The MAX11043 features 4 single-ended or differential
channels of simultaneous-sampling ADCs with 16-bit
resolution. The MAX11043 contains a versatile filter
block and programmable-gain amplifier (PGA) per
channel. The filter consists of seven cascaded 2ndorder filter sections for each channel, allowing the construction of a 14th-order filter. The filter coefficients are
user-programmable. Configure each 2nd-order filter as
lowpass (LP), highpass (HP), or bandpass (BP) with
optional rectification. Gain and phase mismatch of the
analog signal path is better than -50dB.
The ADC can sample up to 800ksps per channel. A
40MHz serial interface provides communication to and
from the device. The SPI™ interface provides throughput of 1600ksps; 4 channels at 400ksps per channel or
2 channels at 800ksps per channel. A software-selectable scan mode allows reading the ADC results while
simultaneously updating the DAC. Other features of the
MAX11043 include an internal (+2.5V) or external
(+2.0V to +2.8V) reference, power-saving modes, and
a PGA with gains of 1 to 64. The PGA includes an
equalizer (EQ) function that automatically boosts lowamplitude, high-frequency signals for applications such
as CW-chirp radar.
The MAX11043 includes two 8-bit coarse DACs that set
the high and low references for a second-stage 12-bit
fine DAC, typically used for VCO control. Use software
controls to write to the DAC or step the DAC up and
down under hardware control in programmable steps.
The device operates from a +3.0V to +3.6V supply. The
MAX11043 is available in a 40-pin, 6mm x 6mm TQFN
package and operates over the extended -40°C to
+125°C temperature range.
OSCIN), clock divider set to 4, SHDN = DACSTEP = UP/DWN = DGND, CONVRUN = DVDD, all analog inputs driven directly through
a series 150Ω/330pF anti-alias filter, PGA gain = 1. Default filters and gain settings. T
A
= T
MIN
to T
MAX
, unless otherwise noted (Note 1).
Typical values are at T
A
= +25°C.)
Stresses beyond those listed under “Absolute Maximum Ratings” may cause permanent damage to the device. These are stress ratings only, and functional
operation of the device at these or any other conditions beyond those indicated in the operational sections of the specifications is not implied. Exposure to
absolute maximum rating conditions for extended periods may affect device reliability.
AVDD to AGND ....................................................-0.3V to +4.0V
DVDD to DGND .....................................................-0.3V to +4.0V
DVREG to DGND...................................................-0.3V to +3.0V
AGND to DGND.....................................................-0.3V to +0.3V
Analog I/O, REFDACH, REFDACL, REFA, REFB, REFC, REFD,
AOUT, REFDAC, REFBP to AGND......-0.3V to (AVDD + 0.3V)
UP/DWN, CONVRUN, SHDN, DACSTEP, EOC, Digital I/O,
OSCIN, OSCOUT to DGND ....................-0.3V to (DVDD + 0.3V)
Maximum Current into Any Pin except AVDD, DVDD, DVREG,
OSCIN), clock divider set to 4, SHDN = DACSTEP = UP/DWN = DGND, CONVRUN = DVDD, all analog inputs driven directly through
a series 150Ω/330pF anti-alias filter, PGA gain = 1. Default filters and gain settings. T
A
= T
MIN
to T
MAX
, unless otherwise noted (Note 1).
Typical values are at T
A
= +25°C.)
DYNAMIC PERFORMANCE (PGA Enabled, PGA Gain = 8 x (25kHz -1dB Full-Scale Signal))
Maximum Full-Scale InputADC modulator gain = 1150mV
Input-Referred Noise Spectral
Density
Second Harmonic to
Fundamental
Third Harmonic to Fundamental-94dB
Spurious-Free Dynamic RangeSFDR100dB
Channel-to-Channel Isolation
Channel Phase Matching
DYNAMIC PERFORMANCE (PGA Enabled, PGA Gain = 16 x (25kHz -1dB Full-Scale Signal))
Maximum Full-Scale InputADC modulator gain = 175mV
OSCIN), clock divider set to 4, SHDN = DACSTEP = UP/DWN = DGND, CONVRUN = DVDD, all analog inputs driven directly through
a series 150Ω/330pF anti-alias filter, PGA gain = 1. Default filters and gain settings. T
A
= T
MIN
to T
MAX
, unless otherwise noted (Note 1).
Typical values are at T
A
= +25°C.)
Channel-to-Channel Isolation
Channel Phase Matching
DYNAMIC PERFORMANCE (All Modes)
Conversion Rate
Minimum Throughput5ksps
Power-Supply Rejection RatioDCPSRR50dB
ANALOG INPUTS (AINAP/AINAN, AINBP/AINBN, AINCP/AINCN, AINDP/AINDN)
Absolute Voltage Any Input(Note 4)0AVDDV
Input Impedance (Note 5)
Input CapacitanceEQ mode only50pF
EQ FILTER (Analog and Digital)
Unity-Gain FrequencyDefault5kHz
Lower Transition FrequencyDefault, from 40dB/decade to 0dB/decade190kHz
Upper Transition FrequencyDefault, from 0dB/decade to -80dB/decade205kHz
LP FILTER
-3dB Corner FrequencyDefault205kHz
REFERENCE INPUT
REF_ Input Voltage RangeV
Input Current150µA
REFBP Input Voltage RangeV
Input Current700µA
REFDAC Input Voltage RangeV
Input Resistance17kΩ
PARAMETERSYMBOLCONDITIONSMINTYPMAXUNITS
REFBP
REFDAC
REF_
Unused channels are shorted and
unconnected
Between all channels, including complete
analog signal path
OSCIN), clock divider set to 4, SHDN = DACSTEP = UP/DWN = DGND, CONVRUN = DVDD, all analog inputs driven directly through
a series 150Ω/330pF anti-alias filter, PGA gain = 1. Default filters and gain settings. T
A
= T
MIN
to T
MAX
, unless otherwise noted (Note 1).
