DAC8812_15 Datasheet, PDF (20/29 Pages) Texas Instruments – DAC8812 Dual, Serial-Input 16-Bit Multiplying Digital-to-Analog Converter

DAC8812
SBAS349C – AUGUST 2005 – REVISED NOVEMBER 2015
11 Layout
www.ti.com
11.1 Layout Guidelines
A precision analog component requires careful layout, adequate bypassing, and clean, well-regulated power
supplies. The DAC8812 offers single-supply operation, and is often used in close proximity with digital logic,
microcontrollers, microprocessors, and digital signal processors. The more digital logic present in the design and
the higher the switching speed, the more difficult it is to keep digital noise from appearing at the output. This
device has three ground pins: two for analog ground (A_GNDA and A_GNDB) and one for digital ground (DGND),
which are pinned out on opposite sides of the device. Ideally, the analog grounds would be connected directly to
an analog ground plane, and similarly the digital ground connected to a digital ground plane. These planes would
be separated until they were connected at the power-entry point of the system. The power applied to V_DD should
be well-regulated and low-noise. Switching power supplies and dc-dc converters often have high-frequency
glitches or spikes riding on the output voltage. In addition, digital components can create similar high-frequency
spikes as their internal logic switches states. This noise can easily couple into the DAC output voltage through
various paths between the power connections and analog output. V_DD should be connected to a power-supply
plane or trace that is separate from the connection for digital logic until the analog and digital supplies are
connected at the power-entry point. In addition, adding both a 100-pF to 1-nF capacitor and a 0.1-μF to 1-μF
bypass capacitor is strongly recommended. In some situations, additional bypassing may be required, such as a
100-μF electrolytic capacitor or even a pi filter made up of inductors and capacitors – all designed essentially to
provide low-pass filtering for the supply and remove the high-frequency noise.
The compensation capacitors shown on the layout in Figure 38 are not required for normal operation of the DAC.
However, overshoot of the amplifier output voltage on large code changes is possible. This can be mitigated by
using a compensation capacitor between the IOUTx and RFBx nodes, as shown implemented here.
Performance of the DAC8812 can be compromised by grounding and PCB lead trace resistance. The 16-bit
DAC8812 with a 10-V full-scale range has an LSB size of 153 µV. The ladder and associated reference and
analog ground currents for a given channel can be as high as 2 mA. With this 2-mA current level, a series wiring
and connector resistance of only 76 mΩ causes 1 LSB of voltage drop. The preferred PCB layout for the
DAC8812 is to have all AGNDx pins connected directly to an analog ground plane at the device. The
noninverting input for the transimpedance amplifier of each channel should also either connect directly to the
analog ground plane or have an individual sense trace back to the AGNDx pin connection. The feedback resistor
trace to the transimpedance amplifier should also be kept short and have low resistance in order to prevent IR
drops from contributing to gain error. Therefore, it is important to place the transimpedance amplifier as close to
the DAC as possible. This attention to wiring and placement ensures the optimal performance of the DAC8812.
20
Submit Documentation Feedback
Product Folder Links: DAC8812
Copyright © 2005–2015, Texas Instruments Incorporated