eeTimes
eeTimes
eeTimes eeTimes EDN
Forgot password Register
Print - Send - -

Conversion

Cloning an A/D converter that hit end-of-life presented engineering challenges

March 21, 2012 | Tony Khazen | 222903188
Cloning an A/D converter that hit end-of-life presented engineering challenges Tony Khazen, Director, DATEL, a Business Unit of Murata Power Solutions outlines some of the challenges faced when selecting a 'cloned' analog-to-digital converter to replace an obsolete model.
Page 1 of 3

Data Converter products are widely available from multiple sources. There are very few pin-to-pin replacement and directly compatible products within the industry. An analog-to-digital converter of a particular resolution and speed from several suppliers may be quite dissimilar in package size or type, pin-out, functionality and performance.

Some converters may require multiple supply voltages while others may require only one. Some may need multiple ground pins while others may use separate analog and digital ground returns. The digital timing and analog front end are typically different between manufacturers. This dissimilarity came to light when a prominent military equipment supplier was hit with an end-of-life situation on one of the A/D converters that was used in their equipment.

This supplier of radar systems was in urgent need of a drop-in replacement A/D converter. They approached Datel seeking a cloned A/D for the obsolete model in the original design. Our engineering team accepted the opportunity but the challenges were more than anticipated. Certainly, the first task was to fit the replica in the same size package and map the circuit to the same pin-outs.

Timing was a tough undertaking, especially the output data valid time that had to mesh with the existing system timing. Aperture Delay is an important parameter for radar applications, especially for I and Q demodulation. Much design time was required to perfectly match the performance of the obsolete part. Because two units are required per radar system, (one for the I and one for the Q channel) the aperture delay could not vary from one unit to another.

Figure 1 shows the measurement of the Aperture Delay while Figure 2 shows the variance of Aperture Delay over temperature and between different units. Similarly, Aperture Jitter is another key parameter when sampling a high input frequency waveform. The design required sub-4 picoseconds in order to preserve the accuracy of the conversion and maintain the desired Signal-to-Noise ratio of 83 dB.

1 | 2 | 3 | Next page







Please login to post your comment - click here
Related News
    No news
MOST POPULAR NEWS
Interview
Technical papers
Poll
What are your most important criteria when selecting an analog supplier?