Typical values are at T
A
= +25°C.)
PARAMETERSYMBOLCONDITIONSMINTYPMAXUNITS
REFDAC_ Input Voltage RangeV
REFDAC_
01.4V
Input Resistance150kΩ
INTERNAL REFERENCE
Reference VoltageV
Reference Temperature
Coefficient
REFBP
2.452.52.55V
100ppm/°C
CRYSTAL OSCILLATOR (Max ESR 100Ω, 22pF Load Capacitors to DGND)
Maximum Crystal Operating
Frequency
External Clock Input Frequency
Range
Epson Electronics MA-505 (16MHz)16MHz
External clock applied to OSCIN440MHz
StabilityExcluding crystal25ppm
Startup TimeEpson Electronics MA-505 (16MHz)10ms
OSCIN Input Low VoltageWhen driven with external clock source
OSCIN Input High VoltageWhen driven with external clock source
0.7 x
DVDD
0.3 x
DVDD
OSCIN Leakage Current-5+5µA
DIGITAL INPUTS
Input High VoltageV
Input Low VoltageV
IH
IL
0.7 x
DVDD
0.3 x
DVDD
Input Hysterisis15mV
Input Leakage CurrentI
Input CapacitanceC
IN
V
= 0 or DVDD-1+1µA
IN
IN
15pF
DIGITAL OUTPUTS
DVDD
-
Output-Voltage HighV
Output-Voltage LowV
OH
OL
I
I
= 0.8mA
SOURCE
= 1.6mA0.4V
SINK
0.6
Three-State Leakage CurrentDOUT only-1+1µA
Three-State Output CapacitanceDOUT only15pF
VOLTAGE REGULATOR
Regulated Digital Supply VoltageDV
REG
Internal use only2.5V
POWER REQUIREMENTS
Analog Supply Voltage3.03.6V
Digital Supply Voltage3.03.6V
V
V
V
V
V
MAX11043
4-Channel, 16-Bit, Simultaneous-Sampling ADCs
with PGA, Filter, and 8-/12-Bit Dual-Stage DAC
OSCIN), clock divider set to 4, SHDN = DACSTEP = UP/DWN = DGND, CONVRUN = DVDD, all analog inputs driven directly through
a series 150Ω/330pF anti-alias filter, PGA gain = 1. Default filters and gain settings. T
Note 1: Devices 100% production tested at TA= +125°C. Guaranteed by design and characterization to TA= -40°C.
Note 2: Full scale in analog EQ mode decreases with increasing frequency at a rate of 20dB/decade from 5kHz. If digital EQ is also
used, full scale decreases with increasing frequency at 40dB/decade from 5kHz.
Note 3: SFDR in the EQ mode is normalized to the input by subtracting the analog EQ gain at each frequency (20dB/decade) from
the FFT results.
Note 4: The absolute input voltage range is 0 to AVDD. For optimal performance, use a common-mode voltage of AVDD/2.
Note 5: Switched capacitor input impedance is proportional to 1/fC. Where f is the sampling frequency and C is the input capacitance.
ELECTRICAL CHARACTERISTICS (continued)
((V
AVDD
= +3.0V to +3.6V, V
DVDD
= +3.0V, C
DVREG
= 10µF, V
AGND
= V
DGND
= 0, common-mode input voltage = AVDD/2, V
REFBP
=
V
REFA
= V
REFB
= V
REFC
= V
REFD
= +2.5V (external reference), V
REFDAC
= V
REFDACH
= +1.25V (external reference), V
REFDACL
= 0,
C
REFBP
= C
REFA
= C
REFB
= C
REFC
= C
REFD
= C
REFDAC
= 1µF, f
SCLK
= 38.4MHz, f
EXCLK
= 38.4MHz (external clock applied to
OSCIN), clock divider set to 4, SHDN = DACSTEP = UP/DWN = DGND, CONVRUN = DVDD, all analog inputs driven directly through
a series 150Ω/330pF anti-alias filter, PGA gain = 1. Default filters and gain settings. T
A
= T
MIN
to T
MAX
, unless otherwise noted (Note 1).
Typical values are at T
A
= +25°C.)
Typical Operating Characteristics
(V
AVDD
= +3.3V, V
DVDD
= +3.0V, f
SCLK
= f
EXCLK
= 19.2MHz, V
REFBP
, V
REF_
= +2.5V, common-mode input voltage = AVDD/2,
TA = +25°C, unless otherwise noted.)
