host device (or end user) can monitor these flags to determine whether conditions exist that are likely to have caused a transceiver link to fail (alarm flags) or whether conditions exist which predict that a failure is likely to occur soon. Examples of such conditions might be a laser bias current which has fallen to zero, which is indicative of an immediate failure of the transmitter output, or a laser bias current in a constant power mode which exceeds its nominal value by more than 50%, which is an indication of a laser end-of-life condition. Thus, the automatically generated limit flags are useful because they provide a simple pass-fail decision on the transceiver functionality based on internally stored limit values.

[0059] In a preferred embodiment, fault control and logic circuit 133 logically OR's the alarm and warning flags, along with the internal LOS (loss of signal) input and Fault Input signals, to produce a binary Transceiver fault (TxFault) signal that is coupled to the host interface, and thus made available to the host device. The host device can be programmed to monitor the TxFault signal, and to respond to an assertion of the TxFault signal by automatically reading all the alarm and warning flags in the transceiver, as well as the corresponding monitored signals, so as to determine the cause of the alarm or warning.

[0060] The fault control and logic circuit 133 furthermore conveys a loss of signal (LOS) signal received from the receiver circuit (ROSA, FIG. 2) to the host interface.

[0061] Yet another function of the fault control and logic circuit 133 is to determine the polarity of its input and output signals in accordance with a set of configuration flags stored in memory 128. For instance, the Loss of Signal (LOS) output of circuit 133 may be either a logic low or logic high signal, as determined by a corresponding configuration flag stored in memory 128.

[0062] Other configuration flags (see Table 4) stored in memory 128 are used to determine the polarity of each of the warning and alarm flags. Yet other configuration values stored in memory 128 are used to determine the scaling applied by the ADC 127 when converting each of the monitored analog signals into digital values.

[0063] In an alternate embodiment, another input to the controller 102, at the host interface, is a rate selection signal. In FIG. 3 the rate selection signal is input to logic 133. This host generated signal would typically be a digital signal that specifies the expected data rate of data to be received by the receiver (ROSA 102). For instance, the rate selection signal might have two values, representing high and low data rates (e.g., 2.5 Gb/s and 1.25 Gb/s). The controller responds to the rate selection signal by generating control signals to set the analog receiver circuitry to a bandwidth corresponding to the value specified by the rate selection signal.

[0064] Another function of the fault control and logic circuit 133 is to disable the operation of the transmitter (TOSA, FIG. 2) when needed to ensure eye safety. There is a standards defined interaction between the state of the laser driver and an internal Tx Disable output, which is implemented by the fault control and logic circuit 133. When the logic circuit 133 detects a problem that might result in an eye safety hazard, the laser driver is preferably disabled by activating an internal Tx Disable signal output from the controller, as described in further detail below. The host device can reset this condition by sending a command signal on the external Tx Disable line 13 (FIG. 2) into the controller from the host. Further details of this functionality can be found below in relation to FIGS. 4-7.

[0065] FIG. 4 is a more detailed block diagram of the connections between the controller 110 (FIG. 2) and the laser driver 105 and post-amplifier 104. Optical signals received by the optical receiver in the ROSA 102 are transmitted along a received power connection 402 to the postamp 104. The postamp 104 generates a fixed output swing digital signal which is connected to the host, and/or controller 110 (FIG. 2), via RX+ and RX connections 404. The postamp circuit preferably also provides a Loss of Signal (LOS) indicator to the host, and/or controller 110 (FIG. 2), via a LOS connection 406, indicating the presence or absence of suitably strong optical input.

[0066] The host transmits signal inputs TX+ and TX to the laser driver 105 via TX+ and TX connections 420. In addition, the controller 110 (FIG. 2) transmits power to the laser driver via connection 416, and a transmitter disable signal to the laser driver 105 via an internal TX disable connection 418.

[0067] As a laser 410 within the TOSA is not turned on and off, but rather modulated between high and low levels above a threshold current, a modulation current is supplied to the laser 410 via an AC modulation current connection 414. Furthermore, a DC laser bias current is supplied from the laser driver 105 to the laser 410 via a laser bias current connection 412. The level of the laser bias current is adjusted to maintain proper laser output (i.e., to maintain a specified or predefined average level of optical output power by the TOSA 103) and to compensate for variations in temperature and power supply voltage.

[0068] In addition, some transceivers include an output power monitor 422 within the TOSA 103 that monitors the energy output from the laser 410. The output power monitor 422 is preferably a photodiode within the laser package that measures light emitted from the back facet of the laser 410. In general, the amount of optical power produced by the back facet of the laser diode, represented by an output power signal, is directly proportional to the optical power output by the front or main facet of the laser 410. The ratio, K, of the back facet optical power to the front facet optical power will vary from one laser diode to another, even among laser diodes of the same type. The output power signal is transmitted from the output power monitor 422 in the TOSA 103 to the controller 110 (FIG. 2) via a transmitter output power connection 408.

