Self flushing toilet

[pete_c]

@Tom,
 
No but the idea of a urinal came up when looking to build a home in the 1970's and seeing what peers and family had done at the time.
 
One friend (well house poor) built a ~6k sf home and put a father/mother-in-law section in the home with it's own entrance to the garage, kitchen, bedroom and living space.  The bathroom did have a urinal. He was on call 24/7 and at the time didn't want to wake up the rest of the family (2 kids / wife).  We looked too at a student of Frank Lloyd Wright Home in need of an update but a great deal at the time.  All of the bedrooms had their own bathrooms and exits out.  Two of the attached bathrooms had urinals.  Dad built a home in the 70's and lost the urinal battle but did put in a bidet in a new home.

 

[batwater]

These three ideas could work if my cat didn't have such a strange routine before doing his business. As it stands, I'll need a different approach or the toilet will be flushing 5 or 6 times before he even gets around to using it. The other problem is that he's a pretty smart little bugger and at one point I taught him how to flush the toilet but couldn't figure out how to teach him to flush immediately after using the toilet. Instead he would just keep flushing the toilet and watching the water spiral down the drain. I set up a counter on the flush handle and discovered that throughout the day while I was out he would flush it up to 20 times. At that point I had to remove the handle and try to figure out a new method. If I set up either of the two suggestions, not only would the toilet be triggered several times during his pre-poop ritual, but he would also inevitably discover that he can cause the toilet to flush in a certain way and then the games would begin again.

 
What about a 3rd photo sensor with logic behind it, 2 up high, 1 down low, I bet you could come up with a logic tree that would work. And/or a combination of the different sensor types. Perhaps adding a flush trigger that is outside of the bathroom as the final step. Ultimately you are correct your cat will likely figure the pattern out and delight in causing the toilet to flush.  Chemical and color detection seem even more problematic due to maintenance requirements than using physical sensors and the risk of false triggers (kitty playing to get the toilet to flush)  
 
Fun mind bending project.  You do know that ultimately you will loose the battle, cats are extraordinary creatures with much time on their hands to mess with you. Our cats over the years have had so many "curiosity killed the cat" close calls it's amazing they survived. But that is what makes them so endearing, that adventuresome curiosity...  

 

[roussell]

Why not implement a timer to prevent excessive flushes over a period of time. The discussion is fun but it seems like you may be trying to over engineer a solution that may only yield a insignificant efficiency improvement. Detect the motion but count/time instead of flush. Once the duration between motion-events exceeds n seconds/minutes, then flush. That gives you a couple of tunable parameters and I'd imagine with observation and tuning you can get pretty close to an ideal flushing pattern. If it doesn't work, the process itself might uncover some additional patterns or parameters that you can monitor to tune further.


Sent from my iPhone using Tapatalk

 

[Quixote_1]

I'm not convinced a timer with motion sensing will be sufficient. It will definitely play a role, though.
I set up a similar system that used multiple motion sensors in a hallway/stairwell to determine if an occupant was arriving or departing the premises, and while it was a limited success, we were not trying to fool the system and our movements were very predictable. In a situation like this, I need to know that I'm not going to come home to a flooded apartment because the little monster decided to drop one of his toys in the toilet and then flush it 20 times.
 
Right now I'm toying with the idea of using a load cell sensor to detect when he loses a little weight while on the seat.

 

[giesen]

I'm guessing probes in the water passing a current and then measuring the change in resistance is a no go?

Sent from my SM-N910W8 using Tapatalk

 

[Quixote_1]

I'm guessing probes in the water passing a current and then measuring the change in resistance is a no go? Sent from my SM-N910W8 using Tapatalk

Nope -- nothing touching the water or otherwise encroaching on the normal use of the toilet. (That includes multiple cameras.  hahaha!)

 

[Bzncrewjr]

Second cat didn't take to potty training.   He kept resisting.  So we abandoned the process.   Wish we didn't.  First cat still gets it and will occasionally use the toilet.  
 
One day we should try again.   It's just so hard to start it up and dedicate a bathroom for the process.
 
--Russ
 

Thanks for the leads. Both of those products look interesting, but again, I need more control over when the toilet is flushed to prevent my little friend from purposely triggering it repeatedly. The first product gave me a good idea -- I can detect motion and then keep track of the time spent on the seat. That way when he jumps up, starts pawing the tray and jumps down right after, maybe I can time it correctly so that it won't flush unless he's there for a specific amount of time.
I still really want to find a way of analyzing the colour of the water or reflected IR light level.
 
By the way, how did your second cat ruin the program?

 

[RAL]

Maybe this CatGenie self-flushing litter box would work better than trying to rig up something to a regular toilet.

 

[Quixote_1]

Maybe this CatGenie self-flushing litter box would work better than trying to rig up something to a regular toilet.

He's already trained to use the toilet. No way am I ever going back to a litter box, even a fancy one. They invariably track litter around the bathroom and house.

 

[Quixote_1]

I've been meditating on it and the load cell method would be too complicated. The cat doesn't stay still long enough to get an accurate enough reading to then detect when he drops his payload. I'm back to the idea of image processing and this looks practically perfect:  https://www.atlas-scientific.com/product_pages/probes/ezo-rgb.html

 

[mikefamig]

I've been meditating on it and the load cell method would be too complicated. The cat doesn't stay still long enough to get an accurate enough reading to then detect when he drops his payload. I'm back to the idea of image processing and this looks practically perfect:  https://www.atlas-scientific.com/product_pages/probes/ezo-rgb.html

 
That looks like it will tell you if the water changes color but where would you mount it? If you put it in the water and find a way to route the wire without creating a leak it will be impossible to keep the sensor clean. Even ifinstalled above the water line it will be nasty dirty in a short time.
 
Mike.

 

[LarrylLix]

How about just flushing once per day at 3:00 PM when all cats left alone become nocturnal and are sleeping?

 

[Del91574]

I'd go with a switch on the seat and then put a debounce on it so it takes X amount of triggers or add the timer to the mix as well.

 

[Quixote_1]

That looks like it will tell you if the water changes color but where would you mount it? If you put it in the water and find a way to route the wire without creating a leak it will be impossible to keep the sensor clean. Even ifinstalled above the water line it will be nasty dirty in a short time.
 
Mike.

 
It would be mounted on the seat that I've designed which is a replacement for the lid. The lid basically just has a hole cut in the center with a textured rubber mat inlaid around it to help the cat from slipping in while he hovers over it. I could drill a hole near the back of the seat and point the sensor directly at the water.
...but this brings me to another point. I need to figure out a way to lower the lid in a controlled fashion should a guest come over and raise the lid then forget to put it down. It's already happened once and my cat decided to bring that to my attention by squatting and pissing on some folded vapor barrier on the floor in the kitchen right in front of me. At least it was on a plastic sheet. Anyway, after I figure out the sensor, I need to find a way to lower the seat as well without using a seat that's already designed for that since I've built my own already.

 

[tmbrown97]

This is a hilarious problem to have!  I'm with the others though - I'd try really hard just to detect his presence, then lack of it for X amount of time - then flush.  Even if his remnants stay in the bowl for an hour, I have to think that once he goes, he's done for a while.
 
That said, if you have the time to pursue this and are this determined to get the exact science down, I do hope you'll document it well so we can all follow along!