INL vs. CODE
MAX11043 toc01
CODE (LSB)
INL (LSB)
491523276816384
-4
-3
-2
-1
0
1
2
3
4
5
-5
065536
LP MODE
GAIN = 1
400ksps FFT
LP MODE
MAX11043 toc02
FREQUENCY (kHz)
AMPLITUDE (dBFS)
18016014012010080604020
-100
-80
-60
-40
-20
0
-120
0200
fIN = 50kHz
GAIN = 1
800ksps FFT
LP MODE
MAX11043 toc03
FREQUENCY (kHz)
AMPLITUDE (dBFS)
35030025020015010050
-120
-100
-80
-60
-40
-20
0
-140
0400
fIN = 50kHz
GAIN = 1
PARAMETERSYMBOLCONDITIONSMINTYPMAXUNITS
SPI INTERFACE
SCLK Clock Periodt
SCLK Pulse-Width Hight
SCLK Pulse-Width Lowt
SCLK Rise to DOUT Transitiont
CS Fall to SCLK Rise Setup Timet
SCLK Rise to CS Rise Setup Timet
DIN to SCLK Rise Setup Timet
DIN to SCLK Rise Hold Timet
CS Pulse-Width Hight
CS Rise to DOUT Disablet
CS Fall to DOUT Enablet
EOC Fall to CS Fallt
CP
CH
CL
DOT
CSS
CSH
DS
DH
CSPWH
DOD
DOE
RDS
C
= 20pF115ns
LOAD
= 20pF20ns
C
LOAD
C
= 20pF1ns
LOAD
25ns
10ns
10ns
10ns
5ns
10ns
0ns
10ns
10ns
FINE DAC SETTLING
75% TO 25% FS STEP
MAX11043 toc10
500mV/div
0V
FINE DAC SETTLING
1% STEP-UP
MAX11043 toc11
1μs/div
20mV/div
1200mV
MAX11043
4-Channel, 16-Bit, Simultaneous-Sampling ADCs
with PGA, Filter, and 8-/12-Bit Dual-Stage DAC
8, 22DVDDDigital Supply. Bypass each DVDD with a nominal 1µF capacitor to DGND.
9DVREGRegulated Digital Core Supply. Bypass DVREG to DGND with a 10µF capacitor.
10UP/DWNDAC Step Direction Select. Drive high to step up, drive low to step down when DACSTEP is toggled.
11DACSTEP
12CONVRUN
13CSActive-Low Serial-Interface Chip Select
14DOUTSerial-Interface Data Out. Data transitions on the rising edge of SCLK.
15DINSerial-Interface Data In. Data is sampled on the rising edge of SCLK.
16SCLKSerial-Interface Clock
17, 35I.C.Internally Connected. Connect to either AGND or DGND.
18EOCActive-Low End-of-Conversion Indicator. EOC asserts low to indicate that new data is ready.
19OSCINCrystal Oscillator/External Clock Input
20OSCOUTCrystal-Oscillator Output. Leave unconnected when using external clock.
21SHDNActive-High Shutdown Input. Drive high to shut down the MAX11043.
25AOUTBuffered 12-Bit Fine DAC Output
27REFDACLFine DAC Low Reference Bypass. Bypass REFDACL with a nominal 1µF capacitor to AGND.
28REFDACHFine DAC High Reference Bypass. Bypass REFDACH with a nominal 1µF capacitor to AGND.
29REFDACCoarse DAC Reference Bypass. Bypass REFDAC with a nominal 1µF capacitor to AGND.
30REFDChannel D Reference Bypass. Bypass REFD with a nominal 1µF capacitor to AGND.
31AINDNChannel D Analog Negative Input
32AINDPChannel D Analog Positive Input
34REFBPMain Reference Bypass. Bypass REFBP with a nominal 1µF capacitor to AGND.
36AINCNChannel C Analog Negative Input
37AINCPChannel C Analog Positive Input
38REFCChannel C Reference Bypass. Bypass REFC with a nominal 1µF capacitor to AGND.
39REFBChannel B Reference Bypass. Bypass REFB with a nominal 1µF capacitor to AGND.
40AINBPChannel B Analog Positive Input
—EP
DAC Step Input. Drive high to move the DAC output in the direction of UP/DWN on the next rising
edge of the system clock.
Convert Run. Drive high to start continuous conversions on all 4 channels. The device is idle when
CONVRUN is low.
Exposed Pad. Connect EP to a ground plane on the PCB to enhance thermal dissipation. Internally
connected to AGND. Not intended as an electrical connection point.
The MAX11043 features 4 single-ended or differential
channels of simultaneous-sampling ADCs with 16-bit
resolution. The MAX11043 contains a versatile filter
block and PGA per channel. The filter consists of seven
cascaded 2nd-order filter sections for each channel
allowing the construction of a 14th-order filter. The filter
coefficients are user-programmable. Configure each
2nd-order filter as a LP filter, HP filter, or BP filter with
optional rectification. Gain and phase mismatch of the
analog signal path is better than -50dB.
The ADCs can sample up to 800ksps per channel. A
40MHz serial interface provides communication to and
from the device. The SPI interface provides throughput
of 1600ksps; 4 channels at 400ksps per channel or 2
channels at 800ksps per channel. A software-selectable scan mode allows reading the ADC results while
simultaneously updating the DAC. Other features of the
MAX11043 include an internal (+2.5V) or external
(+2.0V to +2.8V) reference, power-saving modes, and
a PGA with gains of 1 to 64. The PGA includes an EQ
function that automatically boosts low-amplitude, highfrequency signals for applications such as CW-chirp
radar.