[0069] In a preferred embodiment, certain of the components within the fiber optic transceiver include monitoring logic that outputs digital fault conditions. For example, the laser driver 105 may output a out of lock signal 424 if a control loop monitoring the modulation current is broken. These digital fault condition outputs may then be used to notify the host of fault conditions within the component, or shut down the laser.

[0070] FIG. 5A is a block diagram 500 of a high-resolution alarm system 502 and a fast trip alarm system 504, for monitoring and controlling the operation of the fiber optic transceiver to ensure eye safety. The fast trip alarm system 504 is used to quickly generate flag used to shut down the laser 410 (FIG. 4). The fast trip alarm system 504 uses an analog comparator 522 to achieve a quick response. The high resolution alarm system 502 does not generate a flag to shut down the laser as quick as the fast trip alarm system 504. However, the high resolution alarm system 502 is more accurate than the fast trip alarm system 504. To achieve this accuracy, the high resolution alarm system 502 uses digital comparators 512. In use, the high resolution alarm system 502 and the fast trip alarm system 504 operate simultaneously. If the fast trip alarm system 504 does not generate a flag quickly, the high resolution alarm system 502 will identify the fault and generate a flag to shut down the laser.

[0071] The high-resolution alarm system 502 and fast trip alarm system 504 are preferably contained within the controller 110 (FIG. 3). Both the high-resolution alarm system 502 and fast trip alarm system 504 are coupled to an input signal 506. In a preferred embodiment this input signal is an analog signal. It should be noted that FIG. 5A shows the high-resolution alarm system 502 and fast trip alarm system 504 for a single input signal 506. However, in a preferred embodiment, identical alarm systems 502 and 504 are provided for each of several signals 506, including several different types of input signals.

[0072] The input signals processed by the alarm systems 502 and 504 preferably include: power supply voltage, internal transceiver temperature (hereinafter temperature), laser bias current, transmitter output power, and received optical power. The power supply voltage 19 (FIG. 3) is preferably the voltage in millivolts as measured by the Vcc sensor 126 (FIG. 3). The temperature is preferably the temperature in C. as measured by the temperature sensor 125 (FIG. 3). The laser bias current is preferably the laser bias current in microamps supplied to the laser 410 (FIG. 4) via the laser bias current connection 412 (FIG. 4). The received optical power is the power in microwatts received at the ROSA 102 (FIG. 4) via the received power connection 402 (FIG. 4). Finally, the optical output power (FIG. 4) is the optical power output in microwatts, from the power monitor 422 (FIG. 4) as received by the controller 110 (FIG. 2) via the output power connection 408 (FIG. 4).

[0073] The high-resolution alarm system 502 preferably utilizes all of the above described input signals to trigger warnings and/or shut down at least part of the fiber optic transceiver. In other embodiments the high-resolution alarm system 502 utilizes a subset of the above described input signals to trigger warnings and/or alarms. The high-resolution alarm system 502 includes one or more analog to digital converters 124 (see also FIG. 3) that are configured to receive the analog input signal 506. Each type of analog input signal is preferably converted to a digital input signal using a calibration factor 508 for the particular type of input signal received. For example, a supply voltage in millivolts is converted to a 16 bit digital number by multiplying a supply voltage millivolt value by a supply voltage calibration factor. These calibration factors are predetermined and are preferably stored in the diagnostic value and flag storage 128 (FIG. 3). Alternatively, such calibration factors 508 may be stored in the general purpose EEPROM 120 (FIG. 3).

[0074] The analog to digital converter 124 is also coupled to multiple comparators 512. In a preferred embodiment, the comparators 512 form a portion of the value comparison and other logic 131 (FIG. 3) in the controller 110 (FIG. 2). In a preferred embodiment, these comparators 512 are digital comparators.

[0075] Also coupled to the comparators 512 are high-resolution setpoints 510(1)-(N). In a preferred embodiment, four predetermined setpoints 510(1)-(4) (for each type of input signal 506) are stored in the diagnostic value and flag storage 128 (FIG. 3). These four predetermined setpoints are: a high alarm setpoint 510(1), a high warning setpoint 510(2), a low warning setpoint 510(3), and a low alarm setpoint 510(4). The comparators 512(1)-(N) are configured to compare the input signal 506 with the predetermined setpoints 510(1)-(4). In a preferred embodiment, the digital equivalent of the input signal 506 is simultaneously compared by the comparators 512(1)-(N), to each of the four digital predetermined setpoints 510(1)-(N) for the particular type of input signal received. Also in a preferred embodiment, the setpoints 510(1)-(N) and the digital equivalents to the input signals 506 are preferably sixteen bit numbers. Of course, in other embodiments there may be more or fewer setpoints 510, and the setpoints 510 and input signal could be digitally represented by more or fewer than sixteen bits.

