5V Regulator for PIC in Sleep Mode

[MikeMl]

I have a PIC project in which the PIC goes into sleep mode where it draws a few uA. While awake, the PIC circuitry draws about 30mA @ 5V. Input voltage is from a 12V lead-acid battery, so ranges from 11.5V to 14.5V.

317 and 7805 type regulators have quiescent currents that are huge compared to the PIC's sleep current, so I'm looking for a regulator that has a minimum current draw of 50uA or less. The PIC must be continuously powered, so I'm not looking for a regulator with a shut-down pin.

 

[colin55]

Can you use the micro to turn on its own regulator?

 

[MikeMl]

Can you use the micro to turn on its own regulator?

No, because the PIC must be powered while in Sleep Mode, so it can wake up on Port change. It can stay in sleep mode for months. The PIC and its regulator are a parasitic load on the battery, so that is why I'm trying to minimize it.

 

[dknguyen]

The ADP1714 (300mA), ADP1716 (500mA), and ADP1707 meets your requirements. have a minimum Iq around 65uA or so. However, max input voltage is 5.5V. It may be a technological limitation you are seeing for the minimum Iq on regulators that can handle 12V inputs.

But whether you find a regulator or not, perhaps you should use a 6V lead-acid battery with double the capacity intead. That'd nearly double efficiency and may let you use regulators better optimized for low Iq.

EDIT: Hunted around and found these ones with even better Iq and fits your voltage range:
https://www.electro-tech-online.com/custompdfs/2010/08/3009fc.pdf
http://cds.linear.com/docs/Datasheet/30105fd.pdf
http://cds.linear.com/docs/Datasheet/1761sff.pdf

Each datsheet also has a related products section at the end that has a bunch more possible parts too.

 

[colin55]

No, because the PIC must be powered while in Sleep Mode

I did not ask this question.
I asked if you had a spare line to power its own regulator.

Place a 5v1 zener across the micro.
Power the chip from a 10k resistor. Add 100u electrolytic.
When chip wakes up it turns on its own 5v regulator via a transistor on the common pin of the regulator.

 

[ericgibbs]

I have a PIC project in which the PIC goes into sleep mode where it draws a few uA. While awake, the PIC circuitry draws about 30mA @ 5V. Input voltage is from a 12V lead-acid battery, so ranges from 11.5V to 14.5V.

317 and 7805 type regulators have quiescent currents that are huge compared to the PIC's sleep current, so I'm looking for a regulator that has a minimum current draw of 50uA or less. The PIC must be continuously powered, so I'm not looking for a regulator with a shut-down pin.

hi Mike,
Two Vregs here.

 

[MikeMl]

I did not ask this question.
I asked if you had a spare line to power its own regulator.

Place a 5v1 zener across the micro.
Power the chip from a 10k resistor. Add 100u electrolytic.
When chip wakes up it turns on its own 5v regulator via a transistor on the common pin of the regulator.

I already have a pin which goes high when the PIC wakes up. It is controlling a high-current, High-side PFet which applies battery voltage to a SMPS, so I already have a switched source of 12V.

Running the numbers on a Zener shunt regulator doesn't look good, however. A series 10K resistor would have (13-5)/10k = 0.8mA, which is too high. Zeners that I have around here do not do well with less than 1mA of bias current.

 

[colin55]

Use a 15k to 22k. You don't need the zener to regulate. It simply will not let the voltage rise above 5v1.

 

[crutschow]

ericgibbs' recommendations look good to me. They have a low bias current and a low drop-out voltage.

 

[MikeMl]

Thanks to all who responded. I learned something...

I'm trying to do it with what I have on hand; ordering a LM2936 and waiting for the mail is a PITA.

Based on Colin's input, here is what I came up with:

 

[crutschow]

As long as the zener maintains a voltage above the minimum operating voltage of the PIC at the low bias current, then that should work also.

 

[colin55]

A zener does not "turn on" until the zener voltage is reached. But the actual zener voltage can be a margin below the marked value and the zener will "leak" when this lower value is reached and continue to "leak more" as the voltage rises - to a point where it "leaks considerably" when the rated voltage is reached. For a 5v6, this lower voltage can be as low as 4.7v. When a zener is supplied with a very low current this leakage will be noticeable and it will appear that the zener is only working at 4.7v. As the available current is increased, the voltage across the zener will increase. This is because the leakage current creates a lower voltage-drop across the "current limiting resistor."

 

[ericgibbs]

hi Mike,
This is another option.
LTS.asc

EDIT: added 2nd file
If the series 2N7000 is replaced with a Si4404 and R3 a 4.7R , the +5V regulator will pass 50uA thru 500mA subject to loading.

 

[MikeMl]

Thanks Eric. Before I used the simple Zener circuit suggest by Colin, I Spiced a series regulator based on the TL431. It was simpler than yours, it took advantage of not needing a low drop-out voltage, and it used a series NFet with the TL431 shunting the gate of the FET to ground. I biased the TL431 so that it had much less than the minimum 1mA specified on the data sheet. It simed ok, and had low quiescent current, but when breadboarding it, it didn't regulate any better than the simpler Zener version, and used a lot more parts...