The MAX11043 includes two 8-bit coarse DACs that set
the high and low references for a second-stage 12-bit
fine DAC, typically used for VCO control. Use software
controls to set the DAC, or step the DAC up and down
using hardware control in programmable steps.
MAX11043 Signal Path
Each of the 4 ADC channels features a PGA and filter
block that feeds the signal to the sigma-delta modulator. The PGA can either be bypassed, which provides a
gain of 1, set to a gain of 8, a gain of 16, or set to analog EQ mode. For more amplification, set the ADC modulator gain to one, two, or four. After the modulator, the
result passes through the sinc 5 filter and decimator.
Seven biquad programmable digital filters isolate the
band of interest. Read the result using the 40MHz SPI
interface. See Figure 1.
Analog-to-Digital Converter
The MAX11043 features a quad sigma-delta ADC architecture with 4 differential input channels. For singleended operation, connect the N input to the
common-mode voltage or bypass to AGND with a 10µF
capacitor. All inputs feature a programmable bias generator; see the
CONFIG_ Register (0Ch–0Fh)
section.
All four ADCs convert simultaneously with a maximum
modulator sampling rate of 9.6Msps; decimated by 12
or 24 for output rates of 800ksps and 400ksps, respectively. The SPI bus limits the maximum output data rate
to 40Mbps.
Sinc 5 Filter
The sinc 5 filter removes high-frequency noise from the
output of the sigma-delta modulator. It also decimates
the modulator data by a factor of 12, providing a maximum of 800ksps to the programmable filters when the
modulator is operating at 9.6Msps. Figure 2 shows the
frequency characteristics of the sinc 5 filter with the
modulator running at 9.6Msps. Operating the modulator
at a lower sample rate causes a proportional reduction
in the frequency response of the sinc 5 filter. The total
attenuation of the MAX11043 is the sum of the analog
filtering, the sinc 5 filter, and the seven stages of programmable filters.
Equalizer (EQ)
The EQ matches the frequency/gain characteristics of
CW-chirp radar systems where the distance to the target is proportional to the measured frequency. Distant
targets not only have a higher frequency, they have a
weaker signal. Hence, higher frequencies need more
amplification than lower frequencies. The EQ provides
gain proportional to frequencies up to 190kHz, at which
point the gain rolls off at 80dB/decade.
The EQ consists of an analog section in the PGA and a
digital EQ created from the biquad filters. The analog
EQ (PGA) provides 20dB/decade of gain and the
default digital EQ provides an additional 20dB/decade
of gain. Together they provide 40dB/decade of gain up
to 190kHz with a gain of 0dB at 5kHz.
Variations in the manufacturing process affect the gain
and phase of the analog filter. Compensation for these
variations include adjustments to the digital filter during
the manufacture of the MAX11043. Use the analog and
digital EQs together for optimal performance. For a
detailed description of digital-filter customization, refer
to the
MAX11043 User’s Guide
.
Conversion and ADC Reading
Drive CONVRUN high to initiate a continuous conversion on all 4 channels. Keep CONVRUN high for the
entire conversion process. Do not pulse CONVRUN.
EOC asserts low when new data is available. Initiate a
data read prior to the next rising edge of EOC or the
result is overwritten. EOC asserts high upon read completion of all active channels. Use ConfigA, ConfigB,
ConfigC, and ConfigD registers to read single channel
data. Concatenated data is available in the ADCAB,
ADCCD, and ADCABCD registers. Use concatenated
registers to ensure simultaneous results are read. See
the
Register Functions
section for more details.
A software-selectable scan mode automatically sends
the result from selected channels following the CS
falling edge and allows other registers to be simultaneously updated. To enable scan mode, set SCHAN_ bits
high. See the
Configuration Register (08h)
section for a
detailed description. The ADC output is presented in
two’s complement format (Figure 3).
Digital Filter
Seven cascaded, individually configurable, 2nd-order
filter elements make up the digital filter. Figure 4 shows
the structure of a single filter section. Configure these
elements as LP, BP, HP, or all pass (AP) filters with
optional rectification. Filter configuration is transferred
from the flash to coefficient RAM (C-RAM) on power-up.
Store custom filters permanently in the flash or write
directly to C-RAM each time on power-up. Two separate sets of programmable coefficients exist for each
filter. Dual coefficient sets allow rapid filter reconfiguration. These filter coefficients are programmed to LP and
EQ modes at the factory. Multiple flash memory pages
exist so that custom filters can be created while preserving factory-programmed filter coefficients.
SINC 5 FILTER AT 9.6Msps
MAX11043 fig02
FREQUENCY (kHz)
ATTENUATION (dB)
16001200800400
-100
-80
-60
-40
-20
0
-120
02000
Figure 2. Sinc 5 Filter Frequency Response
0 +1+FS-1-FS
1000 0000 0000 0000
1000 0000 0000 0001
1000 0000 0000 0010
INPUT VOLTAGE (LSB)
BINARY OUTPUT CODE
0111 1111 1111 1101
0111 1111 1111 1110
0111 1111 1111 1111
0000 0000 0000 0000
0000 0000 0000 0001
1111 1111 1111 1111
Figure 3. Two’s Complement Transfer Function
MAX11043
4-Channel, 16-Bit, Simultaneous-Sampling ADCs
with PGA, Filter, and 8-/12-Bit Dual-Stage DAC
Filter coefficients A1 and B1 are always 1. B3 is limited
to -1, 0, and 1.