[0076] The comparators subsequently generate high-resolution flags 514(1)-(N), which are input into the general logic and fault control circuit 133 (FIG. 3) to either provide a warning to the host computer, or to shut down at least part of the fiber optic transceiver, such as the laser driver 105 (FIG. 4) and/or laser 410 (FIG. 4). Further details of the method for preventing potentially unsafe operation of the fiber optic transceiver, using the high-resolution alarm system 502, are described below in relation to FIG. 7.

[0077] The fast trip alarm system 504 includes multiple temperature dependant setpoints 516. These temperature dependant setpoints 516 are preferably stored in the diagnostic values flag storage 128 (FIG. 3) or the D/A temperature lookup tables 122 (FIG. 3). A multiplexer 518 is configured to supply one of the temperature dependant setpoints 516 to a digital to analog converter 123 (also shown in FIG. 3). The precise temperature dependant setpoint 516 that is supplied depends on the temperature 520 measured by the temperature sensor 125 (FIG. 3). For example, for a first measured temperature, a first setpoint is supplied by the multiplexer 518 to the digital to analog converter 123.

[0078] A separate copy or instance of the fast trip alarm system 504 is provided for each input signal 506 for which a temperature based alarm check is performed. Unlike the high-resolution alarm system 502, the fast trip alarm system 504 preferably utilizes only the following input signals 506: laser bias current, transmitter output power, and received optical power input signals, and thus in the preferred embodiment there are three instances of the fast trip alarm system 504. In other embodiment, fewer or more fast trip alarm systems 504 may be employed. The analog input signals processed by the fast trip alarm systems 504 are each fed to a respective comparator 522 that compares the input signal to an analog equivalent of one of the temperature dependant setpoints 516. In a preferred embodiment, the comparators 522 form a portion of the value comparison and other logic 131 (FIG. 3) in the controller 110 (FIG. 2). In a preferred embodiment, the comparators 522 are analog comparators.

[0079] In a preferred embodiment at least eight temperature dependant setpoints 516 are provided for the laser bias current input signal, with each setpoint corresponding to a distinct 16 C. temperature range. The size of the operating temperature range for each setpoint may be larger or smaller in other embodiments. These temperature dependant setpoints for the laser bias current are crucial because of the temperature compensation needs of a short wavelength module. In particular, at low temperatures the bias required to produce the required light output is much lower than at higher temperatures. In fact, a typical laser bias current when the fiber optic transceiver is at the high end of its temperature operating range will be two or three times as high as the laser bias current when fiber optic transceiver is at the low end of its temperature operating range, and thus the setpoints vary dramatically based on operating temperature. A typical temperature operating range of a fiber optic transceiver is about 40 C. to about 85 C. The temperature dependant setpoints for the laser bias current are also crucial because of the behavior of the laser bias circuit in a fiber optic transceiver that transmits long wavelength energy.

[0080] Also in a preferred embodiment, at least four temperature dependant setpoints 516 are provided for the received optical power and transceiver output power input signals, with each setpoint corresponding to a distinct 32 C. operating temperature range of the fiber optic transceiver. The size of the operating temperature range for each setpoint may be larger or smaller in other embodiments.

[0081] In a preferred embodiment, the above mentioned setpoints 516 are 8 bit numbers, which scale directly to the pin (Bin, Pin, Rin) input voltages at (2.5V(max)/256 counts)=0.0098 volts/count.

[0082] The comparator 522 is configured to compare an analog equivalent of one of the setpoints 516 to the analog input signal 506. In a preferred embodiment, if the analog input signal 506 is larger than the analog equivalent to one of the setpoints 516, then a fast trip alarm flag 524 is generated. The fast trip alarm flag 524 is input into the general logic and fault control circuit 133 (FIG. 3) to either provide a warning to the host computer or shut down at least part of the fiber optic transceiver, such as the laser driver 105 (FIG. 4) and/or laser 410 (FIG. 4). Further details of the method for preventing potentially unsafe operation of the fiber optic transceiver, using the fast trip alarm system 504, are described below in relation to FIG. 6.

[0083] FIG. 5B is a block diagram of logic 530 for disabling the operation of the fiber optic transceiver to ensure eye safety, according to a preferred embodiment of the invention. The high-resolution alarm flags 514(1)-(4), the fast trip alarm system flag 524, and any digital fault condition 532 signals, from FIGS. 4 and 5A, are transmitted to an OR gate 534, which is used to shut down the laser. This is accomplished by sending a signal along the internal Tx disable line 418 (FIG. 4). For example if a digital out of lock signal or a fast trip alarm flag is received, the laser will be shut down. It should be appreciated that more or less alarm flags or digital fault condition signals may be supplied to the OR gate 534. For instance, in one preferred embodiment, the inputs to the OR gate 534 include only the high and low alarm flags 514(1), 514(4), the fast trip alarm flag 524 and the digital fault condition(s) signal 532. In other words, in this preferred embodiment, the warning flags 514(2) and 514(3) are not used to generate the internal Tx disable signal 418.

