MB86060PMCR-G-BNDE1 [FUJITSU]
D/A Converter, 1 Func, Parallel, Word Input Loading, 0.004us Settling Time, PQFP80, ROHS COMPLIANT, PLASTIC, LQFP-80;![MB86060PMCR-G-BNDE1](http://pdffile.icpdf.com/pdf2/p00237/img/icpdf/MB86060PMCR-_1391551_icpdf.jpg)
型号: | MB86060PMCR-G-BNDE1 |
厂家: | ![]() |
描述: | D/A Converter, 1 Func, Parallel, Word Input Loading, 0.004us Settling Time, PQFP80, ROHS COMPLIANT, PLASTIC, LQFP-80 转换器 |
文件: | 总44页 (文件大小:745K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
![](http://public.icpdf.com/style/img/ads.jpg)
Datasheet
February 2009
Version 2.0
MB86060
FME/MS/SFDAC1/DS/4250
16-bit Interpolating Digital to Analog Converter
The Fujitsu MB86060 is a high performance 12-bit, 400MSa/s
digital to analog converter (DAC) enhanced with a 16-bit
PLASTIC PACKAGE
LQFP-80
interpolation filtering front-end. Use of novel techniques for the
converter architecture delivers high speed operation consistent
with BiCMOS or bipolar devices but at the low power of CMOS.
Fujitsu’s proprietary architecture is the subject of several patent
applications. Additional versatility is provided by selectable
input interpolation filters, programmable dither and noise
shaping facilities. Excellent SFDR performance coupled with
high speed conversion rate and low power make this device
particularly suitable for high performance communication
systems, in particular direct IF synthesis applications.
This device is a pin and functional replacement for,
• MB86060PFV-G-BND-FG (RoHS-5)
• MB86060PFV-G-BNDE1 (RoHS-6)
PIN ASSIGNMENT
FEATURES
• 16-bit Interpolating Digital to Analog conversion
• x2 or x4 interpolation filtering
• 100MSa/s input, with x4 interpolation enabled
• Programmable highpass filtered dither
• Selectable 2nd order noise shaping
DITH0
DITH1
DITH2
VSS
VDD
TWOC
MUL0
MUL1
MUL2
VSS
VDD
CKOUT
CKOUTB
VSS
N/C
VSS
VDD
DSUB
OVER
ASUB
FILTS
VDD
• Versatile CMOS digital interface
• Internal programmable clock multiplier
• Low power, 3.3V operation (345mW @32MSa/s input, x4)
• Performance enhanced pinout with on-chip decoupling
• 0.35µm CMOS technology with Triple Well
• Industrial temperature range (-40°C to +85°C)
DSUB
LOCK
CSUB
NSHAPE
CLKSEL
RESETB
CVDD
CVSS
CLK
MB86060
FILTF
SHUF0
SHUF1
RVSS
BGAP
VREF
CVSS
CLKB
RREF
Pin #1
APPLICATIONS
• Direct IF Synthesis
• Cellular basestations
• Wide-band communication systems
This product has Patents applied for in the US and elsewhere including GB2333191A, EP0935345A, JP11-274934A, GB2333171A, EP0930717A, JP11-
274935A, GB2333190A, EP0929158A, JP11-243339A, GB2335097A, EP0940923A, JP11-317667A, GB2335076A, EP0940852A, JP11-251530A.
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 1 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
Contents
1
Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1 Operating Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Clock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.1 Input Clock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.2 Clock Multiplier Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2.3 Multiple Device Clock Synchronisation. . . . . . . . . . . . . . . . . 8
1.2.4 Clock Out . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.3 Interpolating Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.4 Programmable Dither. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.5 Noise Shaping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.6 Converter Architecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.6.1 Segment Shuffling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.6.2 Converter Overload. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.7 Voltage Reference. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.8 Analog Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.8.1 Analog Output Reference Resistor . . . . . . . . . . . . . . . . . . . 12
1.8.2 Analog Output Scaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.8.3 Analog Output Pins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.9 Digital Data Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.9.1 Data Timing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.10 Power Supplies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.10.1 Substrate Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.10.2 Power Dissipation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.10.3 Pinout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.11 Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2
3
4
Interpolating Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Dither Frequency Spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Application Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.1 Power & Ground Plane Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.2 Power Supplies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.3 Decoupling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.4 Input Data Interfacing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.5 Analog Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.6 Clock Input. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.7 Clock Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
4.8 Signal Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Page 2 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
5
Electrical Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
5.1 Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
5.2 Digital Interface Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
5.3 DC Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
5.4 AC Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
5.5 Clock Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.6 Typical Performance Characterisation Graphs . . . . . . . . . . . . . . . . 36
6
7
Pin Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
6.1 Pin Assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
6.2 Pin Definition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
6.3 Package Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
6.3.1 Thermal Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
6.4 Ordering Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Development Kit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
7.1 Ordering Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 3 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
1
Functional Description
The MB86060 integrates a 12-bit 400MSa/s DAC core with selectable front end processing to provide
input interpolation filtering, dither and noise shaping. Versatile interfacing via the 16-bit parallel
CMOS data input allows different system requirements to be accommodated, with either offset binary
or 2’s complement data formats selected by an input format control pin.
The device is manufactured in a 0.35µm advanced CMOS process with Triple Well extension giving
improved isolation between analog blocks and digital-analog.
A functional block diagram is shown in Figure 1.
Clk Select
CLK in (diff.)
Crystal
Clock
Multiplier
2
Dither
Programmable
Dithermay be
excluded from
final MP
Lock
Mult mode
Delay line ctrl
Generator
3
2
devices
HP Filter
Data CLK out
(diff.)
Clock Divider
3
Dither
[x2 slow]
[x2 fast]
x2
NS Enable
16
16
Noise
Shaper
Data In
x2
Data Format
Over
12
DAC
Output
2
DAC
Bandgap
Reference
Filter control
Reset
2
Shuffle
Control
FML Mixed
Figure 1 MB86060 Functional Block Diagram
1.1 Operating Modes
The device can be configured into a number of different operating modes, depending on which clock
source and interpolation filtering mode is selected. The following sections summarise the MB86060
operating modes according to clock multiplier and interpolation filtering configuration.
The MB86060 has two Interpolation filters, that may be operated in any of the following four modes.
• x1 - Interpolating filters disabled, effectively a conventional 12-bit DAC for up to 200MSa/s
• x2 slow - First interpolating filter only, used for generating 0~43MHz, assuming 100MSa/s data
• x2 fast - Second interpolating filter only, used for generating 0~74MHz, assuming 200MSa/s data
• x4 - Full interpolation, for generating 0~43MHz with 100MSa/s data & maximum DAC rate
Page 4 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
1.2 Clock
The MB86060 incorporates a clock multiplier. This may be used to generate the internal x1, x2 and
x4 clock signals required to clock the DAC core when the interpolation filters are enabled, or as a
general purpose clock multiplier to allow lower frequency clock sources to be used.
The clock multiplier is based on a delay-lock-loop whose delay is adjusted by a charge pump
controlled by a phase detector. A ‘Lock’ indicator pin is provided so that the system can monitor the
multiplier’s condition.
For systems where a high frequency clock is available, or the lowest possible jitter is required, the
clock multiplier should be disabled and an external clock applied directly as the clock multipliers
systematic jitter will cause jitter spurs to appear in the analog output
1.2.1
Input Clock
The input clock is selectable between either a differential system clock, typically a sine wave source
of amplitude 0dBm, or an external crystal using the internal oscillator circuit. A CMOS single ended
clock can also be connected to XIN. See Table 1 for details of these configurations.
Table 1: Input Clock Source Selection
Clock multiplier
mode
Clock
source
CLKSEL
Function
0
x1 (either filter)
or x2 to x8
Crystal
oscillator
Connect crystal between XIN and XOUT, or
connect CMOS clock to XIN
1
Any
Differential Connect differential clock source to CLK and
clock CLKB
When using the internal oscillator with an external crystal, or connecting a single-ended
CMOS clock to XIN, the clock multiplier must be set to multiply modes x2 to x8 so that XIN
is enabled. The operating speed of the internal crystal oscillator circuit is limited. See
section 5.5.
The differential input clock pins CLK and CLKB are internally biased to have a common mode level
equivalent to the voltage applied to pin VREF. When using either the internal oscillator or an external
single ended clock connected to XIN, pins CLK and CLKB should be linked to CVSS.
If minimum jitter is required, then the differential clock should be used, with an amplitude sufficient to
ensure that the specification for minimum slew rate is met. For a 250MHz clock this represents 0dBm,
with higher amplitudes required for lower clock rates. A sine signal is recommended over a square
wave to avoid unwanted harmonics.
