F75393

F75393

May, 2008

V0.16P

-6-

These five events are wired-NOR together. This means that when one of out-of-limit event occurs, the ALERT# output goes low if the

MASK control is disabled. ALERT# signal can be used as an IRQ-like interrupt or as an SMBALERT. When ALERT# acts as an

IRQ-like interrupt, the ALERT# will be de-asserted until the following 2 conditions are matched:

(1). The abnormal condition is gone

(2). Reading the Status register to clear the status

When ALERT# acts as a SMBALERT, the ALERT# will be de-asserted until the following 3 conditions are matched:

(1). The abnormal condition is gone

(2). Reading the Status register to clear the status

(3). The ALERT# has been serviced by the SMBus master reading the device address.

For more information about SMBALERT, please see SMBus 1.1 specification.

6.6

THERM#

Either VT1(Local) or VT2(Remote) temperature exceeds the corresponding THERM limit, the THERM# output will assert low. The

asserted output will be de-asserted until the temperature goes below (THERM Limit – Hysteresis). The hysteresis default value is

10°C and it can be programmed. Both VT1 and VT2 have their own THERM limits and Hysteresis values.

6.7

ADC Conversion Sequence

If a START command is written, both channels are converted and the results of both measurements are available after the end of

conversion. A BUSY status bit in the status byte shows that the device is actually performing a new conversion; however, even if the

ADC is busy, the results of the previous conversion are always available.

6.8

Thermal Mass and Self Heating

Thermal mass effect can seriously degrade the F75393’s effective accuracy. The thermal time constant of the SOP package is about

140 in still air. For the F75393’s junction temperature to settle to within +1°C after a sudden +100°C change requires about five time

constants or 12 minutes. The use of smaller packages for remote sensors such as SOT23, improves the situation. Take care to

account for thermal gradients between the heat source and the sensor package do not interfere with measurement accuracy.

Sel-heating does not significantly affect measurement accuracy. Remote sensor self-heating due to the diode current source is

negligible. For the local diode, the worst case error occurs when auto-converting at the fastest rate and simultaneously sinking

maximum current at the ALERT# output. For instance, at an 64Hz rate and ALERT# sink around 0.7mA when pull up resistor 4.7K

3.3V and no copper PC board heat-sinking, the resulting temperature rise is:

dT = 1.01mW x 120 °C/W = 0.12 °C

Even with these contrived circumstances, it is difficult to introduce significant self-heating errors.