[0084] FIG. 6 is a flow chart of a method 600 for reducing or preventing potentially unsafe operation of a fiber optic transceiver using the fast trip alarm system 504 of FIG. 5A. Once the fast trip alarm system 504 (FIG. 5A) has started at step 602, an input signal is acquired, at step 604. In a preferred embodiment, the input signal is preferably an analog signal of: laser bias current in milliamps, received optical power in microwatts, or transceiver output power in microwatts. A temperature of the fiber optic transceiver is obtained at step 606. Step 606 may be performed before, after or at the same time as input signal acquisition step 604.

[0085] The multiplexer 518 (FIG. 5A) uses the input signal and the measured temperature to determine, at step 608, which setpoint 516 (FIG. 5A) to use for comparison with the input signal. For example, if the input signal is laser bias current, then the multiplexer looks up a setpoint for laser bias current based on the obtained temperature 520 (FIG. 5A).

[0086] In a preferred embodiment, this setpoint is then converted from a digital to analog value, at step 610 by the digital to analog converter 123 (FIG. 5A). Thereafter, the comparator 522 (FIG. 5A) compares the input signal to the setpoint, at step 612, to determine whether there is a conflict, at step 614. In a preferred embodiment, a conflict occurs where the input signal is higher than the setpoint (or an analog equivalent of the setpoint). Alternatively, a conflict may occur where the input signal is lower than the setpoint (or an analog equivalent of the setpoint).

[0087] If no conflict exists (614No), then the method 600 repeats itself. However, if a conflict does exist (614Yes), then a fast trip alarm flag 524 (FIG. 5A) is generated at step 616. In a preferred embodiment the fast trip alarm flag 524 (FIG. 5A) is then used to shut down at least part of the fiber optic transceiver, at step 618, by applying a signal to the internal TxDisable connection 418 (FIG. 4). In a preferred embodiment the fast trip alarm flag 524 (FIG. 5A) is used to disable the laser driver 105 (FIG. 4) and/or laser 410 (FIG. 4), so that no potential eye-damage can occur.

[0088] The alarm flag 524 (FIG. 5A) can be used to control the laser driver via the internal Tx Disable Output (Dout) and signal the fault to the host system via the Tx Fault Output (Fout). These outputs can also respond to the Tx Fault Input (Fin), if that signal exists in any given implementation, and the Tx Disable Input (Din) which comes into the fiber optic transceiver from the host.

[0089] FIG. 7 is a flow chart of a method 700 for reducing or preventing potentially unsafe operation of a fiber optic transceiver using the high-resolution alarm system 502 of FIG. 5A. Once the high-resolution alarm system 502 (FIG. 5A) has started at step 702, an input signal is acquired, at step 704. In a preferred embodiment, the input signal is preferably an analog signal of: power supply voltage 19 (FIG. 3) in millivolts; the temperature in C.; the laser bias current 412 (FIG. 4) in microamps; the received optical power 420 (FIG. 4) in microwatts; and the output power 408 (FIG. 4) in microwatts. In other embodiments, the input signal(s) may be scaled in accordance with other units.

[0090] An analog to digital converter 124 (FIGS. 3 and 5) then converts the analog input signal 506 (FIG. 5A) to a digital equivalent, preferably a 16 bit number, at step 706. Conversion of the analog input signal 506 (FIG. 5A) to a digital equivalent performed includes multiplying the input signal 506 (FIG. 5A) by a calibration factor 508 (FIG. 5A), at step 708, for the particular type of input signal received, as described above in relation to FIG. 5A.

[0091] The comparators 512 (FIG. 5A) then compare the digital equivalent of the input signal to the setpoints 510(1)-(N) (FIG. 5A), at step 710, to determine whether there is a conflict. In a preferred embodiment, conflicts occur when the digital equivalent of the input signal is: higher than the high alarm setpoint 510(1) to produce a high-alarm flag 514(1) (FIG. 5A); higher than the high warning setpoint 510(2) (FIG. 5A) to produce a high warning flag 514(2) (FIG. 5A); lower than a low warning flag 510(3) (FIG. 5A) to produce a low warning flag 514(3) (FIG. 5A); or lower than a low alarm flag 510(4) (FIG. 5A) to produce a low alarm flag 514(4) (FIG. 5A). It should, however, be appreciated that other types of alarms or warnings may be set.