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 5 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
tsl
thi
CLK In
Data CLK
XIN
tdel
tlo
tdelx
Figure 2 CLK In and Data CLK Timing
1.2.2
Clock Multiplier Modes
The clock multiplier can be set in one of eight modes. These take the form of two basic groups, either
multiplier functioning or bypassed.
Multiplier Bypassed
With the multiplier in bypass mode, MUL[2:0] = 000, then the clock frequency applied at CLK In, (F ),
IN
will be the update frequency used by the DAC core, (F
). The frequency available at Data CLK,
DAC
(F
), will be dependant on the interpolation filter settings. If no filters are selected, then F
=
DATA
DATA
F
F
. If the x2 filter, slow or fast, is selected, then F
= F
/2. If x4 filtering is selected, then
DAC
DATA
DAC
= F
/4. See Table 2.
DATA
DAC
Multiplier Enabled
With the multiplier enabled (modes x1 to x8, MUL[2:0] = 001 to 111), then the clock frequency at CLK
In (F ) will be applied to the input of the clock multiplier (F ), and multiplied by the chosen clock
IN
MULIN
multiplier setting. This will be the frequency available at Data CLK out, (F
). The update
DATA
frequency used by the DAC core, (F
) will then depend on the interpolation filter settings. If no
DAC
filters are selected, F
= F
. If the x2 filter, slow or fast, is selected, then F
= 2.F
. If x4
DAC
DATA
DAC
DATA
filtering is selected, then F
= 4.F
. Hence the DAC core sampling rate will be between 1.F
DAC
DATA IN
and 32.F . See Table 2.
IN
Page 6 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
Table 2: Multiplier and Filter Mode to DAC Update Frequency Cross Reference Matrix
Multiplier Mode
Filter
Mode
Bypass
x1
x2
x3
x4
x5
x6
x8
FDAC = FIN
FDAC = FIN FDAC = 2.FIN FDAC = 3.FIN FDAC = 4.FIN FDAC = 5.FIN FDAC = 6.FIN FDAC = 8.FIN
x1
x2
x4
FDATA = FDAC FDATA = FIN FDATA = 2.FIN FDATA = 3.FIN FDATA = 4.FIN FDATA = 5.FIN FDATA = 6.FIN FDATA = 8.FIN
FDAC = FIN FDAC = 2.FIN FDAC = 4.FIN FDAC = 6.FIN FDAC = 8.FIN FDAC = 10.FIN FDAC = 12.FIN FDAC = 16.FIN
FDATA = FDAC/2 FDATA = FIN FDATA = 2.FIN FDATA = 3.FIN FDATA = 4.FIN FDATA = 5.FIN FDATA = 6.FIN FDATA = 8.FIN
FDAC = FIN FDAC = 4.FIN FDAC = 4.FIN FDAC = 12.FIN FDAC = 16.FIN FDAC = 20.FIN FDAC = 24.FIN FDAC = 32.FIN
FDATA = FDAC/4 FDATA = FIN FDATA = 2.FIN FDATA = 3.FIN FDATA = 4.FIN FDATA = 5.FIN FDATA = 6.FIN FDATA = 8.FIN
F
and F
must not exceed limits set in Section 5.5. When the clock multiplier is
DATA
DAC
enabled (modes x1 to x8), F (F
) must be within the operational input range of the
IN
MULIN
clock multiplier. See Section 5.5.
Table 3: Clock Multiplier Configuration
MUL[2:0]
Clock Multiplier Mode
2
0
0
0
0
1
1
1
1
1
0
0
1
1
0
0
1
1
0
0
1
0
1
0
1
0
1
Bypass
x1
x2
†
x3
x4
†
x5
x6
x8
† Clock Multiplier modes x3 and x5 are implemented with a multiply by 6 and divide by 2,
and a multiply by 10 and divide by 2, respectively. Ensure that OSC[1:0] is set for the clock
frequency produced by the x6 or x10 multiplication. F
multiplication frequency.
(max) will apply to the
DAC
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 7 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
The OSC[1:0] setting controls the delay response within the delay-lock-loop. Different settings are
required to enable best jitter performance to be achieved and should be set according to the DAC
clock rate being used.
Table 4: Clock Multiplier Oscillator Configuration
DAC Update Frequency (F
)
OSC[1:0]
DAC
Mode
1
0
Min.
Max.
0
0
1
1
0
1
0
1
Fastest
250 MHz
150 MHz
80 MHz
250 MHz
150 MHz
80 MHz
Slowest
1.2.3
Multiple Device Clock Synchronisation
To allow multiple devices to be used with a common clock source, a clock synchronization function
is included. This will ensure that the clock out to clock in phase relationship is maintained.
Multiplier Bypassed
With the clock multiplier bypassed, the input clock (CLK In) signal bypasses the clock multiplier block
and connects directly to the clock divider. The divider generates clock out (Data CLK) depending
upon the interpolation filter settings. The phase relationship between clock in and the divided clock
out will vary between devices because the time taken for the divider to start up from power-up will
vary, which will cause the dividers to start from different input clock edges.
Clock synchronisation can be achieved between multiple devices by applying a short FULL-RESET
condition simultaneously during the positive half cycle of CLK In. This simultaneous reset will start
the dividers in each device from a common state, hence synchronising the clock outputs upon the
next input clock edge. The reset pulse must be shorter than one half of one CLK In cycle.
As such a short pulse may be difficult to produce, the input clock may be cleanly stopped for long
enough to create a reset pulse. When the clock is restarted it is important to ensure that no glitches
occur that may falsely trigger the clock input.
Multiplier Enabled
With the clock multiplier enabled, the input clock (CLK In) signal is routed through the multiplier block
before connecting to the clock divider. The multiplier block maintains the phase relationship between
clock in and clock out by producing a re-synchronization pulse from CLK In that sets the clock divider
blocks back to a known state every CLK In cycle.
Page 8 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
1.2.4
Clock Out
A differential output clock (Data CLK) signal is available to act as a reference to clock data into the
device. Data CLK is available on pins CKOUT and CKOUTB. These pins have a nominal output
resistance of 25Ω each, and are designed to drive a bridged load to reduce the effect of package
inductance. The output waveform will be a square wave.
The Data CLK output frequency (F
bus D[15:0].
) will always be matched to the sample rate of the input data
DATA
1.3 Interpolating Filters
The interpolating filters are configured as two cascaded, independently selectable low-pass stages.
The first filter being slower with sharp roll-off, and the second faster but with relaxed roll-off. It is
important to note that when the interpolation stages are not selected they are not clocked,
significantly reducing the power consumption for either x2 or x1 modes, compared to x4.
Table 5: Interpolating Filter Configuration
Filter Configuration
Reduction in Q noise
FILTF
(Fast filter)
FILTS
(Slow filter)
Noise Shaping
Noise Shaping
enabled
Mode
disabled
x1
x2 slow
x2 fast
x4
0
0
1
1
0
1
0
1
N/A
3dB
3dB
6dB
N/A
5dB
5dB
22dB
Further information on the interpolating filters, including frequency response, is given in Section 2.
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 9 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
1.4 Programmable Dither
Dither can be added to improve low-level performance and reduce effects due to nonlinearities within
the DAC, and reducing DNL and glitch energy. The dither has programmable amplitude, see Table
6, and is high pass filtered to fall out of the pass band.
Table 6: Programmable Dither
Dither Setting
DITH1
Dither Amplitude
DITH2
DITH0
rms
Disabled - no dither added
-33.6 dBFS (480 LSBs ) -27.0 dBFS (1458 LSBs )
peak
0
0
0
0
0
1
16
16
0
0
1
1
1
1
0
0
0
1
0
1
-27.6 dBFS (960 LSBs ) -21.0 dBFS (2916 LSBs )
16 16
-21.6 dBFS (1920 LSBs ) -15.0 dBFS (5832 LSBs )
16
16
-15.6 dBFS (3840 LSBs ) -9.0 dBFS (11664 LSBs )
16
16
-9.6 dBFS (7680 LSBs ) -3.0 dBFS (23328 LSBs )
16
16
1
1
1
1
0
1
Reserved for factory use only
Reserved for factory use only
The frequency characteristics of the added dither is illustrated in Section 3.
1.5 Noise Shaping
Second order noise shaping can be applied to interpolated data prior to being passed to the DAC
core. When enabled this provides an additional reduction in quantisation noise to that gained through
the use of interpolation filtering. For the x4 interpolation mode this improvement will be 16dB,
equivalent to 2.7 bits.
