We have a development card using 2 x AD9706 DACs to drive a zero-IF IQ modulator that requires a DC bias of 1.15V. We have set the full scale current to 5mA (Radj = 6.45K). We have set the OTCM to 0.5V, derived from a ADP223 Dual LDO. We have a load resistance of 470 ohms. When the development card is powered up the OTCM pins on the 2 x DACs are both +0.5V. When the DACs are programmed and are being clocked the OTCM pins on both DACs increases to +2.2V for some reason. I have ADP223 powered on pin 6 from +3V3, the EN1 - pin1 and EN2 - pin2 are connected to +3V3 for autostart, pin 3 is GND, pin 8 - ADJ1 is connected to pin 7 via a 0ohm resistor to provide +0.5V from VOUT1, and pin 4 - ADJ2 is connected to pin 5 via a 0ohm resistor to provide +0.5V from VOUT2. The GND paddle is connected to GND
If the +0.5V is derived from an external bench supply to both DACs the OTCM pins remain at +0.5V even when the DAC clocks are running.
Can anyone offer any insight into why the OTCM pins are increasing in voltage to +2.2V when the DAC clocks are running, when derived from the ADP223 LDO?
**Update - We tried adding a potential divider to the ADP223 between the ADJ and VOUT pins, with values of 200k/200k. With no load we have +1V output from VOUT1 and VOUT2 as expected, but when we apply a load both outputs drop to +1.65V, which is around +0.5V less than where we were with 0R between the ADJ and VOUT pins, so there seems to be some kind of relationship between load/no load. Any ideas?