[0092] If no conflict exists (712No), then the method 700 repeats itself. However, if a conflict does exist (714Yes), then a high-resolution flag 514(1)-(N) (FIG. 5A) is generated, at step 714. In a preferred embodiment, the high-resolution flags 514(1)-(N) (FIG. 5A) are a high alarm flag 514(1), a high warning flag 514(2), a low warning flag 514(3), and a low alarm flag 514(4), as shown in FIG. 5A. Also in a preferred embodiment, the high alarm flag 514(1) (FIG. 5A) and the low alarm flag 514(4) are used to shut down at least part of the fiber optic transceiver, at step 716, by applying a signal to the internal TxDisable connection 418 (FIG. 4). The part of the fiber optic transceiver shut down preferably includes the laser driver 105 (FIG. 4) and/or the TOSA 103 (FIG. 4). The high and low warning flags 514(2) and 514(3) (FIG. 5A) preferably merely provide a warning to the host and do not shut down the laser driver 105 (FIG. 4) and/or the TOSA 103 (FIG. 4).

[0093] The alarm flags 514(1)-(N) (FIG. 5A) can be used to control the laser driver via the internal Tx Disable Output (Dout) and signal the fault to the host system via the Tx Fault Output (Fout). These outputs can also respond to the Tx Fault Input (Fin), if that signal exists in any given implementation, and the Tx Disable Input (Din) which comes into the fiber optic transceiver from the host.

[0094] In a preferred embodiment, the high-resolution alarm system 502 (FIG. 5A) updates the high-resolution alarm flags at a rate of approximately once every 0.015 seconds (15 milliseconds), and more generally at least 50 times per second. Thus, the high-resolution alarm flags are set within 0.015 seconds of the detection of an alarm condition. In some embodiments the high-resolution alarm flag update rate is between about 50 times per second and 200 times per second. However, the fast trip alarm system 504 (FIG. 5A) preferably updates the fast trip alarm flags a rate that is faster than once every 10 microseconds. In some embodiments the fast trip alarm system 504 updates the fast trip alarm flags at a rate that is between 50,000 and 200,000 times per second, and more generally at least 50,000 times per second. In a preferred embodiment, the alarm flags of the fast trip alarm system 504 are updated at a rate that is more than a thousand times faster than the update rate of the high-resolution alarm flags. In other embodiments the alarm flags of the fast trip alarm system 504 are updated at a rate that is between 250 and 4000 times faster than the update rate of the high-resolution alarm flags.

[0095] To further aid the above explanation, two examples are presented below, where a single point failure causes an eye safety fault condition that is detected, reported to a host coupled to the fiber optic transceiver, and/or a laser shutdown is performed.

Example 1

[0096] The power monitor 422 (FIG. 4) in a fiber optic transceiver that includes a power monitor, or its associated circuitry, fails, indicating no or low output power when the laser is in fact operating. The laser bias driver will attempt to increase the transmitter output power by increasing laser bias current. Since the feedback is interrupted, the laser is driven to its maximum capability, perhaps exceeding the eye safety alarm setpoints. The fast trip alarm flag will be generated in less than 10 microseconds after the failure and this fast trip alarm flag can be used to shut down the laser driver via the internal Tx disable (Dout) output. If the fast trip alarm fails or is not selected in the output logic setup, the high-resolution alarm for laser bias current is generated, and the high-resolution low alarm for power would also occur, either of which could be used to shut down the laser driver and/or TOSA.

Example 2

[0097] The laser driver (in all types of fiber optic transceiver), or its associated circuitry fails, driving the laser to its maximum output. Depending on the specific failure, the laser bias current may read zero or very high, and in a fiber optic transceiver that includes a power monitor, the power will read very high. The fast trip alarm for laser bias current, and the fast trip alarm for transmitted output power will generate an alarm flag within 10 microseconds. If the laser bias current is reading zero, the high-resolution low alarm for laser bias current will generate an alarm flag. This may be indistinguishable from a failure that causes zero light output, like an open laser wire or shorted laser, but the alarm systems preferably err on the side of safety and command the laser to shut down. In this condition, it may not be possible for the logic to physically turn the laser off, if, for example, the fault was caused by a shorted bias driver transistor. In any case, the link will be lost and the Tx fault output will be asserted to advise the host system of the failure. Depending on the configuration of the bias driver circuit, there are non-error conditions which could set some of these flags during a host-commanded transmit disable state, or during startup conditions. For example, if the host commands a transmitter shutdown, some circuits might read zero transmit power, as one would expect, and some might read very large transmit power as an artifact of the shutdown mechanism. When the laser is re-enabled, it takes a period of time for the control circuitry to stabilize, and during this time there may be erratic occurrences of both low, high and fast trip alarms. Programmable delay timers are preferably used to suppress the fault conditions during this time period.