1.6 Converter Architecture
The MB86060 interpolating DAC incorporates a number of novel design aspects that are subject to
patent applications. Key to its operation are the current sources where segmented, common centroid,
interleaved techniques for the most significant bits, as well as load matching ensure good linearity
and low distortion to at least the 12-bit level. In the switch elements tracking capacitance is minimised
to improve settling, while controlled rise and fall times improve SFDR performance. Finally the digital
decoding uses a 3-dimensional addressing approach to minimise propagation delays from latch to
element.
Page 10 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
1.6.1
Segment Shuffling
The DAC core incorporates a proprietary segment shuffling capability which is provided to further
improve linearity, and hence improve SFDR. This feature reduces any signal level dependent effects
on linearity as the same code can be generated by the same number of MSB cells but taken from
any quarter of the MSB segments. Segment shuffling can be selected to operate every 4, 8 or 16
updates of the DAC output using a random shuffle sequence between the four segments. Most
performance improvement will be observed when the device is used in one of the interpolating
modes. The effect of segment shuffling is to produce a spread noise spectrum, raising the overall
noise floor, but reducing the distortion. For minimum distortion when generating low frequency
signals, it is recommended that the shuffling clock rate is no more than 25MHz (F
/ Segment
DAC
Shuffling setting). See Table 7. However, low shuffle clock rates give reduced spreading out of
distortion components.
Table 7: Segment Shuffling Control
Mode
SHUF1
SHUF0
Segment Shuffling
Disabled
Note
0
1
2
3
0
0
1
1
0
1
0
1
Lowest noise
DAC ≤ 100 MSa/s
F
Random - every 4 cycles
Random - every 8 cycles
Random - every 16 cycles
100 ≤ FDAC ≤ 200 MSa/s
200 ≤ F MSa/s
DAC
1.6.2
Converter Overload
Within the front end digital processing there is no automatic protection against converter overload
except for clipping at 12-bit FSD. Warning of 12-bit overload at the input to the DAC is indicated by
a high logic level on the ‘OVER’ status pin. Conditions where care must be taken to avoid problems
due to overload would include input signal level when high levels of dither is selected, and fast edge
input data where inevitable overshoot in the digital filters occurs.
1.7 Voltage Reference
A 1.25V bandgap reference is provided on-chip, although this may be bypassed where an external
reference is to be used. To use the internal bandgap reference pins BGAP and VREF should be
linked via a 50Ω resistor, or smaller if better rejection of reference noise at low frequencies is
required. VREF should be decoupled to Reference Ground (RVSS) with a 100nF capacitor. For
maximum accuracy an external voltage reference is recommended
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 11 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
1.8 Analog Output
The DAC output is a differential current type. A termination resistor should be used appropriate for
the maximum allowable output swing. A power down control places the analog circuitry in a low
power state, switching off the current output drive and reference circuitry. When power down mode
is selected the device enters its reset state setting the input data code to +1/2 in 2’s complement or
0 in unsigned binary. Note that neither IOUT or IOUTB can output zero current.
Table 8: Full Scale Code Representation
Unsigned Binary
2’s Complement
Code
IOUT
IOUTB Effective
Code
IOUT
IOUTB
1023
1
1023
65535
:
1111 1111 1111 1111
1
0111 1111 1111 1111
1
64
/
+2047 /
64
/
1024
1024
2
:
:
:
:
:
:
:
1
1022
1
1
1022
32769 1000 0000 0000 0001
32768 1000 0000 0000 0000
32767 0111 1111 1111 1111
32766 0111 1111 1111 1110
0000 0000 0000 0001
33 /
32
32
/
+1 /
33 /
32
32
/
1024
1024
1024
2
1024
1023
1
1023
33
0000 0000 0000 0000
1111 1111 1111 1111
1111 1111 1111 1110
33
/
+ /
/
1024
1024
2
1023
1022
1
1023
1022
33
33
32
32
/
- /
32
32
/
1024
1024
2
1
1
1
/
33 /
:
-1 /
/
33 /
:
1024
1024
2
1024
1024
:
:
:
:
:
:
1023
1
1023
0
0000 0000 0000 0000
1
1000 0000 0000 0000
1
64
/
-2047 /
64
/
1024
1024
2
Output values are expressed as proportions of I /4
ref
1.8.1
Analog Output Reference Resistor
From the voltage reference a control loop defines the current through an external resistor, Rref,
where the current in the reference resistor is 4 times the internal segment current, and the full scale
output current is defined as,
63
64
Vref
4 × Rref
IOP = 63----- × -------------------- ≈ 16 Iref
therefore,
16 × Vref
Rref = ---------------------
IOP
e.g.Using a 1.25V V , to give a 20mA full scale output => R = 1kΩ
ref.
ref
Page 12 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
1.8.2
Analog Output Scaling
Power savings can be made by reducing the full scale analog output current (I ) by increasing R .
OP
ref
However, to maintain the specified performance, I
should be set to 20mA, and the digital data
OP
should be pre-scaled to achieve full scale deflection at an output current lower than full scale (I ).
OP
1.8.3
Analog Output Pins
The analog outputs, IOUT and IOUTB, are each connected to two pins to reduce output inductance.
These pins should be directly connected together on the PCB.
CLK In
tpd
Analog Out
Figure 3 CLK In to Analog Out Timing
1.9 Digital Data Interface
16-bit data is input through pins D[15:0]. D15 is the MSB. Data may be presented in either Unsigned
Binary or 2’s Complement format, depending upon the setting of the TWOC pin. See Table 9.
Table 9: Digital Data Format Control Pin Function
TWOC
Digital Data Format
0
1
Unsigned Binary
2’s Complement
The Digital Data interface has CMOS inputs. The voltage levels of the input data must not exceed
the specifications in section 5.2. Data from a 5V source must not be presented directly to the Digital
Data Interface.
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 13 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
1.9.1
Data Timing
Data should be clocked into the device with the rising edge of the Data CLK signal. The timing
relationship between the rising edge of Data CLK and the setup and hold times for Data In, forms an
‘eye’ opening, within which data may be presented to the Digital Data Interface. If data is presented
outside of this ‘eye’, significant distortion will occur. See Figure 4.
Data CLK
tdatasu
Data In
tdatah
Figure 4 Data CLK to Data In Timing
1.10 Power Supplies
The MB86060 features separate power and ground supplies for digital data, digital control, analog,
reference and clock circuits. A low jitter supply, free from data dependent signals is required by all
power supply domains. The analog domain requires a supply with low clock and data noise.
All domains are all implemented using Fujitsu’s Triple-Well extension to the standard CMOS process
to provide the necessary electrical isolation. This isolation makes power supply sequencing un-
necessary.
1.10.1 Substrate Connections
Connections to the analog, digital interface, and clock section substrates are provided. These pins
would typically be directly connected to the main digital ground (VSS), so as to direct any noise that
has been collected by the substrates away from the analog blocks.
1.10.2 Power Dissipation
The power dissipation, P , is dependant on specific operating conditions: supply voltage (V ), full
D
DD
scale output current (I ), DAC output update rate (F
) and input data waveform. Equations for
OP
DAC
calculating power dissipation in certain conditions are given in section 5.3.
Depending on these factors, applications requiring high F frequencies and/or extended lifetime
DAC
o
at ambient temperatures > 70 C may need additional cooling.
Page 14 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
1.10.3 Pinout
The MB86060 features a performance enhanced pinout to gain the maximum performance from the
PCB. Ground connection pins are provided adjacent to clock in, clock out and analog out pins to
minimise the loop inductance of the return current path. All critical power supplies are paired on
adjacent pins to minimise the decoupling loop inductance, and small value decoupling is provided
on-chip. Discrete decoupling, typically 100nF, must be provided for each power supply pin pair. See
Section 4.
1.11 Reset
A RESETB pin is provided, which when taken low allows the device to be reset and placed in a low
power state. There is a two cycle latency requiring the device to be clocked in order to reset the
device. On power up the device must be reset before it is operational. Multiple device clock
synchronization (see section 1.2.3), and configuration changes require a device reset to be
performed.
There are two reset modes available determined by the state of the TWOC pin. If TWOC is held low
while RESETB is taken low then a PARTIAL-RESET is performed. This will reset and place in a low
power state all sections of the device except the Clock Multiplier and Voltage Reference sections. If
TWOC is held high while RESETB is taken low, a FULL-RESET is performed. This will reset and
place in a low power state all sections of the device.
