AD580

REV. A

–4–

Figure 3. AD580 Schematic Diagram

Figure 4. AD580 Connection Diagram

VOLTAGE VARIATION VS. TEMPERATURE

Some confusion exists in the area of defining and specifying

reference voltage error over temperature. Historically, references

are characterized using a maximum deviation per degree Centi-

grade; i.e., 10 ppm/

°C. However, because of the inconsistent

nonlinearities in Zener references (butterfly or “S” type charac-

teristics), most manufacturers use a maximum limit error band

approach to characterize their references. This technique mea-

sures the output voltage at 3 to 5 different temperatures and

guarantees that the output voltage deviation will fall within the

guaranteed error band at these discrete temperatures. This ap-

proach, of course, makes no mention or guarantee of perfor-

mance at any other temperature within the operating tempera-

ture range of the device.

The consistent Voltage vs. Temperature performance of a typi-

cal AD580 is shown in Figure 5. Note that the characteristic is

quasi-parabolic, not the possible “S” type characteristics of clas-

sical Zener references. This parabolic characteristic permits a

maximum output deviation specification over the device’s full

operating temperature range, rather than just at 3 to 5 discrete

temperatures.

Figure 5. Typical AD580K Output Voltage vs. Temperature

The AD580M guarantees a maximum deviation of 1.75 mV

over the 0

°C to +70°C temperature range. This can be shown to

be equivalent to 10 ppm/

°C average maximum; i.e.,

1.75 mV max

70

°C

×

1

2.5V

=10 ppm/°C max average

The AD580 typically exhibits a variation of 1.5 mV over the

power supply range of 7 volts to 30 volts. Figure 6 is a plot of

AD580 line rejection versus frequency.

NOISE PERFORMANCE

Figure 7 represents the peak-to-peak noise of the AD580 from

1 Hz (3 dB point) to a 3 dB high end shown on the horizontal

axis. Peak-to-peak noise from 1 Hz to 1 MHz is approximately

600

µV.

Figure 6. AD580 Line Rejection Plot

Figure 7. Peak-to-Peak Output Noise vs. Frequency