[0098] While the combination of all of the above functions is desired in the preferred embodiment of this transceiver controller, it should be obvious to one skilled in the art that a device which only implements a subset of these functions would also be of great use. Similarly, the present invention is also applicable to transmitters and receivers, and thus is not solely applicable to transceivers. Finally, it should be pointed out that the controller of the present invention is suitable for application in multichannel optical links.

TABLE-US-00001 TABLE 1 MEMORY MAP FOR TRANSCEIVER CONTROLLER Name of Location Function Memory Location (Array 0) 00h-5Fh IEEE Data This memory block is used to store required GBIC data 60h Temperature MSB This byte contains the MSB of the 15-bit 2's complement temperature output from the temperature sensor. 61h Temperature LSB This byte contains the LSB of the 15-bit 2's complement temperature output from the temperature sensor. (LSB is 0b). 62h-63h V.sub.cc Value These bytes contain the MSB (62h) and the LSB (63h) of the measured V.sub.cc (15-bit number, with a 0b LSbit) 64h-65h B.sub.in Value These bytes contain the MSB (64h) and the LSB (65h) of the measured B.sub.in (15-bit number, with a 0b LSbit) 66h-67h P.sub.in Value These bytes contain the MSB (66h) and the LSB (67h) of the measured P.sub.in (15-bit number, with a 0b LSbit) 68h-69h R.sub.in Value These bytes contain the MSB (68h) and the LSB (69h) of the measured R.sub.in (15-bit number, with a 0b LSbit) 6Ah-6Dh Reserved Reserved 6Eh IO States This byte shows the logical value of the I/O pins. 6Fh A/D Updated Allows the user to verify if an update from the A/D has occurred to the 5 values: temperature, Vcc, B.sub.in, P.sub.in, and R.sub.in. The user writes the byte to 00h. Once a conversion is complete for a give value, its bit will change to 1. 70h-73h Alarm Flags These bits reflect the state of the alarms as a conversion updates. High alarm bits are 1 if converted value is greater than corresponding high limit. Low alarm bits are 1 if converted value is less than corresponding low limit. Otherwise, bits are 0b. 74h-77h Warning Flags These bits reflect the state of the warnings as a conversion updates. High warning bits are 1 if converted value is greater than corresponding high limit. Low warning bits are 1 if converted value is less than corresponding low limit. Otherwise, bits are 0b. 78h-7Ah Reserved Reserved 7Bh-7Eh Password Entry Bytes The four bytes are used for password entry. PWE Byte 3 (7Bh) The entered password will determine the MSByte user's read/write privileges. PWE Byte 2 (7Ch) PWE Byte 1 (7Dh) PWE Byte 0 (7Eh) LSByte 7Fh Array Select Writing to this byte determines which of the upper pages of memory is selected for reading and writing. 0xh (Array x Selected) Where x = 1, 2, 3, 4 or 5 80h-F7h Customer EEPROM 87h DA % Adj Scale output of D/A converters by specified percentage Memory Location (Array 0) 00h-FFh Data EEPROM Memory Location (Array 0) 00h-Ffh Data EEPROM Memory Location (Array 3) 80h-81h Temperature High The value written to this location serves as 88h-89h Alarm the high alarm limit. Data format is the 90h-91h Vcc High Alarm same as the corresponding value 98h-99h B.sub.in High Alarm (temperature, Vcc, B.sub.in, P.sub.in, R.sub.in). A0h-A1h P.sub.in High Alarm R.sub.in High Alarm 82h-83h Temperature Low The value written to this location serves as 8Ah-8Bh Alarm the low alarm limit. Data format is the same 92h-93h Vcc Low Alarm as the corresponding value (temperature, 9Ah-9Bh B.sub.in Low Alarm Vcc, B.sub.in, P.sub.in, R.sub.in). A2h-A3h P.sub.in Low Alarm R.sub.in Low Alarm 84h-85h Temp High Warning The value written to this location serves as 8Ch-8Dh Vcc High Warning the high warning limit. Data format is the 94h-95h B.sub.in High Warning same as the corresponding value 9Ch-9Dh P.sub.in High Warning (temperature, Vcc, B.sub.in, P.sub.in, R.sub.in). A4h-A5h R.sub.in High Warning 86h-87h Temperature Low The value written to this location serves as 8Eh-8Fh Warning the low warning limit. Data format is the 96h-97h Vcc Low Warning same as the corresponding value 9Eh-9Fh B.sub.in Low Warning (temperature, Vcc, B.sub.in, P.sub.in, R.sub.in). A6h-A7h P.sub.in Low Warning R.sub.in Low Warning A8h-AFh D.sub.out control 0-8 Individual bit locations are defined in Table C5h F.sub.out control 0-8 4. B0h-B7h, C6h L.sub.out control 0-8 B8h-BFh, C7h C0h Reserved Reserved C1h Prescale Selects MCLK divisor for X-delay CLKS. C2h D.sub.out Delay Selects number of prescale clocks C3h F.sub.out Delay C4h L.sub.out Delay C8h-C9h Vcc-A/D Scale 16 bits of gain adjustment for corresponding CAh-CBh B.sub.in-A/D Scale A/D conversion values. CCh-CDh P.sub.in-A/D Scale CEh-CFh R.sub.in-A/D Scale D0h Chip Address Selects chip address when external pin ASEL is low. D1h Margin #2 Finisar Selective Percentage (FSP) for D/A #2 D2h Margin #1 Finisar Selective Percentage (FSP) for D/A #1 D3h-D6h PW1 Byte 3 (D3h) The four bytes are used for password 1 MSB entry. The entered password will determine PW1 Byte 2 (D4h) the Finisar customer's read/write privileges. PW1 Byte 1 (D5h) PW1 Byte 0 (D6h) LSB D7h D/A Control This byte determines if the D/A outputs source or sink current, and it allows for the outputs to be scaled. D8h-DFh B.sub.in Fast Trip These bytes define the fast trip comparison over temperature. E0h-E3h P.sub.in Fast Trip These bytes define the fast trip comparison over temperature. E4h-E7h R.sub.in Fast Trip These bytes define the fast trip comparison over temperature. E8h Configuration Override Location of the bits is defined in Table 4 Byte E9h Reserved Reserved EAh-EBh Internal State Bytes Location of the bits is defined in Table 4 ECh I/O States 1 Location of the bits is defined in Table 4 EDh-EEh D/A Out Magnitude of the temperature compensated D/A outputs EFh Temperature Index Address pointer to the look-up Arrays F0h-FFh Reserved Reserved Memory Location (Array 4) 00h-Ffh D/A Current vs. Temp #1 (User-Defined Look-up Array #1) Memory Location (Array 5) 00h-Ffh D/A Current vs. Temp #2 (User-Defined Look-up Array #2)