Table 10: Reset Modes
TWOC
RESETB
Function
Sections Reset
All sections except Clock Multiplier and
Voltage Reference
0
1 > 0
PARTIAL-RESET
1
1 > 0
FULL-RESET
All sections
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 15 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
2
Interpolating Filters
The integration of interpolating filters provides a number of benefits to the system implementation. In
general, improved performance can be gained by using a higher DAC conversion rate effectively
providing a higher level of oversampling from the generated signal. For the designer, the problem
with this approach is generating the required high speed digital data, especially when considering
high performance wide-band designs with up to 50MHz of signal. Integrating this processing on-chip
with the DAC alleviates this problem for the designer.
Other benefits include a reduced effect due to the sinx/x roll-off due to the DAC S&H output stage,
which for a conventional DAC represents -4dB at Nyquist, compared to only -0.22dB when operating
in the x4 interpolating mode. Also the digital interpolation filters sharp cutoff and effective stop-band
attentuation improves both in and out-of-band SFDR. This is illustrated in figure 5.
Conventional DAC
-4dB
-13dB
Sinx
x
-18dB
0
50
100
200
300
400
500
FDAT = FDAC
MHz
* Reduced data generation rate while maintaining high DAC rate
* Integral digital interpolating filters
4x Interpolating DAC
* Relaxed analog reconstruction filter
* Reduced effect due to Sinx/x function
* Improves both In and Out-of-band SFDR
Sinx
x
-0.22dB
-4dB
0
50
100
FDAT
200
300
400
FDAC
500
MHz
Figure 5 Benefits of Interpolating Filters
The MB86060 features four interpolation filter modes x1, x2(slow), x2(fast) and x4. x1 is as per a
conventional DAC, and choosing between the remaining three modes would depend on the system
requirements. x2(slow) may be advantageous to a system requiring the benefits of interpolation
filtering but saving some power by not running the DAC core at full rate. x2(fast) gives access to the
wider band, slower roll-off interpolation filter allowing wider band signals to be generated compared
to the other modes, for example 74MHz (-0.1dB) for 200MSa/s data rate. x4 for the complete
interpolation filter operation. These different modes are illustrated in Figure 6.
Page 16 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
-4dB
Sinx
x
-13dB
-18dB
x1
mode
0
100
DAT & FDAC
200
200
300
400
400
500 MHz
F
Filter Pass Band = 0.43.FDAT (-0.1dB)
Sinx
x
-0.9dB
X2 (slow)
mode
-13dB
0
100
300
500 MHz
FDAT
FDAC
Filter Pass Band = 0.37.FDAT (-0.1dB)
Sinx
x
X2 (fast)
mode
-0.9dB
0
100
200
300
400
500 MHz
FDAT
FDAC
Filter Pass Band = 0.43.FDAT (-0.1dB)
Sinx
x
x4
mode
-0.22dB
0
100
FDAT
200
300
400
FDAC
500 MHz
Figure 6 Interpolating Filter Modes
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 17 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
Simulated slow and fast x2 interpolation filter responses
20
x2 Slow Filter
Pass Band 0.43fs (-0.1dB)
0
Stop Band -75dBFS from 0.59fs
[note: Frequency axis normalised to
input data rate]
-20
-40
-60
-80
dB
x2 Fast Filter
Pass Band 0.74fs (-0.1dB)
Stop Band -83dBFS from 1.54fs
[note: Frequency axis normalised to
input data rate for x4 interpolation
mode. With only x2 Fast selected then
the input data rate is normalised to 2.0
Frequency]
-100
-120
0
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Frequency
Figure 7 Slow & Fast Filter Characteristics
Simulated overall x4 interpolation filter response
20
0
Combined Filters, x4 Mode
Pass Band 0.43fs (-0.1dB)
Stop Band -75dBFS from 0.59fs -
(excluding transition band at around
1.5fs)
-20
-40
-60
-80
dB
[note: Frequency axis normalised to
input data rate]
-100
-120
0
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Frequency
Figure 8 Combined Filter Characteristics
Page 18 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
CLK In
14 x CLK In cycles
D15
IOUT
Data CLK
A: x1 Interpolation Filter, Clock Multiplier Bypassed
CLK In
49 x CLK In cycles
D15
IOUT
Data CLK
B: x2 slow Interpolation Filter, Clock Multiplier Bypassed
CLK In
39 x CLK In cycles
D15
IOUT
Data CLK
C: x2 fast Interpolation Filter, Clock Multiplier Bypassed
CLK In
110 x CLK In cycles
D15
IOUT
Data CLK
D: x4 Interpolation Filter, Clock Multiplier Bypassed
Figure 9 Pipeline Delays
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 19 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
3
Dither Frequency Spectrum
The use of dither in data converter applications is not uncommon, where improvements in low-level
performance and reduced effects due to nonlinearities can be achieved. For dither to be used
effectively both amplitude and frequency characteristics must be carefully considered. Obviously the
dither amplitude should be larger than the nonlinearities to be masked, but levels significantly larger
than this will ultimately limit available dynamic range for the wanted signal. Similar considerations
should be made for the frequency characteristics, which in the MB86060 the dither is highpass
filtered such that the majority of the energy is concentrated at Nyquist of the DAC output rate.
In many systems a simple calculation can be used to determine the maximum input signal level for
a given dither amplitude, and in most applications this applies. However, in multi-tone systems such
as discrete multi-tone (DMT) or multi-channel communication systems a more statistical approach
may be adopted where the probability of converter overloads occurring is considered.
The frequency characteristics of the highpass filtered dither is shown in Figure 10.
Dither highpass filter response
0
-10
-20
-30
-40
-50
-60
dB
-70
-80
-90
-100
0
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Frequency
Figure 10 Dither Frequency Characteristics
Page 20 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
4
Application Notes
4.1 Power & Ground Plane Regions
The following guidelines are suggested to obtain the specified performance. Any departure from
these recommendations should be investigated to confirm that performance in the application is
acceptable.
The device should be used with a PCB utilizing a minimum of four layers for separate power and
ground planes, manufactured with tolerances capable of producing exact impedance tracking. When
using a four layer board critical analog signals should be routed on the external layer adjacent to the
ground plane (typically layer 1). The power and ground planes should be split to isolate digital, clock
and analog regions of the circuitry to prevent supply noise coupling from one to another. These
separated regions should only be connected together at one place, a star point located underneath
the device, which should also be used as a connection point to the PSU. These isolated regions
should only extend across the PCB as far as necessary, and avoid other sections of the application
circuit that could introduce noise. Signal regions such as the Analog Out and Clock In/Out can be
separated from the remainder of the application circuit by introducing a transformer as an isolator.
The connection to the PSU should also be arranged as a star point, with all other sections of the
application circuit joined at this point. Tracks to this point should be made as wide as possible, and
if they are located in the ground plane layer, should be positioned under static pins. No connection
to the supply tracks should be made midway. See figure 11.
DVSS
VSS
CVSS
PSU Star Point
Device Star Point
Application
Circuit
AVSS
PSU
RVSS
Figure 11 Power Supply Distribution Through Star Points
The DVDD/DVSS and VDD/VSS pins can be connected to the same region, but normally the Digital
supply and ground plane regions should be split further to isolate the Digital control and Digital data
blocks. The Digital data region will normally extend into the application circuit, and as such will be
subject to significant noise from the data source.
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 21 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
Another significant reason for splitting the digital data ground plane from the main digital ground is
that when using a remote data generator (e.g. a benchtop pattern or data generator) there is a
tendency for noise to be injected on the data ground by the equipment. This is significant because
the data rate is high, the data bus is wide, and it’s correlation with the signal can cause spurious tones
which degrade SFDR. The coupling mechanism is from the fast-slewing data inputs via the
capacitance of the input pins/pads/protection diodes into the internal circuits. The transient currents
through these parasitics can be several hundred milliamps. For this reason, it is recommended that
this region is not connected to the device star point but to the PSU star point directly.
The data supplies are only used for the input section of the device, so noise in this region cannot
couple into the DAC core. The digital supplies connect to the digital circuits (filters, noise shaper and
ditherer) inside the DAC, including those inside the DAC core. The control inputs can use this supply
because they toggle more slowly (if at all) and aren't correlated with the data. The analog sections
(Analog, Clock and Reference) have separate supply connections as transition dependant currents
from the digital sections will cause delay modulation in the clock path, and amplitude modulation in
the analog output section.
Each supply should be decoupled, producing a low impedance shunt at high frequency. The Digital,
Analog, Clock and Reference sections can be connected directly to the device star point, but
preferably through a small inductor. If a fully split power (VDD) plane is not desired, then as a
minimum only the ground plane need be split as described. However it is very important to isolate the
I/O supply (DVDD) from all other supplies in some way, possibly feeding the supply through a low-R
resistor or ferrite bead. This will help to filter out noise.
