Post Go back to editing

No lock on IF synthesizer after slow power turn off

Category: Hardware
Product Number: ADF4218L

We are using the ADF4218L dual synthesizer in place of the long-extinct LMX2330L from National/TI.

The ADF is powered at 3.0V by an LDO switched on/off by our microprocessor. This LDO has very good risetime (<10us) during turn on. But the voltage decays very slowly when the LDO is turned off. This long fall-off in voltage is expected and due to the LDO's output capacitance slow discharge through the surrounding circuitry. During discharge, this LDO output voltage hovers around 1V for 10s of milliseconds before finally winding its way down to ~ 0V. The LDO voltage then turns back on a few hundred milliseconds later. This on/off cycle continues as long as the radio operates. 

We have found that a small number of synthesizers will not lock properly after the first 'off' period after startup. The R synthesizer locks properly but the I synthesizer doesn't lock. Correct operation can only be recovered by a complete reset of radio. 

Our strong suspicion is that the slow discharge time - and excessive amount of time spent at ~1V - is somehow causing distress to the synthesizer and causing it to lock up.

We've done two modifications to the board that reinforce our suspicions. Increasing the output capacitance of the LDO will make the problem worse. Conversely, putting in a constant 100 ohm discharge resistor at the LDO output will cause the output voltage to discharge much quicker when the LDO is turned off.  

Both of these tests hint that the problem is related to the slow turn off of the LDO. This very slow discharge voltage evidently put a small number of synthesizers into a fugue state. Those devices cannot be reprogrammed unit power is completely reset.

We can fix this by using an LDO with a auto-discharge feature when the LDO is turned off. This discharges the output voltage instantly at turn off. We never see the same synthesizer lock up problem using an LDO with this pull-down feature. 

Can ADI could provide any guidance as to what may be happening here? The auto-turnoff LDO seems to fix the problem. We just need some help understanding the failure mechanism. 

Parents
  • In the initial set-up where the lock issue is present, was the part fully reset (reaches 0V) prior the power-up? 

    •      In this scenario - slow power down, the part may be powered back up prior power down reaching 0V. In cases like this the part will be below operating range but not low enough to achieve a full reset. Another reason is that the slow ramp could not trigger a proper internal reset causing the device to power back up in random state. 

    • a clean power sequence is required since there is no software reset nor a reset pin for this part. 

Reply
  • In the initial set-up where the lock issue is present, was the part fully reset (reaches 0V) prior the power-up? 

    •      In this scenario - slow power down, the part may be powered back up prior power down reaching 0V. In cases like this the part will be below operating range but not low enough to achieve a full reset. Another reason is that the slow ramp could not trigger a proper internal reset causing the device to power back up in random state. 

    • a clean power sequence is required since there is no software reset nor a reset pin for this part. 

Children
No Data