Filter coefficients A2, A3, and B2 are stored as 16-bit
two’s complement values in the range of -4 to +4.
Gain is limited to the following values 2
4
, 22, 20, 2-2, 2-4,
2
-6
, 2-8, and 2
-10
. For better gain resolution, adjust the
Fine Gain A/B/C/D Registers at the input of each filter
set. Fine gain adjustment has a resolution of 16 bits and
a gain range of -4 to +4. Set the RECT bit to rectify the
filter output.
Figures 5–8 show the response to a step input of the
default filters used for ADC trimming.
1/A1
-A2
Z
-1
-A3
B1
B2
B3
+
+
+
+
OUT
IN
X
Y
RECT
G
ABS
Z
-1
Figure 4. Single Programmable 2nd-Order Filter Section
Each ADC channel features an input buffer with input
impedance of at least 5kΩ and programmable gain of
eight or 16. When set to a gain of one, the signal
bypasses the PGA to reduce noise.
The PGA features an optional 20dB/decade analog EQ
mode, with a gain of 0dB at 5kHz and attenuation
above 190kHz to reduce out-of-band noise. Using the
digital EQ filter adds another 20dB/decade gain.
Control the EQ and PGA gain from their respective
CONFIG_ registers. For additional filtering and equalization, use the integrated digital filters. Refer to the
MAX11043 User’s Guide
for more information.
Digital-to-Analog Converter
The MAX11043 features a 12-bit fine DAC with high and
low reference inputs set by the 8-bit, dual tap coarse DAC
or driven externally. The output buffer of the fine DAC has
a gain of two and can drive 10kΩ and 200pF in parallel.
Bypass the REFDACH and REFDACL with a 1µF capacitor when using the coarse DAC to set the reference
values, or power down the buffers and drive REFDACH
and REFDACL with external references. Alternatively
drive one of the fine DAC references using the coarse
DAC and the other using an external reference.
The fine DAC register contains the current value of the
output. The output value changes by writing to this register or by the rising edge of the DACSTEP input. The
DAC register updates on the next rising edge of the
system clock following the rising edge of the DACSTEP
input. The programmable DACSTEP register contains
the step size. The UP/DWN input sets the direction of
the step. Drive UP/DWN high to step up, drive low to
step down.
The coarse 8-bit, dual tap DAC generates the high and
low reference values for the fine DAC. Obtain the
coarse DAC reference from the main reference or by
driving the REFDAC input externally. The main reference, REFBP, is divided by two before the coarse DAC.
When driving REFDAC, REFDACH, or REFDACL directly, ensure the voltage to the fine DAC does not exceed
AVDD/2 to prevent the output amplifier from saturating.
EQ FILTER OUTPUT
MAX11043 fig06
SAMPLE
OUTPUT (LSB)
80602040
-15,000
-10,000
-5000
0
5000
10,000
15,000
20,000
25,000
30,000
35,000
-20,000
0100
Figure 6. EQ Filter Response to a Step Input
STAGE 1 FILTER OUTPUT
MAX11043 fig08
SAMPLE
OUTPUT (LSB)
40103020
0
500
1000
1500
2000
2500
3000
3500
-500
050
Figure 8. Stage 1 Default Filter Response to a Step Input
LP FILTER OUTPUT
MAX11043 fig07
SAMPLE
OUTPUT (LSB)
80604020
500
1000
1500
2000
2500
0
0100
Figure 7. LP Filter Response to a Step Input
MAX11043
4-Channel, 16-Bit, Simultaneous-Sampling ADCs
with PGA, Filter, and 8-/12-Bit Dual-Stage DAC
The MAX11043 features an internal 2.5V bandgap reference. Bypass REFBP with a 1µF capacitor or power
down the buffer amplifier and drive REFBP with an
external reference. In internal reference mode, REFBP
provides the main reference voltage for the MAX11043.
Refer to www.maxim-ic.com/references for a list of
available precision references.
In addition to the integrated main reference, there are
seven separate references derived from REFBP, one for
each ADC channel, one for the coarse DAC, and two
(one high and one low) for the fine DAC. When using
the main reference, bypass each of the references with
a 1µF capacitor or set the appropriate bits (7–0), in the
reference (10h) register, to power down the references
and drive externally. Use external references capable
of driving a 700µA or total load.
Clock Sources
The MAX11043 features an internal 16MHz oscillator
that supports either an external crystal or ceramic resonator. For highest performance, set bit 15 in the configuration register to 1 and use an external clock (EX
clock) source, up to 40MHz, to drive OSCIN. A programmable clock divider divides the EX clock by 2, 3,
4, or 6 to generate the ADC sample clock. The system
clock, used for all digital timing, is twice the ADC sample clock. Ensure that the minimum EX clock high or
low time is greater than 25ns when using the divide-by2 or divide-by-3 mode.
The system clock, used for all internal timing, is derived
from the clock divider setting and the input clock.