If the data (signal) and control lines are connected to the same device (e.g. an ASIC or FPGA), then
generally this should have been designed to support separate supply and ground pins for the digital
data bus. The ground plane at the generating device (ASIC or FPGA) then becomes the star point
for the data, requiring cuts in the ground/supply planes on either side of the data bus, and looping
under the DAC. The digital data decoupling at the DAC should also be inside this loop. This gives a
"U" shaped cut in the planes with the open end at the data source (with decoupling) and the closed
end at the DAC (with decoupling). All the data return currents will then be confined inside this "U",
and so none of them can couple into the analog ground planes to degrade SFDR. It may be advisable
to bury the digital data bus tracks on an internal layer, with data ground planes above, below and
either side of the tracks (the ground layers connected together with a “picket fence” row of vias) to
shield against RF radiation.
Figure 12 shows these principles applied to the ground plane of an application board. The pad on the
left represents the PSU star point, and the pad in the center represents the device star point. These
points could be realized with a via, so that the connection from the PSU out to other star points could
be made on another layer if necessary. The positioning of the plane breaks are also shown. The
breaks in the planes between each section should mark the boundary of that section. It is very
important to ensure that there are no tracks crossing these boundaries, or any splits in the planes
that tracks must cross, as this will create current loops within the plane itself.
The power supply track region is shown extending to the side of the device for reference purposes.
If the PSU region is not required, then the region should be merged with the Digital region.
Page 22 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
DATA REGION
DIGITAL REGION
DIGITAL REGION
PSU Star point
Pin #1
CLOCK REGION
REFERENCE REGION
ANALOG REGION
Figure 12 Recommended Ground Plane Splits
4.2 Power Supplies
Only one clean low-impedance power supply is required. Power distribution should be organized as
shown in Figure 11, with a main star point at the PSU supplying the data (DVDD/DVSS) block, with
a secondary device star point supplying the Digital, Analog, Clock and Reference blocks. If this
supply is used to supply any other circuits, they must not introduce any modulation onto the supply,
or SFDR will be degraded.
If the impedance of the supply is not low enough to prevent modulation by the currents drawn by the
data source, then a separate supply for the Data source should be used. If the Digital, Analog, Clock
and Reference block supply is still not low enough impedance to prevent power supply modulation
being introduced by the Digital block, then a further supply for the Analog block alone must be
introduced.
Bulk decoupling of 100uF or more at the power supply star point is recommended to remove any low
frequency ripple. Smaller value decoupling of around 10uF at the device star point is recommended
to apply a low impedance shunt at high frequency. High quality, very low ESR capacitors such as
solid aluminium types are recommended.
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 23 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
Power supply tracks should be kept as short and wide as possible. The ground and supply tracks
should run adjacent to each other for as far as possible, whilst avoiding all signal tracks that may
couple noise into the supply.
4.3 Decoupling
All supplies and references should be decoupled to the appropriate ground plane using surface
mount 100nF capacitors, placed as close as possible to the device. For each pair of VDD/VSS pins
it is recommended that the capacitor is located on the reverse of the PCB, immediately under the
device, with vias to the supply and ground planes as close as physically possible to the device and
capacitor. This layout minimizes the total length of track, including the plated through hole, and hence
keeps loop inductance to a minimum. An example of the recommended layout, using 0603 format
surface mount capacitors and a four layer PCB is illustrated in figure 13.
Layer 1 pads
MUL0
MUL1
MUL2
VSS
DITH0
DITH1
DITH2
VSS
VDD
VDD
CKOUT
CKOUTB
VSS
TWOC
N/C
VSS
VDD
VDD
DSUB
LOCK
CSUB
NSHAPE
CLKSEL
RESETB
CVDD
CVSS
CLK
DSUB
OVER
ASUB
FILTS
FILTF
SHUF0
SHUF1
RVSS
BGAP
VREF
RREF
CVSS
CLKB
Pin #1
Layer 4 pads
N.B. Not to scale. All vias connect to the appropriate Ground or Power plane.
Figure 13 Recommended Supply Decoupling Layout
Page 24 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
4.4 Input Data Interfacing
The data input interface is a 16-bit wide CMOS bus, operating at speeds up to 200MHz. As such this
bus can radiate large amounts of RF radiation and generate substantial ground noise problems if it
is routed for any distance across the PCB. In addition to following the PCB layout guidelines in
section 4.1, it may be possible to minimise these problems by transmitting data across the PCB as
LVDS, and only converting to CMOS locally to the DAC.
LVDS is a low-voltage differential signalling format that is becoming increasingly popular for data
transmission, and as such is now well supported by the majority of programmable logic devices. It
has a signal swing of 350mV, and a common mode level of 1.25V. As the signal is transmitted as a
differential signal, immunity to noise pick-up is very high, radiation is minimal, and a direct ground
connection is not required between transmitter and receiver.
Four 4-bit LVDS to CMOS receivers (available from a number of manufacturers) could be used to
interface the MB86060’s 16-bit data interface to an LVDS bus. Many receivers have a flow through
pin-out design, with the power supply pins and CMOS data pins on one side, and the LVDS pins on
the other side of the package. This design makes PCB ground plane layout especially easy as the
data generator and DAC PCB regions can be kept some distance apart. The data generating device
ground plane can extend to just under the LVDS pins on one side of the converters package (to
maintain track impedance of 100Ω), but no further so as not to merge or connect to the DAC’s CMOS
ground plane on the other side. The CMOS data signal tracks from the converter to the DAC can now
be limited to just a few tens of millimetres. The converters 3.3V supply would be connected to the
DAC’s DVDD/DVSS supply region. See figure 14.
DVDD
DVDD
D0
100Ω
D0
D0
GND
FPGA
DVSS
LVDS to CMOS
DVSS
MB86060
System Ground
Figure 14 Using LVDS to CMOS Converters
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 25 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
4.5 Analog Output
To provide a differential analog output which is both isolated form the analog ground plane, and which
gives good common mode rejection, a two stage transformer circuit can be used. The recommended
devices are Mini-Circuits (www.minicircuits.com) ADTT1-1, 1:1 transformer, and ADTL1-4-75
transmission line transformer. The primary of the ADTL1-4-75 is connected to IOUT and IOUTB,
(terminated as shown in figure 15) and the secondary is connected to the ends of the secondary of
the ADTT1-1. The center tapping of the secondary of the ADTT1-1 will be linked to the analog ground
plane. The primary of the ADTT1-1 will be terminated as required by the application circuit. See figure
15.
For optimum performance the transformer should be positioned as close to the device as physically
possible, and should be connected to the analog output pins IOUT and IOUTB with 50Ω tracks. The
connections to the analog ground plane should be made through the same vias as the decoupling
capacitors, which are shown in figure 13 as being on the outside of the device pad pattern, so that
the track length and hence loop inductance can be kept to a minimum.
IOUT
ADTT1-1
ADTL1-4-75
50Ω
50Ω*
100Ω
AVSS
Analog out
AGND
50Ω
*
22pF
IOUTB
AGND
MB86060
* 0.1% Precision resistors
Figure 15 Analog Out Transformer Coupling
4.6 Clock Input
The reference input clock can be connected in a number of ways, depending on the clock source
used. For optimum performance the MB86060 should have ground isolation from the source by the
use of a coupling transformer.
The simplest method is to use the internal oscillator, with a crystal connected across XIN and XOUT.
A 1MΩ resistor should be connected in parallel with the crystal, and capacitors (typically 22pF,
depending on crystal used) should be connected from both XIN and XOUT to the clock ground
region. See figure 16a.
A single ended CMOS crystal oscillator can be used, connected to XIN. This source can be
connected to the clock ground region as long as there are no other devices connected to the clock
signal, i.e. the oscillator has a fan-out of one. See figure 16b.
Page 26 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
If a common, system wide logic level clock source is to be used, this should be transformer coupled
to remove common mode noise and isolate the clock ground region. The recommended 1:1
impedance transformer is a Mini-Circuits, ADT1-1. The secondary will be connected to the differential
clock inputs CLK and CLKB, and terminated with a 100Ω resistor. The PCB tracks to the device
should be 50Ω tracks. The primary of the transformer will be connected between the digital clock
source and the digital clock source ground. A 100Ω source resistor is used, and the PCB tracks to
the clock source should be 100Ω. See figure 16c.
