•  
• EDN Europe MAGAZINE
•  
• Free SUBSCRIPTION
• Subscription Renewal
• View Media Kit

• EDN
• EDN Australia
• EDN Asia
• EDN China
• EDN Japan

Baker's Best: System or technology dictates ADC choice

How do you decide which ADC technology to use in your applications before you do a thorough system evaluation? Maybe you prefer SAR (successive-approximationregister) ADCs because you assume they are easy to use and a bit faster than delta-sigma converters. Then again, you may select delta-sigma converters because you assume they are slower but have good resolution. Or your decision process may be even simpler: maybe you choose the ADC that you have always used.

When selecting a converter, you usually base your decision on the ENOB (effective number of bits); accuracy; repeatability, or noise; and output data rate. You may assume that SAR ADCs produce accurate outputs with medium output speeds and that deltasigma converters produce lower-noise output signals with slower output data rates. These assumptions may no longer guide you when deciding between a SAR ADC and a delta-sigma ADC.

Think about changing your design paradigm from focusing on individual devices to considering the complete system. You will find that both ADC architectures might be appropriate for a given application. For instance, if you know the system ENOB, you may find that combining an analog gain stage with a SAR ADC matches the performance of a higher-speed deltasigma converter.

A system evaluation includes inspecting the system's sampling speeds, analyzing its accuracy, and comparing its repeatability, or noise-level, capability. To inspect sampling speeds, select a single clock frequency and allow time for the analog components to fully settle before conversion. With system accuracy, you combine the dc performance characteristics intoatotal-unadjustable error figure-of-merit for comparison.

The repeatability evaluation differs from the accuracy evaluation in that it defines how consistently a value from one conversion to the next repeats itself. With a repeatability evaluation, you can combine the noise performance of the signalchain devices in terms of effective resolution.

As we examine the accuracy and repeatability in our system evaluations, we will use Table 1 as our starting point. The circuits in the table encompass handheld-meter, data-logger, automotive- system, monitoring-system, and many other applications. Each system's gain ranges from one to 128. Column 2 in the table lists the ideal system's fullscale range referred to the input of the system. The system's LSB (least-significant bit, Column 3) is equal to the system's full-scale range divided by the number of system codes: 4096.

In my next column, I'll delve into the conversion speeds of these designs. In future columns, I'll examine the differences between a 12-bit SAR ADC, a multiplexed PGA (programmablegain- amplifier)-SAR ADC, and a 24-bit multiplexed delta-sigma converter. The analog or digital gain range for each system will be 1 to 128V/V, and the power-supply voltage will be 5V. I'll also investigate the accuracy and repeatability of these systems.

Here is the trillion-dollar question: Which system is best for the listed applications— a PGA-SAR ADC or a delta-sigma ADC? You can reach me at ti_bonniebaker@list.ti.com with your best guess. Be sure to include a description of your application with its basic requirements.

REFERENCES

Baker, Bonnie C, "Analog-to-Digital Converters, Part 4 of 6: Move Your System Strategy to the Forefront in Your ADC Designs," Texas Instruments, Dec 3, 2009, www.en-genius.net/includes/files/acqt_113009.pdf. "Nuts and Bolts of the Delta-SigmaConverter," Texas Instruments, www.ti.com/nutsandboltsvideo-ca.

Author Information

Bonnie Baker is a senior applications engineer at Texas Instruments and author of A Baker's Dozen: Real Analog Solutions for Digital Designers.