For optimal performance, derive the SPI clock and system clock from the same source.
Power Saving
The MAX11043 features an active-high power-down
input, as well as an SPI-controlled power-down bit that
places the MAX11043 in low-power mode. In addition,
the MAX11043 features an independent, SPI-controlled,
power-down for each ADC channel, the DAC, and the
oscillator. See the
Configuration Register (08h)
section
for more details.
Serial Communication
The SPI-compatible interface allows synchronous serial
data transfers up to 40Mbps. The bandwidth is divided
between the DACs and the ADC. Maximum conversion
throughput depends on which read commands are
used. The highest conversion rates are obtained by
using the scan mode. The second highest rate is
obtained by reading concatenated registers. The slowest method is to read the results individually.
Configure the SPI master for SCLK to idle low (SCLK is
low when CS is asserted). The data at DIN is latched on
the rising edge of SCLK. Data at DOUT transitions
immediately after the rising edge of SCLK.
All SPI transactions start with a command byte. The
command byte selects the address of the register and
the mode of operation (read/write).
SPI Command Byte
START<7>: Start bit. This bit must be 0 for normal
operation.
ADR_<6:2>: Device register address bits. See the register map in Table 1.
R/W<1>: Read/write bit. 1 = read from device. 0 = write
to device.
The ADC channel A, B, C, and D result registers provide the result data from the 4 ADC channels. EOC
asserts low when new data is available. Initiate a data
read prior to the next rising edge of EOC or the result is
overwritten. Set bit 5 of the configuration register 08h
high to read the data out in 24-bit resolution or set bit 5
low to read the data out in 16-bit resolution.
ADCAB, ADCCD, and ADCABCD
Result Registers (04h–06h)
Registers ADCAB, ADCCD, and ADCABCD contain
concatenated ADC results ensuring simultaneous
results are read. This reduces the risk of reading samples delayed by one cycle from channel to channel.
Set bit 5 of the configuration register 08h high to read
the data out in 24-bit resolution or set bit 5 low to read
the data out in 16-bit resolution.
Status Register (07h)
The status register contains the channel overflow flags
and POR bits.
X<7:6>: Don’t-care bits.
Flash Busy<5>: Do not start a new flash operation until
this is 0.
BOOT<4>: Power-on reset flag.
OFLG_<3:0>: Channel overflow flag, one per channel.
Configuration Register (08h)
EXTCLK<15>: External clock select.
1 = logic-level clock supplied on OSCIN.
0 = crystal or resonator connected between OSCIN
and OSCOUT (default).
CLKDIV1:CLKDIV0<14:13>: Clock divider ratio (EX
clock : ADC sample clock).
00 = 1:2 clock divider.
01 = 1:3 clock divider.
10 = 1:4 clock divider.
11 = 1:6 clock divider (default).
PD<12>: Power-down analog circuitry (reference and
SPI interface remains active).
1 = low-power mode.
0 = normal operation (default).
PD_<11:8>: ADC power-down for each channel (A, B,
C, and D).
1 = powers down analog signal path.
0 = normal operation (default).
PDDAC< 7>: DAC power-down.
1 = fine DAC buffer powered down.
0 = normal operation (default).
PDOSC<6>: Oscillator power-down.
1 = oscillator powered down (disconnects EX clock in
EX clock mode).
0 = normal operation (default).
24BIT<5>: ADC output data format.
1 = ADC data output as 24 bits.
0 = ADC data output as 16 bits (default).
Use the 24-bit ADC output in conjunction with external
digital filtering to improve signal-to-noise ratio.
BIT 7BIT 6BIT 5BIT 4BIT 3BIT 2BIT 1BIT 0
XXFlash BusyBOOTOFLGAOFLGBOFLGCOFLGD
BIT 15BIT 14BIT 13BIT 12BIT 11BIT 10BIT 9BIT 8
EXTCLKCLKDIV1CLKDIV0PDPDAPDBPDCPDD
BIT 7BIT 6BIT 5BIT 4BIT 3BIT 2BIT 1BIT 0
PDDACPDOSC24BITSCHANASCHANBSCHANCSCHANDDECSEL
MAX11043
4-Channel, 16-Bit, Simultaneous-Sampling ADCs
with PGA, Filter, and 8-/12-Bit Dual-Stage DAC
1 = ADC channel data is output on DOUT each time a
new result is valid in the sequence, A, B, C, and D.
0 = ADC data is not presented automatically for this
channel (default).
When SCHAN_ = 1, the selected ADC channel data is
automatically presented on DOUT each time EOC
asserts low in the sequence A, B, C, and D with the
unselected channels omitted. The data transitions on
the rising edge of SCLK. Force CS low to initiate transmission. CS can go high between results. The MSB of
the first selected ADC channel outputs immediately
after the falling edge of EOC. EOC goes high after the
last bit of the selected channels clocks out or one clock
cycle before the next result is ready. Insufficient SCLK
pulses result in truncated data. Extra clock pulses give
an undefined output. In scan mode, keep DIN high or
write data to the MAX11043 as usual. In scan mode,
the MAX11043 ignores requests for data reads.
DECSEL<0>: Decimate select.
1 = decimate by 12.
0 = decimate by 24 (default).
Set DECSEL high to decimate the ADC result by 12,
doubling the number of samples. The SPI interface is
limited to 40Mbps.
