Battery powered PIR

[Mrlynx]

What is downside to instead using 3xAA and a good LDO regulator with a modified sensor?
This way you dont get any ripple and dont need any step up booster.

[therik]

@mrlynx
Downside to using 3xAA would be size.

@Andreas-Maurer
I would prefer to use the 3 pin header and not have wires coming from multiple places, but you are correct.

[m26872]

@therik
Nice work! Do you know the overall sleep mode power/current? And what battery life time one can expect from a real application?
@Andreas-Maurer
If you leave the regulator, isn't there the a power draw from the reverse feeding like with the Arduino pro minis?
@Mrlynx
Which LDO do you have in mind? Maybe the battery economy would be a little worse as well. At least with 3AAs and >500mV drop-out..

[darazs]

So, I'm quite a rookie when it comes to electronics, but I made the 3.3V modification to this PIR sensor too. I had trouble with stability as well when I removed the voltage regulator and the diode, so I removed the two capacitors from the bottom (one 47uF, and the other 22uF) that I assumed are the buffers on both sides of the voltage regulator. Now the sensor is fed directly from my 3.3V step-up converter (based on NCP1402SN33T1) and it's very reliable and stable.

I have 2 of these sensors connected to the two interrupt pins and my sensor is going strong since 2 weeks on a single AAA NiMH battery (750mAh). I think it has at least another 2 weeks in it.

I'm using these to trigger the lights in my entrance room, so I'd notice if they wouldn't work. The board also wakes up every minute to send the battery voltage and I don't see unusual wakeups or unresponsiveness.

[therik]

@darazs Good to know. I think the output of the booster is very dependent on the current draw and, of course, every booster is a little different.

[darazs]

@therik Yeah, now that I've read through your "Efficiency of Voltage Boosters" topic, I understand things a little better, and I really have no clue why removing the caps made the PIR more stable. BUT your topic helped me solve my packet loss/acks not arriving problem by adding a bigger capacitor to my radio, so what I want to say is that you're awesome, thanks! :)

[Mrlynx]

@m26872 I am looking at xc6206 and tc2117 that have a low quiescent current and a dropout of 50-100 mV at low currents.

[Yveaux]

@therik I recently observed the same behaviour with a step-up regulator based on a 2108A.
The strange thing is that the sensor behaves erratic when powered from 2 fresh AA alkaline batteries (approx. 3.2V input to the regulator) but works just fine when powered by some older AA alkaline batteries (approx. 2.6V input).
I quickly inspected the output of the regulator with a scope, but saw no clear differences between the two. Maybe the output was less stable at 3.2V input.
Next I'll try using a diode to lower the voltage a little, either before or after the regulator.

[riaan003]

I am trying to figure out if running this modification from 3 AAs at 4.5V will work (it is for a setup with a attiny85 that I want to run from 4.5 down to 3.3v - it currently works well from 5 down to 4.5 with exactly the erratic behaviour explained above from about 4.4v and down). When looking at the datasheet for the BIS0001 then it seems like it will fine? What am I missing? :)

[therik]

A couple of questions.

1. Did you remove the voltage regulator from the PIR?
   a. if not, I would suspect that the voltage regulator on the PIR drops out and cannot supply the voltage for the PIR IC below 4V5.
2. Where do you derive the power for the PIR?
   a. If it is from the µC, the µC may be adding noise on the voltage supply to the PIR and with the PIR voltage regulator removed the supply voltage may be too noisy.

[riaan003]

The regulator was not removed yet, no, the behaviour was with my unmodified sensor. I have actually now started looking at an alternative power setup so that i can use this mod at 3.3v and drop the requiment for it to run from 4.5v (which was based on running it upregulated from 3 AAs with an attiny85 woken up from sleep mode - which worked well and consumed ~0.1mA overall). Back to the drawing board :)

[Igor Katkov]

How did you produce that nice ripple charts? Oscilloscope connected to a computer? What is the model?

[Haakon]

Thanks for sharing, @therik. Where can I get the high-ripple-rejection-ratio regulator you are describing? I cannot find an FT531IA on Ebay. (Or if therik is no longer following the topic: Anyone else know a specific regulator I gan get from Ebay?)

Thanks in advance.

[LastSamurai]

It's still on aliexpress. Could a AMS1117 work too though? It's ripple rejection isn't bad for what I have read. Although I don't get the part of the specs.. if you use a fixed version do you still need a capacitor on the GND (adj) pin for the ripple rejection?

[Haakon]

Thanks, @LastSamurai! I have ordered a few to try them out. (Not sure about the question about AMS1117, hopefully there are someone here who are.)

[n1ck1355]

Hi :)
does anyone know where i actually find the FT531IA? on aliexpress there are only FT531EA with output of 2.5V :(