TABLE-US-00002 TABLE 2 DETAIL MEMORY DESCRIPTIONS-A/D VALUES AND STATUS BITS Byte Bit Name Description Converted analog values. Calibrated 16 bit data. (See Notes 1-2) 96 All Temperature MSB Signed 2's complement integer temperature (60h) (40 to +125 C.) Based on internal temperature measurement 97 All Temperature LSB Fractional part of temperature (count/256) 98 All Vcc MSB Internally measured supply voltage in transceiver. Actual voltage is full 16 bit value * 100 uVolt. 99 All Vcc LSB (Yields range of 0-6.55 V) 100 All TX Bias MSB Measured TX Bias Current in mA Bias current is full 16 bit value *(1/256) mA. 101 All TX Bias LSB (Full range of 0-256 mA possible with 4 uA resolution) 102 All TX Power MSB Measured TX output power in mW. Output is full 16 bit value *(1/2048) mW. (see note 5) 103 All TX Power LSB (Full range of 0-32 mW possible with 0.5 W resolution, or 33 to +15 dBm) 104 All RX Power MSB Measured RX input power in mW RX power is full 16 bit value *(1/16384) mW. (see note 6) 105 All RX Power LSB (Full range of 0-4 mW possible with 0.06 W resolution, or 42 to +6 dBm) 106 All Reserved MSB Reserved for 1.sup.st future definition of digitized analog input 107 All Reserved LSB Reserved for 1.sup.st future definition of digitized analog input 108 All Reserved MSB Reserved for 2.sup.nd future definition of digitized analog input 109 All Reserved LSB Reserved for 2.sup.nd future definition of digitized analog input 110 7 TX Disable Digital state of the TX Disable Input Pin 110 6 Reserved 110 5 Reserved 110 4 Rate Select Digital state of the SFP Rate Select Input Pin 110 3 Reserved 110 2 TX Fault Digital state of the TX Fault Output Pin 110 1 LOS Digital state of the LOS Output Pin 110 0 Power-On-Logic Indicates transceiver has achieved power up and data valid 111 7 Temp A/D Valid Indicates A/D value in Bytes 96/97 is valid 111 6 Vcc A/D Valid Indicates A/D value in Bytes 98/99 is valid 111 5 TX Bias A/D Valid Indicates A/D value in Bytes 100/101 is valid 111 4 TX Power A/D Valid Indicates A/D value in Bytes 102/103 is valid 111 3 RX Power A/D Valid Indicates A/D value in Bytes 104/105 is valid 111 2 Reserved Indicates A/D value in Bytes 106/107 is valid 111 1 Reserved Indicates A/D value in Bytes 108/109 is valid 111 0 Reserved Reserved