Fine DAC Register (09h)
X<15:12>: Don’t-care bits.
DAC_<11:0>: Contains current fine DAC output value.
When using the DACSTEP input to change the DAC
value, this register updates to the new value on the
next rising edge of the system clock following the rising
edge of DACSTEP. The power-on default is 0.
DACSTEP Register (0Ah)
X<15:12>: Don’t-care bits.
DACSTEP11:DACSTEP0<11:0>: Provides the size of
the DAC step. The value is positive only and the
UP/DWN input is used to set the direction. The value in
the fine DAC register updates on the next rising edge
of the system clock following the rising edge of the
DACSTEP input. The power-on default is 0.
Coarse DACH/DACL Register (0Bh)
DACH7:DACH0<15:8>: High coarse DAC value.
DACL7:DACL0<7:0>: Low coarse DAC value.
Coarse DAC sets high and low references for the fine
DAC. The power-on default is 0.
00h = first available sample is presented (default).
1Fh = 31 results are discarded.
Digital filters retain a history of past input data. At
power-up and when changing the signal path, old data
requires purging before new output data is valid.
PURGE4(MSB):PURGE0 determine the number of samples to discard before a new result is valid. Each time
CONVRUN is taken high, N results are discarded
before EOC asserts low (where N is the decimal equivalent of the binary representation of PURGE4:PURGE0).
Results prior to N+1 are overwritten. EOC asserts for
results N+1, N+2, N+3, etc., as long as CONVRUN
remains high. Taking CONVRUN low and then high
invokes another purge.
Purging of the sinc 5 filter requires five readings if
DECSEL (configuration register 08h, bit 0) = 1 and
three readings if DECSEL = 0. The minimum total purge
interval of the seven cascaded filters is one reading if
not used. If the filters are used, the total latency of the
programmable filters is the sum of the latency caused
by each stage. Set the appropriate delay for filter purging and settling time.
This is a 16-bit register that contains the data for a flash
write operation. Default = 0.
Flash Data Out Register (1Bh)
This is a read-only register. Data is valid only if flash
busy is zero.
This is a 16-bit register that contains the data for a flash
read operation.
Flash and C-RAM Register Map
The flash memory consists of 2048 words by 16 bits.
The 3 MSBs of the flash address select one of eight
pages of 256 words each. Page zero contains the
default filter coefficients for channels A and B. Page
one contains the default filter coefficients for channels
C and D. Use pages two and three for the coefficients
of custom filters. When the first word on page two contains a nonzero value, the MAX11043 loads these
pages into C-RAM at power-up instead of the default
values from pages zero and one. Flash pages zero and
one include trim data. Unique trim data optimizes the
performance of each MAX11043. To maintain optimum
performance when using custom filters, copy the trim
data from flash pages zero and one to the corresponding locations in flash pages two and three or to C-RAM
when writing directly to C-RAM.
Further optimization of the MAX11043 is achieved
through stage one filter coefficients for each channel.
When using custom filters, copy stage one coefficients
from pages zero and one to the corresponding locations in flash pages two and three or to C-RAM when
writing directly to C-RAM. Table 2 identifies the default
stage one filters (EQ and LP) for the MAX11043. For
custom filters, use stages two through seven first, and
only change the stage one coefficients when all seven
stages require customization.
The flash addresses below are for channel A; for channel B add 80h, for channel C add 100h, and for channel
D add 180h. To write to pages two and three of flash,
add 200h to the above values.
To load the coefficients directly to C-RAM, create a 32bit data word by concatenating the data in adjacent
flash locations as shown in Table 3. The C-RAM
addresses below are for channel A; for channel B add
40h, for channel C add 80h, and for channel D add
C0h.
Multiple addresses exist for some stage 1 filter coefficients as shown in Table 3. The address accessed by
the filter depends on the configuration bits as shown in
Table 2.
Table 2. Stage One Filter Selection
Table 3. C-RAM and Flash Memory Map
FILTER FIRST STAGEEQPGAPDMODGPGAG
EQ filter stage 1 (C-RAM address 03h–05h)10XXX
LP filter for ADC gain of 1, 2, and 4; stage 1 (C-RAM address 1Dh–1Fh)X1XXX
LP filter for ADC gain of 8; stage 1 (C-RAM address 3Dh–3Fh)00000
LP filter for ADC gain of 16; stage 1 (C-RAM address 23h–25h)00XX1
C-RAM
ADDRESS
00h
01h
02h
03h
FLASH
ADDRESS
00h—Not used
01h*EQ gain trim for gain = 1—
02h—Not used
03hUser trim for EQ gain, default = 2000h—
04h—Not used
05hNot used—
06h*—EQ filter gain for filter stage 1
07h*EQ filter coefficient A2 for filter stage 1—
MSB FOR C-RAMLSB FOR C-RAM
MAX11043
4-Channel, 16-Bit, Simultaneous-Sampling ADCs
with PGA, Filter, and 8-/12-Bit Dual-Stage DAC
When erasing or programming the flash, maintain the
system clock between 14MHz and 27MHz to satisfy
flash timing requirements and ensure CONVRUN = 0.
The system clock used for all digital timing is twice the
ADC sample clock (2 x EX clock/divider).
Always erase the flash page before writing new data.
