As my forum name would suggest I’m not interested in chipping myself (when internal blood glucose testing kits and integrated e-ink tattoos become a working thing THAT’s when I’d want to be biohacked) but I am interested in building RFID sensors for our cats.
I’m in the UK and I believe we only have one standard so I’d be interested in any info on what kind of antenna setup I would need to read them. The main idea would be to track our cats around the house so I’d want to set up something with a range of about a foot or two (ideally a door’s width). Being able to tinker with making things like automatic feeders and cat flaps would be great as well (the commercial ones have some limitations, not being wifi linked).
A foot might be pushing it slightly.
But you know what I’d do if I were you? I’d get a bunch of SureFlap RFID cat flaps, remove the flap itself, unsolder the latch coil and connect a relay instead, and voila: you have yourself a very effective, el-cheapo cat detector that you can interface with whatever the hell you want. For 60 quid, you got yourself a nice cat-sized “detector tunnel” with a battery holder (the batteries last forever, even with hyperactive cats), the circuitry to read FDX-B and EM41xx chips that truly never misses a read, and an easily hacked interface. Strategically place the hacked catflaps where your cats have to go through them and you’ll be able to track their movements.
I know because I hacked three of those things to do almost what you want to do: I had two of them one after the other - as an “extended tunnel” - to detect my cats going in or out and prevent them from getting in with dead critters at night, and the other was used to release food into their bowls when I was away on vacation - all that controlled by my home server under Linux.
UHF tags are much better suited for your application. They can have a range measured in meters instead of mm like the pet chips do.
Downside is probably cost. Readers aren’t too cheap but I haven’t really looked whats out there.
I assumed the OP’s cats were already tagged - hence the requirement for LF RFID. But yeah, if the kitties have collars, then it’d be just as simple and cheap to have them carry around BLE tags and stick a bunch of bluetooth receivers around the house.
These look decently awesome, to get such a reliable read I could see these being useful for plenty of human projects too
The performance of the reader is pretty awesome considering it’s battery-powered. It got a solid read out of my EM implant anywhere within the frame of the catflap and it put many more powerful and more expensive readers to shame.
Sadly, I could never figure out how to get the UID out of it. There are test points on the PCB, but they’re not usable. So you can only use them as detectors for up to 32 predefined chips.
Would it be possible for you to post a picture of the primary and secondary sides of the PCB, as well as closeups of the reader ICs with a magnifier app?
Sadly no: I left all that junk behind when I moved to Finland. That’s what I was writing in the past tense.
There’s a lot of stuff you have to throw away when you emigrate. It’s the 7th time I do it, and as always, I’m left with a bunch of cardboard boxes and a few suitcases Until I find a new house to buy and settle down for good here, I’m sort of camping with the few things I brought with me.
Unfortunately our cats are chipped but not collared. Both of them are known for getting into scrapes so we wouldn’t put it past them to throttle themselves if they had a collar. Without a collar the only chip that we can track them with would be an implanted one.
As I’ve seen plenty of RFID readers at around £20-£30 and the price of the flap on the site you linked was £60 (and the cheapest I could find it on google shopping was £47) I think you and I have a slightly different definition of “el cheapo”.
That said, at £60 it’s not totally out of the bounds of possibility,
Ideally it’s for tracking the cats indoors without getting in the way of the humans (me and my wife). As there aren’t many doorways that the cats could get through without going at ground level my idea was to have one at the side of the door.
If I could get a reader/antenna at ground level (low profile enough to fit in a mat) then the needed range could be reduced but then I think the chips would be shoulder level for the cats so it’d still be about 6-8 inches (or something like that).
Well, you’re right: £60 - or even £47 - ain’t cheap. But I’ll tell you something else I forgot to mention, that you should factor in: good luck finding a bona fide medium range FDX-B reader with a USB port for under £500. The only “affordable” computer-attachable reader I know of that does FDX-B is the Elatec TWN4 MultiTech2, with which the read range for implants is contact at best, and which requires a flexible understanding of affordable.
Unless you want to roll your own - there are China-made modules for $30, and you can coil your own inductor - there just isn’t any ready-made solution available on the market. I figured the path of least resistance to solve your problem is reusing a not-so-cheapo, but readily available and easily hackable catflap.
Why are you looking to track them in the house? Could you not have a occupancy system for example the cat is in then if the tag is read then the cat must have gone out?
I am setting up a tracking system for my cat soon^tm using cams and object detection. Mainly so that when I am away for a few days I know what cam to look at to see her
Are you using Raspberry Pi(s) to run that? TensorFlow?
Not sure yet. Likely something more powerful than a pi but we will see. I have lots of options and ideas.
More powerful than a Pi and camera just to remotely watch webcam footage of Puss…Never mind, I see where you are going with this
Ohh, no, I have network cams. I will likely run the image recognition on a server blade with my other smart home stuff. Although I have been toying with the idea of a pi cluster just so I can mess around with orchestration tech.
Goto thingiverse and look at 6" rack and raspberry pi blade!