TABLE-US-00003 TABLE 3 DETAIL MEMORY DESCRIPTIONS-ALARM AND WARNING FLAG BITS Alarm and Warning Flag Bits Byte Bit Name Description 112 7 Temp High Alarm Set when internal temperature exceeds high alarm level. 112 6 Temp Low Alarm Set when internal temperature is below low alarm level. 112 5 Vcc High Alarm Set when internal supply voltage exceeds high alarm level. 112 4 Vcc Low Alarm Set when internal supply voltage is below low alarm level. 112 3 TX Bias High Alarm Set when TX Bias current exceeds high alarm level. 112 2 TX Bias Low Alarm Set when TX Bias current is below low alarm level. 112 1 TX Power High Alarm Set when TX output power exceeds high alarm level. 112 0 TX Power Low Alarm Set when TX output power is below low alarm level. 113 7 RX Power High Alarm Set when Received Power exceeds high alarm level. 113 6 RX Power Low Alarm Set when Received Power is below low alarm level. 113 5-0 Reserved Alarm 114 All Reserved 115 All Reserved 116 7 Temp High Warning Set when internal temperature exceeds high warning level. 116 6 Temp Low Warning Set when internal temperature is below low warning level. 116 5 Vcc High Warning Set when internal supply voltage exceeds high warning level. 116 4 Vcc Low Warning Set when internal supply voltage is below low warning level. 116 3 TX Bias High Warning Set when TX Bias current exceeds high warning level. 116 2 TX Bias Low Warning Set when TX Bias current is below low warning level. 116 1 TX Power High Set when TX output power exceeds high Warning warning level. 116 0 TX Power Low Set when TX output power is below low Warning warning level. 117 7 RX Power High Set when Received Power exceeds high Warning warning level. 117 6 RX Power Low Set when Received Power is below low Warning warning level. 117 5 Reserved Warning 117 4 Reserved Warning 117 3 Reserved Warning 117 2 Reserved Warning 117 1 Reserved Warning 117 0 Reserved Warning 118 All Reserved 119 All Reserved

TABLE-US-00004 TABLE 4 Byte Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 X-out cntl0 T alrm hi T alrm lo V alrm hi V alrm lo B alrm hi B alrm lo P alrm hi P alrm lo set set set set set set set set X-out cntl1 R alrm hi R alrm lo B ft hi set P ft hi set R ft hi set D-in inv D-in set F-in inv set set set set X-out cntl2 F-in set L-in inv L-in set Aux inv Aux set T alrm hi T alrm lo V alrm hi set set hib hib hib X-out cntl3 V alrm lo B alrm hi B alrm lo P alrm hi P alrm lo R alrm hi R alrm lo B ft hi hib hib hib hib hib hib hib hib X-out cntl4 P ft hi hib R ft hi hib D-in inv D-in hib F-in inv F-in hib L-in inv L-in hib hib hib hib X-out cntl5 Aux inv Aux hib T alrm hi T alrm lo V alrm hi V alrm lo B alrm hi B alrm lo hib clr clr clr clr clr clr X-out cntl6 P alrm hi P alrm lo R alrm hi R alrm lo B ft hi clr P ft hi clr R ft hi clr D-in inv clr clr clr clr clr X-out cntl7 D-in clr F-in inv F-in clr L-in inv L-in clr Aux inv Aux clr EE clr clr clr X-out cntl8 latch invert o-ride data o-ride S reset HI enable LO enable Pullup select select data enable Prescale reserved reserved Reserved reserved B.sup.3 B.sup.2 B.sup.1 B.sup.0 X-out delay B.sup.7 B.sup.6 B.sup.5 B.sup.4 B.sup.3 B.sup.2 B.sup.1 B.sup.0 chip address b.sup.7 b.sup.6 b.sup.5 b.sup.4 b.sup.3 b.sup.2 b.sup.1 X X-ad scale 2.sup.15 2.sup.14 2.sup.13 2.sup.12 2.sup.11 2.sup.10 2.sup.9 2.sup.8 MSB X-ad scale 2.sup.7 2.sup.6 2.sup.5 2.sup.4 2.sup.3 2.sup.2 2.sup.1 2.sup.0 LSB D/A cntl source/sink D/A #2 range source/sink D/A #1 range 1/0 2.sup.2 2.sup.1 2.sup.0 1/0 2.sup.2 2.sup.1 2.sup.0 config/O- manual manual manual EE Bar SW-POR A/D Manual reserved ride D/A index AD alarm Enable fast alarm Internal D-set D-inhibit D-delay D-clear F-set F-inhibit F-delay F-clear State 1 Internal L-set L-inhibit L-delay L-clear reserved reserved reserved reserved State 0 I/O States 1 reserved F-in L-in reserved D-out reserved reserved reserved Margin #1 Reserved Neg_ Neg_ Neg_ Reserved Pos_Scale Pos_Scale Pos_Scale Scale2 Scale1 Scale0 2 1 0 Margin #2 Reserved Neg_ Neg_ Neg_ Reserved Pos_Scale Pos_Scale Pos_Scale Scale2 Scale1 Scale0 2 1 0