My board can use either 2 or 3 1.5V battery cells. My input voltage is therefore between 2.6 and 4.6V. The radio module can only handle 2.4V to 3.8V, so I need to clamp this voltage to 3.8V (or preferably 3.3V).

If I use a 3.3V very low quiescent current LDO regulator, and the input voltage is 2.6V, will the pass transistor be turned on so that 2.6V (minus the LDO drop) appears on the output (at 0 - 40mA output current range)? The radio module will be asleep most of the time but when transmitting or receiving could take 18mA or so.

The LDO I'm thinking of using has versions working down to 1.2V, so I think that the logic inside will be functional and the pass transistor will turn on since the output is less than the desired output. Any thoughts?

Thanks!

Mark.

Originally Posted by crutschow

I would think that would be true for most LDOs.

Yes, I think it'll be OK too. The problem is finding an LDO with extremely low working quiescent current because the voltage has to be clamped even when the radio module is asleep. I've found one that's speced at 1.5uA max (S-1206 series from Seiko).

Thanks.

Originally Posted by crutschow

National LM2936 - Ultra-Low Quiescent Current LDO Voltage Regulator and TI Linear Regulators - Single Channel LDO - TPS78230 - TI.com also have some low quiescent current LDO regulators.

Thanks! I wasn't aware of the TI part, this is actually a better spec than the The Seiko S-1206.

The National part though does indeed have low quiescent power when shut down, but it takes up to 20uA when enabled. Need to watch the specs, sometimes they quote 'ultra low quiescent' but neglect to mention the ground current when powered up.

Originally Posted by crutschow

Well, 20µA is still only about 0.1% of your radio module current so it's not that significant.

The radio module will be asleep most of the time (several seconds, possibly as much as 10), and only awake for enough time to detect a small preamble. So if if that duty cycle was, say, 1:200 and ignoring the sleep current of the radio and micro (which is only a few uA) that is 100uA average, and then 20uA starts to become significant. The lower the duty cycle the more significant it gets. Best to get it down if possible.