Battery powered rfid reader

I'm just looking for some advice on creating a battery powered RFID reader. The reader would read a passive RFID bracelet when a user touches the reader with the bracelet. It needs to be battery-powered and last a minimum of 2 months. I've got everything working but am stuck on how to keep the amps to a minimum. I have been trying out some ideas:

  • pir sensor that turns on my esp32 from deep sleep and then powers the RFID for a few seconds to see if the person that came near it reads their bracelet. If nothing is read in 5 seconds, shut back down. Problem: pir is slow and there may be a lot of people around, so it could go on more than needed. I only need short range.
  • esp32 that goes into deep sleep and then powers up reader every second to see if a tag is near the reader. Problem: I've got this working using this code: Power down modes? · Issue #269 · miguelbalboa/rfid · GitHub, but it's reading a lot more amps than that person found and it only reads about 50% of the time (maybe I can improve).
  • ir sensor/break beam sensor. Problem: uses too much power
  • getting the RFID reader to be in a low power mode and wake the esp32 when it detects a tag. Problem: From what I can tell, RFID readers (mfrc-522 and pn532) can't read a passive tag in low power mode.
  • use a reed (magnetic) switch and attach a magnet to the bracelet. When the bracelet comes close, I power on the esp32 with the switch and then read the tag. Problem: haven't tried yet, but worried the magnet will interfere with reading the RFID tag and would prefer to use an unmodified RFID bracelet.

None of these solutions seem ideal does anyone have any other ideas? I'm thinking maybe a momentary switch on the front of the device that they push their bracelet against, but this seems awkward.

Thanks for any input.

It won’t, but a reed switch will not trigger unless it is very close to the magnet. A Hall effect switch would not be sensitive enough either, but a magnetometer sensor would do the job. However that takes current to run, read the data sheet to find out how much.

Hi,
Simple press button.
Make it so you need to press the sense plate with the tag, microswitch underneath.
Extremely low power.
You need to keep the controller boot time very low however so the read is almost instantaneous.
And have a READ LED, two colour, RED reading, GREEN read and OK, flashing RED, invalid RFID tag.

Tom... :grinning: :+1: :coffee: :australia:

Most RFID readers need a second or so to power up.

Thanks for these ideas. Yeah, I think a push-button is the simplest but most effective solution. Just need to figure out the case design and button that is easy to press.

I did some tests earlier in the project and if I cut the power to the mcu and then plug it back in, the RFID reader (mfrc-522) starts to work almost instantly. So the button should work. I could be wrong, but I'm pretty sure it was instantaneous. I'll set it up again.

I'll look into the magnetometer sensor as well.

Also thought of using an ultrasonic rangefinder. It looks like it runs off 2ma at 3v, so that would work to last a while. Need to make sure it can wake up the esp.

Hesitant to use buttons because this thing needs to be bombproof...

Hi,
What does your project do with the RFID info if it is valid or invalid?
What is your projects OUTPUT?

Thanks.. Tom... :grinning: :+1: :coffee: :australia:

It sends the rfid via espnow to another esp which sends it to a server

It doesn’t matter too much if it’s invalid, the server will handle parsing the ids and I’ll probably ignore any invalid ids

I think a physical push button switch would be bomb proof.

Ah new information that changes any advice you have been given. If it is in a climbing wall why not have mains powered readers as these are going to be in a fixed location?
He readers should be able to read through the wall as I assume it is not made of metal.

Hi,
Can you please tell us the complete scope of your project, what is the RFID needed for.
What is the whole system?

WiFi activation is going to consume power, do you know how much?

Tom... :grinning: :+1: :coffee: :australia:
PS @Grumpy_Mike looks like you have been sidelined by @vednus deleting the posts refering to climbing wall. Lucky you quoted it. :+1:

Sorry, just getting scared about competitors following my trail

Hi,
Okay, now why does the battery have to last 2months minimum.
Climbing wall seems to hint at a commercial wall climbing centre, not a remote rarely visited site.

If it is being used nearly everyday, that is humans are approaching it nearly everyday, why not change to a removeable battery nearly everyday? The battery life being say 3 or 4 days.

Tom... :grinning: :+1: :coffee: :australia:

There could be 200 climbs in a commercial gym. Changing the batteries isn’t feasible, but I think I’ve found a solution. I’ll post it once I’ve built it

I don’t see why the power supply to a reader has to be physical close to that reader. You could even build in a network of sockets from the power supply to make swapping the readers monitoring the holds easy.