For connection to a low noise RF source, a Balun transformer should be used. The recommended
transformer is a Mini-Circuits ADTL1-4-75. The secondary terminals should be connected through a
100nF capacitor to the differential clock inputs CLK and CLKB. The clock inputs should be terminated
with a 50Ω resistor. The primary dot should be connected to the RF signal with a 50Ω PCB track, and
the other primary connection should be connected to RF ground. The primary ground may be
connected to clock ground if necessary. See figure 16d. This configuration would allow for a RF
signal level of between -6dBm and +24dBm, giving a differential signal level of up to 5V pk-pk at CLK
and CLKB.
XIN
CLKOUT
XIN
22pF
OSC
MB86060
1MΩ
XTAL
MB86060
GND
CGND
22pF
XOUT
CVSS
CGND
B: Using a dedicated CMOS oscillator
A: Using the internal crystal oscillator
100nF
CLK
ADT1-1
CLK
RF Source
100Ω
System clock
50Ω
100Ω
MB86060
CLKB
ADTL1-4-75
MB86060
CLKB
System Ground
100nF
RF Ground
C: Using a common system clock source
D: Using an RF source
Figure 16 Clock Input Configurations
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 27 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
4.7 Clock Output
The data clock output is provided as a reference to clock the data into the device. As with the analog
output and the reference input it is recommended that the data clock is isolated from the application
circuit to remove common mode noise in the digital ground plane.
The output pins CKOUT and CKOUTB are designed to drive a bridged load to reduce the effect of
package inductance, and each pin has a nominal output resistance of 25Ω.
A transformer can again be used, the recommended device being a Mini-Circuits ADT4-1WT. The
ends of the secondary should be connected to CKOUT and CKOUTB through series 75Ω resistors,
and 100Ω PCB tracks, and the centre tapping should be left not connected. The primary dot end
should be connected to the wiper of a 100Ω variable resistor to provide an adjustable bias point for
the output signal. The variable resistor should be placed across the supply and ground rails of the
application circuit, and a 100nF capacitor placed from the wiper to application ground. (If the biasing
is not required connect the primary dot to ground). The data clock is then present at the other end of
the primary. A 50Ω PCB track should be used here. See figure 17.
App. Supply
75Ω
ADT4-1WT
100nF
CKOUT
100Ω
MB86060
App. Ground
75Ω
CKOUTB
Data clock out
Figure 17 Data Clock Out Transformer Coupling
Page 28 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
4.8 Signal Routing
Where signals have to be tracked as differential pairs, attention should be paid to the routing and
positioning of these tracks. Wherever possible, differential tracks should be routed parallel to each
other, and kept as close together as possible for the maximum distance possible. The total length of
each track in a differential pair should be equalised, and the same number of vias should always be
used. Where differential tracks have to diverge to connect to component pins, the distance that the
two tracks are routed separately should be equalised. See figure 18.
Where multiple routing layers are available, differential tracking should be kept on the same layer,
rather than as a stacked pair on adjacent layers. (This may introduce more noise onto one track than
the other due to each track being coupled to different ground planes). Where tracks are routed close
to other signals or noise sources, sufficient separation should be maintained so as to not introduce
significantly more distortion onto one track than the other. Particular attention should be paid to the
positioning of through board vias, which may be exposed to noise from tracking on the layers they
pass through. Switching tracking from one layer to another with vias should be avoided if possible,
but where necessary the vias should be positioned close to each other, within the same plane region
on all layers, and located so as to avoid potential sources of noise on all layers.
Careful attention should be paid to the positioning of signal tracking around plane splits. Signal tracks
must never cross a ground or power plane split, as the return current (which tends to follow the signal
track within the plane) will be forced to follow the plane split around to the star point or until it can find
another way to couple back into the plane on the other side of the split. This can create large ground
current loops within the planes that may interfere with other sections of the application.
One via closer to plane
split than other
Via moved away from plane
split and closer together
Un-equal net lengths
Near-equalised net lengths
A: Poor Differential Tracking Layout
B: Correct Differential Tracking Layout
Figure 18 Differential Tracking Examples
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 29 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
5
Electrical Specifications
5.1 Absolute Maximum Ratings
Ratings
Parameter
Notes Symbol
Units
Min.
Typ.
Max.
Supply voltage
Input Voltage
VDD
VIL
VO
ΙO
3.0
3.3
3.6
VDD+0.5
VDD+0.5
21
V
V
Vss -0.2
Vss - 1.0
Output Voltage
Output Current
1
2
0
V
20
25
mA
oC
Storage Temperature
TST
-40
+125
TOP(min) to TOP(max), VDD = DVDD = RVDD = AVDD = CVDD = +3.3V, VSS = DVSS = RVSS = AVSS = CVSS = 0V,
FS = 20mA, Differential Transformer coupled output, 50Ω doubly terminated, unless otherwise specified.
I
1. IOUT & IOUTB
2. For 1 second per pin (at max.)
CAUTION
ELECTROSTATIC DISCHARGE SENSITIVE DEVICE
High electrostatic charges can accumulate in the human body
and discharge without detection. Ensure proper ESD
procedures are followed when handling this device.
Page 30 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
5.2 Digital Interface Specifications
Ratings
Typ.
Parameter
Notes Symbol
Units
Min.
Max.
CMOS inputs
High-level input voltage
Low-level input voltage
High-level input current
Low-level input current
Input capacitance
Setup time
VIH
VIL
IIH
VDD - 1.0
VSS
VDD
VSS + 1.0
+10
V
V
-10
µA
µA
pF
ns
ns
IIL
-10
+10
5
tdatasu
tdatah
1.0
1.2
Hold time
CMOS outputs
High-level output voltage
Low-level output voltage
Reset timing
VOH
VOL
VDD - 0.4
VSS
VDD
V
V
VSS + 0.4
Setup time
trstsu
trsth
1.0
1.2
1
ns
ns
µs
Hold time
Reset delay, analog out
trstdel
TOP(min) to TOP(max), VDD = DVDD = RVDD = AVDD = CVDD = +3.3V, VSS = DVSS = RVSS = AVSS = CVSS = 0V,
FS = 20mA, Differential Transformer coupled output, 50Ω doubly terminated, unless otherwise specified.
I
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 31 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
5.3 DC Specifications
Ratings
Typ.
Parameter
Notes
Symbol
Units
Min.
Max.
DC Accuracy
Integral Non Linearity, (Shuffle Off)
Differential Non Linearity
Analog output
INL
10
5
20
10
LSB16
LSB16
DNL
Full scale output current
Output resistance
1
2
ΙOP
2
20
100
15
21
mA
kΩ
pF
%FS
V
Output capacitance
Gain error
-1
-1
+1
+1
Output voltage (compliance) - Maximum
Output voltage (compliance) - Best performance
CLK In to Analog Out propagation delay
Output settling time (to 0.1%)
Output rise time (10% to 90%)
Output fall time (90% to 10%)
Bandgap Reference
-0.5
+0.5
V
3
4
5
5
tpd
tst
6
4
ns
ns
1.6
1
ns
ns
Reference voltage
VBG
IBG
1.19
0
1.25
1.25
3.3
1.31
100
V
Reference output current
Reference Input
µA
Reference voltage
VREF
IREF
1.19
-1
1.31
+1
V
Reference input current
Power Supply
µA
VDD, DVDD, RVDD, AVDD, CVDD
3.0
3.6
V
Power Dissipation
Multiplier bypassed, shuffle and dither disabled
100MSa/s input, x1 interpolation
100MSa/s input, x2slow interpolation
100MSa/s input, x4 interpolation
Power down current
6
6
6
PD
PD
PD
204
507
867
<1
mW
mW
mW
mA
oC
Ambient Temperature
TA
TJ
-40
25
+85
Junction Temperature
+110
oC
Page 32 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
TOP(min) to TOP(max), VDD = DVDD = RVDD = AVDD = CVDD = +3.3V, VSS = DVSS = RVSS = AVSS = CVSS = 0V,
I
FS = 20mA, Differential Transformer coupled output, 50Ω doubly terminated, unless otherwise specified
1. Distortion increases when IOP decreases. Set to 20mA for optimum performance.
2. VREF = 1.25V. See section 1.8.1
3. Propagation delay does not include the data pipeline delays. See section 2, figure 9.
4. Measured differentially. IOUT and IOUTB doubly terminated (25Ω load to AVSS).
5. Measured single-ended. IOUT and IOUTB doubly terminated (25Ω load to AVSS).
6. Nominal power dissipation
x1 mode PD = 99 + (105 per 100MSa/s) (mW) approx.
x2 slow mode PD = 99 + (204 per 100MSa/s) (mW) approx.
x2 fast mode PD = 99 + (147 per 100MSa/s) (mW) approx.
x4 mode PD = 99 + (192 per 100MSa/s) (mW) approx.
x4 mode and dither enabled PD = 99 + (245 per 100MSa/s) (mW) approx.