The procedure for flash mass erase is as follows:
1) Read the flash mode register (18h); proceed when
the LSB is zero.
2) Write 0000h to the flash address register (19h).
3) Write 60h to the flash mode register (18h).
4) Wait 200ms for erase to complete.
5) FFFFh = flash erased state.
The procedure for flash single page erase is as follows:
1) Read the flash mode register (18h); proceed when
the LSB is zero.
2) Write page address, set word address to 00h in the
flash address register (19h).
3) Write 40h to the flash mode register (18h).
4) Wait 20ms for page erase to complete.
5) FFFFh = flash erased state.
The procedure for flash single word write is as follows:
1) Read the flash mode register (18h); proceed when
the LSB is zero.
2) Write page and word address to the flash address
register (19h).
3) Write the data to the flash data in register (1Ah).
4) Write 20h to the flash mode register (18h).
5) Read the flash mode register (18h); proceed when
the LSB is zero (approx. 40µs).
The procedure for flash single word read is as follows:
1) Read the flash mode register (18h); proceed when
the LSB is zero.
2) Write page and word address to the flash address
register (19h).
3) Write 80h to the flash mode register (18h).
4) Read the flash mode register (18h); proceed when
the LSB is zero (approx. 1µs).
5) Read the data from the flash data out register (1Bh).
The procedure for flash to C-RAM transfer is as follows:
1) Read the flash mode register (18h); proceed when
the LSB is zero.
2) Write A0h to the flash mode register (18h).
3) Read the flash mode register (18h); proceed when
the LSB is zero (approx. 1ms).
4) The content of flash is transferred to C-RAM.
Table 3. C-RAM and Flash Memory Map (continued)
*
Recommended copy to C-RAM or flash for optimum custom-filter performance.
C-RAM
ADDRESS
3Bh
3Ch
3Dh
3Eh
3Fh
FLASH
ADDRESS
76h—Not used
77h*ADC gain trim for gain = 32—
78h—Not used
79h*ADC gain trim for gain = 64—
7Ah*—LP filter gain for filter stage 1, gain = 8
7Bh*
7Ch—Not used
7Dh*
7Eh*—
7Fh*
LP filter coefficient A2 for filter stage 1,
gain = 8
LP filter coefficient A3 for filter stage 1,
gain = 8
LP filter coefficient B2 for filter stage 1,
gain = 8
MSB FOR C-RAMLSB FOR C-RAM
—
—
LP filter coefficient B3 and rectify bit for filter stage 1,
gain = 8
—
MAX11043
4-Channel, 16-Bit, Simultaneous-Sampling ADCs
with PGA, Filter, and 8-/12-Bit Dual-Stage DAC
Filter coefficients A2, A3, and B2 are stored as 16-bit
two’s complement values in the -4 to (4 - 2
-13
) range.
The transfer function equation is as follows:
A2 = int (N x 213)
where N is the decimal coefficient value.
The following are two examples of the transfer function
equation:
Example 1:
N = 2.381
A2 = int (2.381 x 2
13
)
A2 = int (19505.152)
A2 = 19505 = 4C31h (two’s complement)
Example 2:
N = -2.381
A2 = int (-2.381 x 213)
A2 = int (-19505.152)
A2 = -19505 = B3CFh (two’s complement)
B3 Coefficient (55h)
B31:B30<15:14>: Filter coefficient B3.
11 = -1.
00 = 0.
01 = 1.
10 = 0.
X<13>: Don’t-care bit. Not used.
RECT<12>: Rectify bit.
0 = bipolar output.
1 = output rectified. All samples positive.
X<11:0>: Don’t-care bits. Not used.
Power Supplies, Layout, and
Bypassing Considerations
For best performance, use PCBs with ground planes.
Ensure that digital and analog signal lines are separated from each other. Do not run analog and digital lines
parallel to one another (especially clock lines), and do
not run digital lines underneath the MAX11043 package. Use a single-point analog ground (star ground
point) at AGND, separate from the logic ground.
Connect all other analog grounds and DGND to this
star ground point. Do not connect other digital system
grounds to this single-point analog ground. The ground
return to the power supply for this ground should be
low impedance and as short as possible for noise-free
operation. Bypass all supplies to ground with high
quality capacitors as close as possible to the device.
Chip Information
PROCESS: BiCMOS
Package Information
For the latest package outline information and land patterns, go
to www.maxim-ic.com/packages
.
PACKAGE TYPEPACKAGE CODEDOCUMENT NO.
40 TQFNT4066-5
21-0141
BIT 15BIT 14BIT 13BIT 12BIT 11BIT 10BIT 9BIT 8
B31B30XRECTXXXX
BIT 7BIT 6BIT 5BIT 4BIT 3BIT 2BIT 1BIT 0
XXXXXXXX
MAX11043
4-Channel, 16-Bit, Simultaneous-Sampling ADCs
with PGA, Filter, and 8-/12-Bit Dual-Stage DAC
Maxim cannot assume responsibility for use of any circuitry other than circuitry entirely embodied in a Maxim product. No circuit patent licenses are
implied. Maxim reserves the right to change the circuitry and specifications without notice at any time.
32
____________________Maxim Integrated Products, 120 San Gabriel Drive, Sunnyvale, CA 94086 408-737-7600