AD1845 FUNCTIONAL DESCRIPTION
This section overviews the functionality of the AD1845 and is intended as a general introduction to the capabilities of the device. As much as possible, detailed reference information has been placed in “Control Registers” and other sections. The user is not expected to refer repeatedly to this section.
Stereo digital output from the ADCs can be mixed digitally with the input to the DACs. Digital output from the ADCs going out of the data port is unaffected by the digital mix. Along the digital mix datapath, the 16-bit linear output from the ADCs is attenuated by an amount specified with control bits. Both channels of the digital mix datapath are attenuated by the same amount. (Note that internally the AD1845 always works with 16-bit PCM linear data, digital mixing included; format conversions take place at the input and output.)
The AD1845 SoundPort Stereo Codec accepts stereo line-level and microphone-level inputs. The LINE, MIC, AUX1, and post-mixed DAC output are available to the ADC multiplexer. The DAC output can be mixed with LINE, MIC, AUX1, AUX2 and M_IN. Each channel of the MIC inputs can be amplified by +20 dB to compensate for the difference between line levels and typical condenser microphone levels.
Sixty-four steps of –1.5 dB attenuation are supported to –94.5 dB. The digital mix datapath can also be completely muted. Note that the level of the mixed signal is also a function of the input PGA settings, since they affect the ADCs’ output. The attenuated digital mix data is digitally summed with the DAC input data prior to the DACs’ datapath attenuators. The digital sum of digital mix data and DAC input data is clipped at plus or minus full scale and does not wrap around. Because both stereo signals are mixed before the output attenuators, mix data is attenuated a second time by the DACs’ datapath attenuators.
The M_IN mono input signal, MIC, LINE, AUX1 and AUX2 analog stereo signals can be mixed in the analog domain with the DAC output. Each channel of each AUX, LINE and MIC analog input can be independently gained/attenuated from +12 dB to –34.5 dB in 1.5 dB steps or completely muted. M_IN can be attenuated from 0 dB to –45 dB in 3 dB steps or muted. The post-mixed DAC outputs are available on L_OUT and R_OUT and also to the ADC input multiplexer. Even if the AD1845 is not playing back data from its DACs, the analog mix function can still be active. Analog-to-Digital Datapath
The PGA following the input multiplexer allows independent selectable gains for each channel from 0 dB to 22.5 dB in +1.5 dB steps. The codec can operate either in a global stereo mode or in a global mono mode with left-channel inputs appearing at both channel outputs.
In case the AD1845 is capturing data but ADC output data is not removed in time (“ADC overrun”), then the last sample captured before overrun will be used for the digital mix. In case the AD1845 is playing back data but input digital DAC data fails to arrive in time (“DAC underrun”), then a midscale zero will be added to the digital mix data when the DACZ control bit is set to 0, otherwise, the DAC will output the previous valid sample in an underrun condition. Analog Outputs
Stereo and mono line-level outputs are available at external pins. Each channel of this output can be independently muted. When muted, the outputs will settle to a dc value near VREF, the midscale reference voltage. The output is selectable for 2.0 V peak-to-peak or 2.8 V peak-to-peak. When selecting the LINE output as an input to the ADC, the ADC automatically compensates for the output level selection.
The AD1845 ∑∆ ADCs incorporate a fourth-order modulator. A single pole of passive filtering is all that is required for antialiasing the analog input because of the ADC’s high over sampling ratio. The ADCs include linear-phase digital decimation filters that low-pass filter the input to 0.4 × FS. (“FS” is the word rate or “sampling frequency.”) ADC input over range conditions will cause bits to be set that can be read.
Digital Data Types
The AD1845 supports four global data types: 16-bit twos complement linear PCM, 8-bit unsigned linear PCM, companded µ-law, and 8-bit companded A-law, as specified by control register bits. Data in all four formats is always transferred MSB first. All data formats that are less than 16-bits are properly aligned to insure the use of full system resolution.
The ∑∆ DACs are preceded by a programmable attenuator and a low-pass digital interpolation filter. The antimaging interpolation filter over samples and digitally filters the higher frequency images. The attenuator allows independent control of each DAC channel from 0 dB to –94.5 dB in –1.5 dB steps plus full mute. The DACs’ ∑∆ noise shapers also over sample and convert the signal to a single-bit stream. The DAC outputs are then filtered in the analog domain by a combination of switchedcapacitor and continuous-time filters. They remove the very high frequency components of the DAC bit stream output. No external components are required.
The 16-bit PCM data format is capable of representing 96 dB of dynamic range. Eight-bit PCM can represent 48 dB of dynamic range. Companded µ-law and A-law data formats use nonlinear coding with less precision for large-amplitude signals. The loss of precision is compensated for by an increase in dynamic range to 64 dB and 72 dB, respectively.
Changes in DAC output attenuation take effect only on zero crossings, eliminating “zipper” noise on playback. Each channel has its own independent zero-crossing detector and attenuator change control circuitry. A timer guarantees that requested volume changes will occur even in the absence of a zero crossing. The time-out period is 8 milliseconds at a 48 kHz sampling rate and 48 milliseconds at an 8 kHz sampling rate. (Time-out [ms] ≈ 384 ÷ FS [kHz].)
On input, 8-bit companded data is expanded to an internal linear representation, according to whether µ-law or A-law was specified in the codec’s internal registers. Note that when µ-law compressed data is expanded to a linear format, it requires 14 bits. A-law data expanded requires 13 bits.