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 33 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
5.4 AC Specifications
Ratings
Typ.
Parameter
Notes
Symbol
Units
Min.
Max.
Signal to Noise Ratio
Range DC to 50MHz, FDAC = 400 MHz, x4 Interpolation mode,
1
SNR
Noise shaping enabled, Dither disabled
2MHz tone, Segment Shuffling - Off
2MHz tone, Segment Shuffling - On
90
80
dB
dB
Total Harmonic Distortion
Range DC to 50MHz, FDAC = 400 MHz, x4 Interpolation mode,
1
THD
Noise shaping enabled, Dither disabled
2MHz tone, Segment Shuffling - Off
2MHz tone, Segment Shuffling - On
20MHz tone, Segment Shuffling - Off
20MHz tone, Segment Shuffling - On
80
90
65
76
dB
dB
dB
dB
Spurious Free Dynamic Range
Single Tone at -1dBFS, FDAC = 200MHz, range DC to 100MHz
1
SFDR
2MHz tone, Segment Shuffling - Off
2MHz tone, Segment Shuffling - On
15MHz tone, Segment Shuffling - Off
15MHz tone, Segment Shuffling - On
83
93
70
83
dBc
dBc
dBc
dBc
65
4-tones at -15dBFS, FDAC = 200MHz, range DC to 100MHz
19.1, 19.3, 19.7, & 19.9MHz tones, missing centre tone
Spurious tone at 19.5MHz, Segment Shuffling - Off
Spurious tone at 19.5MHz, Segment Shuffling - On
Spurious tones 17.5 - 21.5MHz, Segment Shuffling - Off
Spurious tones 17.5 - 21.5MHz, Segment Shuffling - On
88
95
89
97
dBFS
dBFS
dBFS
dBFS
Adjacent Channel Leakage Ratio
4MHz bandwidth, 5MHz channel spacing
1
ACLR
16MHz centre frequency, 64MSa/s, FDAC=256MSa/s
32MHz centre frequency, 128MSa/s, FDAC=256MSa/s
80
74
dBc
dBc
TOP(min) to TOP(max), VDD = DVDD = RVDD = AVDD = CVDD = +3.3V, VSS = DVSS = RVSS = AVSS = CVSS = 0V,
IFS = 20mA, Differential Transformer coupled output, 50Ω doubly terminated, unless otherwise specified
1. Clock multiplier mode = ‘000’ (bypassed)
Spurious Free Dynamic Range (SFDR) is defined as the highest spurious product (harmonic or non-harmonically related) within a defined
bandwidth while generating a test tone or tones (multi-tone test). SFDR varies with amplitude and frequency of the test tone(s) and should
either be quoted as the difference between the tone and highest spurious component (dBc) or referenced to full scale (dBFS). In both cases
the test tone amplitude and frequency should be quoted as well as the measurement bandwidth. The measurement bandwidth is typically
regarded as DC to Nyquist (of the input data rate where interpolating modes are selected) but occasionally systems will specify an
appropriate narrow band.
Adjacent Channel Power Ratio (ACPR) relates to the ratio of power in an adjacent band compared to that in a wanted transmit band, where
channel bandwidth and channel spacing should be quoted.
Page 34 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
5.5 Clock Specifications
Ratings
Typ.
Parameter
Notes
Symbol
Units
Min.
Max.
Maximum DAC Conversion rate
VDD < 3.3V
1
FDAC
350
400
380
416
MSa/s
MSa/s
VDD ≥ 3.3V
Clock multiplier operational range
Minimum input frequency
Maximum input frequency
Maximum input data rate, interpolation modes
x4 interpolation
FMULIN
10
85
MHz
MHz
FDATA
100
100
200
200
104
104
208
208
MSa/s
MSa/s
MSa/s
MSa/s
x2 slow interpolation
x2 fast interpolation
x1 (no interpolation)
2
Clock in
Low time
tLO
tHI
1
1
ns
ns
High time
Slew rate for minimum wide-band jitter
Common mode input voltage
Signal level (differential)
Clock out
tSL
0.5
V/ns
V
3
4
Vcm
VREF
100
mV
CLK In to Data CLK delay
XIN to Data CLK delay
Crystal oscillator speed
tdel
tdelx
Fxtal
7
ns
ns
5
7.2
40
MHz
TOP(min) to TOP(max), VDD = DVDD = RVDD = AVDD = CVDD = +3.3V, VSS = DVSS = RVSS = AVSS = CVSS = 0V,
FS = 20mA, Differential Transformer coupled output, 50Ω doubly terminated, unless otherwise specified
I
1. Assumes x2 fast or x4 interpolation mode is selected
2. Limited by CMOS digital I/O speed
3. Pins CLK and CLKB are internally biased to the voltage applied to pin VREF
4. Ensure that slew rate specifications are observed
5. CMOS clock signal applied to pin XIN. Clock multiplier in x2 mode
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 35 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
5.6 Typical Performance Characterisation Graphs
95
90
85
80
75
70
65
60
Single Tone Spurious Free Dynamic Range
50MSa/s Input Data Rate
x4 Mode, 200MSa/s DAC Rate
Noise Shaper On, Dither Off
Amplitude = -1dBFS
Shuffle On
Shuffle Off
Figure 19 Single Tone SFDR Performance
0
10
20
30
40
Generated Frequency
(MHz)
0
50MSa/s Input Data Rate
Multi-tone test, 4 tones, 200kHz channel
spacing, missing centre tone
x4 Mode, 200MSa/s DAC Rate
Shuffle On & Noise Shaper On
Dither N/A
-20
4 Tones each at -14.3dBFS
-40
-60
-80
-100
-120
17
18
19
20
21
22
Generated Frequency
(MHz)
Figure 20 Multi-tone Performance
W-CDMA direct-IF channel generation
example
4MHz channel, 5MHz channel spacing
16MHz channel centre frequency
F
F
= 64MSa/s, x4 Interpolation
= 256MSa/s
DATA
DAC
Figure 21 W-CDMA Carrier Direct-IF Gen-
eration
Page 36 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
6
Pin Description
6.1 Pin Assignment
DITH0
DITH1
DITH2
VSS
MUL0
MUL1
MUL2
VSS
VDD
VDD
CKOUT
CKOUTB
VSS
TWOC
N/C
VSS
VDD
VDD
DSUB
DSUB
LOCK
CSUB
NSHAPE
CLKSEL
RESETB
CVDD
CVSS
CLK
MB86060
OVER
ASUB
FILTS
FILTF
SHUF0
SHUF1
RVSS
BGAP
VREF
RREF
CVSS
CLKB
Pin #1
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 37 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
6.2 Pin Definition
Data Interface
Input/
Output
Pin Nos.
Pin Name
Description
†
60
59
D0
D1
I
Input data bit 0 (LSB)
†
I
Input data bit 1
57
D2
I
Input data bit 2
Input data bit 3
Input data bit 4
Input data bit 5
Input data bit 6
Input data bit 7
Input data bit 8
Input data bit 9
Input data bit 10
Input data bit 11
Input data bit 12
Input data bit 13
Input data bit 14
56
D3
I
55
D4
I
54
D5
I
52
D6
I
51
D7
I
50
D8
I
49
D9
I
47
D10
D11
D12
D13
D14
D15
DVDD
DVSS
DSUB
I
46
I
45
I
44
I
42
I
41
I
Input data bit 15 (MSB)
43, 53
48, 58
31, 70
Power
Power
Power
Data interface supply, +3.3V. Decouple to DVSS
Data interface ground, 0V
Data interface substrate. Link to VSS
† Connect unused LSB inputs to DVSS, e.g. D0 & D1 for 14-bit source data
Page 38 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
Digital Control Interface
Input/
Output
Pin Nos.
Pin Name
Description
61
DITH0
DITH1
DITH2
FILTS
FILTF
I
Programmable Dither control bit 0. See table 6
Programmable Dither control bit 1. See table 6
Programmable Dither control bit 2. See table 6
Slow Interpolation Filter control. See table 5
Fast Interpolation Filter control. See table 5
Noise shaper, ‘1’ = enabled. See section 1.5
Segment Shuffling control bit 0. See table 7.
Segment Shuffling control bit 1. See table 7.
Input data format selection, ‘1’ = 2’s Complement
Digital overflow warning. ‘1’ = overflow condition
Clock Select. Differential clock =’1‘, Crystal clock via XIN = ‘0’
Delay line control bit 0. Internal pull-down resistor. See table 4.
