Cirrus Logic CS4234 - Manual (Page 76)

DS899F1 76
CS4234
11.1 Voltage Headroom
Headroom is another word for the static DC offset inserted into the tracking signal. This offset allows the rail
voltages to track the audio amplifier outputs with a sufficient amount of voltage such that a sudden audio
bandwidth transition (i.e. 20 kHz or less) across the full dynamic range of the audio amplifier (i.e from 0 V
input to full scale input) will not cause the amplifier to clip. This control is contained within the DC Offset
block highlighted in yellow in
Figure 47 .
Adjustment to the DC offset is done through the TPS OFFSET[2:0] bits in the "TPS Control" register . The
offset is given in terms of %Full Scale, since the output voltage of the DACs are given relative to VA.
The DC offset applied to the tracking signal affects both the efficiency of the Class H amplifier and the ability
of the SMPS to respond to high amplitude transients. By having a smaller DC Offset, the power that is wast-
ed in the output stage of the audio amplifier is reduced. However, since the speed at which the SMPS rail
voltages change depends heavily on the bulk capacitance attached to its outputs, providing too small of a
DC offset may not provide enough head room for the SMPS to maintain an unclipped audio signal during
the time that the rail voltage transitions are charging the bulk capacitance.
Also, unless the resting voltage of the SMPS is set by some other means within the SMPS, the lowest pos-
sible DC offset value is dictated by the minimum operating threshold of the audio amplifier attached to the
rails of the SMPS. The DC offset, multiplied by the gain of the SMPS “K2”, cannot be less than the amplifier’s
operating threshold. If this occurs, the undervoltage lockout protection of the audio amplifier (if equipped)
engages prematurely when the rails collapse to a level lower than the allowable operating threshold.
11.2 Lead Time
Lead time is a static time interval that allows the SMPS rails to begin to transition before the amplified audio
output signal begins to transition. This is accomplished by delaying the outgoing audio signal sufficiently to
prevent the amplified output signal from transitioning faster than the rails can transition. This control is pro-
vided by the programmable Group Delay block highlighted in red in
Figure 47 .
As was the case with the DC offset, there is a connection between the group delay applied to the audio sig-
nal, the efficiency of the Class H amplifier, and the ability of the SMPS to respond to high amplitude tran-
sients. By having a shorter delay, the output voltage of the SMPS can track the audio signal more closely,
increasing efficiency of the Class H system by reducing wasted power in the amplifier output devices. How-
ever, providing too short of a delay may not provide enough lead time for the SMPS to react to high frequen-
cy, high-amplitude transients, which will result in clipping of the output signal. For these reasons, the group
delay of the CS4234 can be adjusted from 100
s to 500 s, based on the GROUP DELAY[3:0] setting.
11.3 Gain Matching
Gain matching is necessary to ensure that the gain of the path of the SMPS path is the same as the audio
path. The SMPS path consists of the DAC5 path, the SMPS modulator, and the voltage conversion ratio of
the SMPS and any components between DAC5 and the SMPS modulator. The audio path consists of
DAC1-4 path and the audio amplifier and any components between them. The gains present in each of the
blocks of interest are shown in
Figure 47 as K1 through K5. The gain matching block “K1”, highlighted in
violet, provides the means to ensure that the total gain of the SMPS path, which consists of K1*K4(if appli-
cable)*K2, is as close as possible to the audio path gain, comprised of K3*K4*K5. If the “K3” channel gain
is not equal for all channels, use the maximum channel’s gain during this matching calculation to prevent
clipping on any channel.
As much as +6 dB of gain and as much as -83 dB of attenuation can be applied to the tracking signal in
approximately 0.4 dB steps, although a much smaller range centered around 0 dB is likely. To match the
gains between the SMPS and the amplifier, calculate the gain of the audio path and add gain or attenuation
as necessary to the SMPS path to make the gains of both paths equal. The addition of gain or attenuation
is accomplished via the
DAC5 VOLUME[7:0] bits.