Many of today's applications require precision sampling analog-to-digital converters with resolutions of 12 bits or more, since the high resolution lets users make accurate system measurements. But higher resolution also implies the system will be more sensitive to noise. Every time the system resolution increases by 1 bit for example, from 12 bits to 13 system sensitivity rises by a factor of two.
Thus, when designing with A/D converters, it is crucial that designers consider the noise contributions from an often forgotten source: the system power supply. Most Analog Devices A/D data sheets suggest that linear power supplies be used because their noise is lower than that of switching supplies. This article describes an easy way to measure the system's ac power supply rejection and quickly determine how badly the power supplies are corrupting the system dynamics.
Power supply rejection ratio (PSRR) is the ratio of the change in power supply voltage to the resulting change in the A/D's gain or offset error. This can be expressed in fractions of a least significant bit (LSB), as a percentage, or logarithmically, and is usually specified at dc.
But this method only reveals how one specified parameter of the A/D converter may change with a change in power-supply voltage and therefore cannot prove the converter's robustness. A better method is to test supply rejection by riding an ac signal on top of the dc power supply, actively coupling the signal through the A/D's circuitry. This method actively exercises all of the sensitivities that affect converter performance.
An A/D's dynamic performance is commonly specified by its signal-to-noise ratio (SNR), signal-to-noise plus distortion (SINAD), and spurious free dynamic range (SFDR). In other words, any noise in the setup or application will define the A/D's performance. Therefore, power supply noise could seriously degrade the overall performance of the system. It is therefore imperative that the user understand the system supply's noise. Linear-type system supplies are better than switching-type supplies, and their use is thus recommended. Switching-type supplies will have a higher chance of causing problems, so they must be well isolated from the sensitive analog circuitry in cases where they are the only choice for the system.
For example, if an ac signal riding on the power supply “pops up” out of the noise floor and “sits” in the system base-band, the amount of distortion measured could be based on a supply noise spur rather than a harmonic. If the error spur is great enough it can also mix with other errors, further increasing the system noise and limiting the dynamic range.
What to do
There is no possible way to ensure that all supply noise is eliminated in your application. No system will ever be totally immune to unwanted power supply interaction. Therefore, the A/D user must be proactive during the layout stage. Here are some useful tips in maximizing PC board noise immunity to supply changes:
To this point the article has described what to do, or not do, to achieve data sheet PSRR in your application. Now, it is time to focus on a quick way to measure PSRR on your system board. On a PSRR measurement of an A/D on its evaluation board, each supply is measured individually to gain a better perspective on the converter's dynamic behavior when an ac signal rides on the power supply under test. Start with a high capacitor value, such as a 100-microfarad nonpolarized electrolytic. For the inductor, use 1 millihenry to act as the ac blocker to the dc power supply.
Using an analog oscilloscope, such as the Tektronix 475, measure the amplitude of the ac signal with a scope probe applied to the point at which the power supply enters the system board. To make things simple, define the amount of ac signal riding on the supply as a value related to the A/D's input full-scale. For example, if the A/D's full-scale is 2 Vpp, then use 200 mVpp or – 20 dB.
Next, with the input of the A/D grounded (no analog signal applied), look for an error spur at the test frequency coming out of the noise floor on the data capture output response screen or CPU. To calculate PSRR, simply subtract – 20 dB from the error spur value seen on the screen. For example, if the error spur shows up at – 80 dB from the noise floor, then the PSRR is – 80 dB – – 20 dB or – 60 dB, (PSRR = error spur dB – oscilloscope measurement dB). That – 60 dB may not seem like much, but if you look at that in terms of a voltage, it equates to 1 mV/V, which is not uncommon for a PSRR specification in an A/D data sheet.
The next step is to vary the frequency and amplitude of the ac signal in order to characterize the ADC's PSRR in your system board. Most data sheet numbers are typical, but you should specify worst-case operating conditions. For example, the +5-V analog supply might be the worst-performing one over the first Nyquist zone. All other supplies should perform at better than or equal to the specified typical numbers.
The disadvantage of this test is the use of an LC arrangement. When sweeping the frequency band of interest, the signal level required at the output of the waveform generator to achieve the desired input level at the A/D supply pin may be very high. This is because a notch filter has been formed, greatly increasing ground current at the notch that in turn can get into the analog inputs. To get around this, simply swap in new LC values when testing at the frequencies that are causing difficulty.
The author hopes to have given the reader a clear view of PSRR's validity and why it is so important to the user's system dynamics. One should appreciate the layout techniques and hardware required to achieve data sheet specifications of an A/D converter on the system board. An easy and quick method of testing an A/D's PSRR specification is derived using your given setup and a couple of extra passive components.
Test A/D PSSR by injecting noise.
Each supply is measured individually to gauge ADC's dynamic behavior
Source: Analog Devices Inc.