Delay line control bit 1. Internal pull-down resistor. See table 4.
Clock multiplier mode control bit 2. See table 4.
Clock multiplier mode control bit 1. See table 4
Clock multiplier mode control bit 0. See table 4
DLL locked indicator, ’1’ = locked
62
I
63
I
73
I
74
I
28
NSHAPE
SHUF0
SHUF1
TWOC
OVER
CLKSEL
OSC0
OSC1
MUL2
MUL1
MUL0
LOCK
RESETB
N/C
I
75
I
76
I
66
I
71
O
27
I
3
I
4
I
38
I
39
I
40
I
30
26
O
I
Device reset, ’0’ = reset
67
-
No connection. Do not connect
32, 36, 65, 69
33, 37, 64, 68
VDD
Power
Power
Digital supply, +3.3V. Decouple to VSS
VSS
Digital ground, 0V
Clock Interface
Input/
Output
Pin Nos.
Pin Name
Description
21
23
CLKB
CLK
I
Differential input clock (CLK In), negative input.
Differential input clock (CLK In), positive input.
Differential data output clock (Data CLK), negative output.
Differential data output clock (Data CLK), positive output.
Crystal / clock input. Connect to CVSS when not used
Crystal oscillator circuit output
I
34
CKOUTB
CKOUT
XIN
O
35
O
17
I
O
18
XOUT
CVDD
CVSS
CSUB
19, 25
20, 22, 24
29
Power
Power
-
Clock supply, +3.3V. Decouple to CVSS.
Clock ground, 0V
Clock substrate. Link to VSS
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 39 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
Analog Interface
Input/
Output
Pin Nos.
Pin Name
Description
7,8
13,14
IOUTB
IOUT
O
O
DAC current output, negative output. Connect together.
DAC current output, positive output. Connect together.
Analog supply, +3.3V. Decouple to AVSS.
Analog ground, 0V
5, 10, 11, 16
6, 9, 12, 15
72
AVDD
AVSS
ASUB
Power
Power
-
Analog substrate. Link to VSS
Reference Interface
Input/
Output
Pin Nos.
Pin Name
Description
78
79
BGAP
VREF
RREF
RVDD
RVSS
O
I
Bandgap reference. See section 1.7.
Voltage reference input. See section 1.7.
Output reference resistor. See section 1.8.
Reference supply, +3.3V. Decouple to RVSS.
Reference ground, 0V
80
O
2
Power
Power
1, 77
Page 40 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
6.3 Package Data
80-pin plastic LQFP
(FPT-80P-M21)
Note 1) * : These dimensions do not include resin protrusion.
Note 2) Pins width and pins thickness include plating thickness.
Note 3) Pins width do not include tie bar cutting remainder.
14.00±0.20(.551±.008)SQ
*
12.00±0.10(.472±.004)SQ
0.145±0.055
(.006±.002)
60
41
61
40
0.08(.003)
Details of "A" part
1.50 –+00..1200
(Mounting height)
.059 –+..000048
0.10±0.10
(.004±.004)
(Stand off)
INDEX
0˚~8˚
80
21
"A"
0.25(.010)
0.50±0.20
(.020±.008)
0.60±0.15
(.024±.006)
1
20
LEAD No.
0.50(.020)
0.20±0.05
(.008±.002)
M
0.08(.003)
Dimensions in mm (inches).
Note: The values in parentheses are reference values
C
2006 FUJITSU LIMITED F80035S-c-1-1
6.3.1
Thermal Characteristics
o
o
• θJA = 40.4 C/W, θJC = 8 C/W
Figures assume mounting on a 4-layer pcb mounted in free air.
6.4 Ordering Information
The following reference should be used when ordering devices,
• MB86060PMCR-G-BNDE1
The device is RoHS-6 compliant, qualified to Moisture Sensitivity Level (MSL) 3 and a Peak Reflow
o
temperature of 260 C.
Shipment is in plastic trays, each capable of holding 119 devices.
For further assistance please contact your sales representative.
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 41 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
7
Development Kit
A development kit is available for the MB86060 16-bit Interpolating DAC. The kit includes an
evaluation board that enables simple and effective evaluation of the device.
The board provides a complete evaluation environment for the DAC. A transformer coupled
differential output interface is provided to simplify integration into target applications and
development environments. An RF clock source can be connected via the transformer coupled input,
and 16-bit data via a 40-way IDC header or optionally (customer modification) SMA/SMB connectors.
The development kit includes,
• Evaluation board with MB86060 device fitted
• Spare MB86060 for customer development
• User Manual
For further assistance, including price and delivery of the development kit, please contact your sales
representative.
7.1 Ordering Information
The following reference should be used when ordering the customer development kit,
• DK86060-3
Page 42 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
Notes:
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Production
Page 43 of 44
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
February 2009 Version 2.0
FME/MS/SFDAC1/DS/4250
MB86060 16-bit Interpolating Digital to Analog Converter
Worldwide Headquarters
Japan
Asia
Tel: +81 3 5322 3347
Fax: +81 3 5322 3387
Tel: +65 281 0770
Fax: +65 281 0220
Fujitsu Microelectronics Limited
Shinjuku Dai-Ichi Seimei Bldg.
7-1, Nishishinjuku 2-chome
Shinjuku-ku
Fujitsu Microelectronics Asia Pte Ltd
151 Lorong Chauan
New Tech Park
#05-08
Tokyo 163-0722
Singapore 556741
Japan
http://www.fujitsu.com/global/services/microelectronics/
http://www.fujitsu.com/sg/services/micro/semiconductor/
USA
Europe
Tel: +1 408 737 5600
Fax: +1 408 737 5999
Tel: +49 6103 6900
Fax: +49 6103 690122
Fujitsu Microelectronics America, Inc.
Fujitsu Microelectronics Europe GmbH
1250 E. Arques Avenue, M/S 333
Sunnyvale, CA 94088-3470
USA
Pittlerstrasse 47
63225 Langen
Germany
Tel: +1 800 866 8608 Customer Response Center
Fax: +1 408 737 5984 Mon-Fri: 7am-5pm (PST)
http://www.fujitsu.com/us/services/edevices/microelectronics/
http://www.fujitsu.com/emea/services/microelectronics
44
The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU
MICROELECTRONICS sales representatives before ordering.
The information and circuit diagrams in this document presented as examples of semiconductor device applications, and are not
intended to be incorporated in devices for actual use. Also, FUJITSU MICROELECTRONICS is unable to assume responsibility for
infringement of any patent rights or other rights of third parties arising from the use of this information or circuit diagrams. No license
is granted by implication or otherwise under any patent or patent rights of Fujitsu Microelectronics Europe GmbH.
FUJITSU MICROELECTRONICS semiconductor devices are intended for use in standard applications (computers, office
automation and other office equipment, industrial, communications, and measurement equipment, personal or household devices,
etc.).
CAUTION: Customers considering the use of our products in special applications where failure or abnormal operation may directly
affect human lives or cause physical injury or property damage, or where extremely high levels of reliability are demanded (such as
aerospace systems, atomic energy controls, sea floor repeaters, vehicle operating controls, medical devices for life support, etc.)
are requested to consult with FUJITSU MICROELECTRONICS sales representatives before such use. The company will not be
responsible for damages arising from such use without prior approval.
Any semiconductor devices have inherently a certain rate of failure. You must protect against injury, damage or loss from such
failures by incorporating safety design measures into your facility and equipment such as redundancy, fire protection, and prevention
of over-current levels and other abnormal operating conditions.
If any products described in this document represent goods or technologies subject to certain restrictions on export under the Foreign
Exchange and Foreign Trade Control Law of Japan, the prior authorization by Japanese government should be required for export
of those products from Japan.
FME/MS/SFDAC1/DS/4250 2.0
Page 44 of 44
Production
Copyright © 2003-2009 Fujitsu Microelectronics Europe GmbH
Disclaimer: The contents of this document are subject to change without notice. Customers are advised to consult with FUJITSU sales representatives before
ordering.The information and circuit diagrams in this document are presented “as is”, no license is granted by implication or otherwise.
相关型号:
![](http://pdffile.icpdf.com/pdf2/p00225/img/page/MB86061PFQ_1317065_files/MB86061PFQ_1317065_1.jpg)
![](http://pdffile.icpdf.com/pdf2/p00225/img/page/MB86061PFQ_1317065_files/MB86061PFQ_1317065_2.jpg)
MB86061PFQ
D/A Converter, 1 Func, Parallel, Word Input Loading, 0.006us Settling Time, PQFP64, PLASTIC, QFP-64
FUJITSU
©2020 ICPDF网 联系我们和版权申明