Automatic Litterbox Central

http://www.instructables.com/id/Cat-Gen ... nt-electr/

Instructable posted. Lots of pictures and descriptions.

ku4zs wrote:Very nice. It occurred to me that an Arduino could do something similar using analog inputs and programming in a setpoint. That way you could output via ethernet or serial port the CG usage, include a resetter, and various other things. For example, if you want the bowl to fill up higher, then let the Arduino have a timer started after the water hits the sensors.

Agreed, Arduino seems like a pretty good match up for doing extra curricular activities with the C.G. In my case, I had these PSoCs sitting around and really hadn't done anything with them. My goal is to interface the Freshmatic aerosol dispenser to spray a few times after each cleaning cycle

ku4zs wrote:http://www.instructables.com/id/Cat-Genie-three-beep-solution-replacement-electr/

Instructable posted. Lots of pictures and descriptions.

Looks great! Thanks for posting, I'm sure a few others suffering from the poor design of the stock water sensor are grateful!

This has the IR LED correct and J2 has more information on it for those who could not see that 12 v meant 12 volts positive goes here and GND meant the negative from a ps went there. We all aren't electronic techs, and I forget that at times.

This is really a simple circuit, but if I used something else to build it with, I probably would have used Paralax's BasicStamp, since I have many of those around. I would redo the entire Genie's proc with that.

2cats2dogs wrote:



This has the IR LED correct and J2 has more information on it for those who could not see that 12 v meant 12 volts positive goes here and GND meant the negative from a ps went there. We all aren't electronic techs, and I forget that at times.

This is really a simple circuit, but if I used something else to build it with, I probably would have used Paralax's BasicStamp, since I have many of those around. I would redo the entire Genie's proc with that.

Good updates on the schematic...

CORRECTION: my transistors are 2n2222, not 2n2222A! I just looked to verify and they are not the A version. Probably doesn't matter, I see on Google searches that the A is just an improved version and can be used in circuits with the non-A version.

a capacitive sensor would emiminate all of those problems with corrosion on a conductive sensor and contamination on a photo-optic sensor.
A capacitive sensor can be made with two strips of PC board and sealed with a coating to protect from corrosion.

jfischer2 wrote:a capacitive sensor would emiminate all of those problems with corrosion on a conductive sensor and contamination on a photo-optic sensor.
A capacitive sensor can be made with two strips of PC board and sealed with a coating to protect from corrosion.

Yes a capacitive sensor would work too, but you would also have "gunk" that would accumulate on the surface of the PCB. Really in any case you will have sensor maintenance.

My current solution is having petnovations replace my processing unit. We'll see how that goes. Your solution is next in line though as long as it can be attached without voiding the warranty. I think its a great solution and a better implementation than petnovations current water sensor.

How about Sonar?

http://www.robotshop.ca/dfrobot-urm-ult ... ensor.html

JasonT

Or a float with a ball switch?

Niosh wrote:My current solution is having petnovations replace my processing unit. We'll see how that goes. Your solution is next in line though as long as it can be attached without voiding the warranty. I think its a great solution and a better implementation than petnovations current water sensor.

Good luck, but I doubt replacing the head unit will help, unless you really do have a hardware problem.

As far as other sensors, there's probably a million ways to electronically detect water levels. The advantage of a "simple" solution is that it is easy to work on. Think of the old model cars, without alot of fancy computer controlled systems, etc. The home mechanic could work on them because of their simplicity. What happens if your sonar sensor starts going nuts, how will you diagnose the problem, buy another one?

Just something to think about, simplicity is simple.

I will be building a kit this weekend for some lucky CG owner, just as a test run. First person to PM me can lay claim on it. After researching parts prices, I think $35 is fair for the kit, including free Priority USPS shipping (The power supplies are more expensive than I thought they would be). These kits are just for those folks who don't have the expertise or time to build one. I will just build these using the same proto board like what I have in my photos, but it will be cleanly soldered and built.

ku4zs wrote:Good luck, but I doubt replacing the head unit will help, unless you really do have a hardware problem.

As far as other sensors, there's probably a million ways to electronically detect water levels. The advantage of a "simple" solution is that it is easy to work on. Think of the old model cars, without alot of fancy computer controlled systems, etc. The home mechanic could work on them because of their simplicity. What happens if your sonar sensor starts going nuts, how will you diagnose the problem, buy another one?

Just something to think about, simplicity is simple.

It seems like a hardware problem to me, but it could just be a design issue. Right now it errors out after every run. I can sometimes get it to fill with water once, but after that it will always think it's full of water until I mess around with it. Here is a picture of my current "dirty" water sensor.

My CG just finished baking and gave me the 3 beep error, I took this out, cleaned it, put it back in, and now I get the 1 beep "dirty or missing sensor" error. And I've been trying both sensors that came with the unit. What really boggles my mind is how the unit still attempts to dry itself after having this 3-beep/clog error. If the unit really is clogged and full of water why the hell would you want to run the heater on it for 30 minutes and still have it error out and think its full of water. The error handling is what really makes me hate this machine. That and the whole leave us a message and we'll maybe call you back sometime in the next 36 hours.

I would love it if they just had a simple ball float to detect water levels. You could even semi-accurately detect the actual level if you wanted to go crazy with it (aside from the current accuracy of full or not full). I've had zero ball floats fail on every toilet I've ever owned. I've also never had to clean one or had one falsely report that the tank was full when it was empty.

Niosh wrote:

ku4zs wrote:Good luck, but I doubt replacing the head unit will help, unless you really do have a hardware problem.

As far as other sensors, there's probably a million ways to electronically detect water levels. The advantage of a "simple" solution is that it is easy to work on. Think of the old model cars, without alot of fancy computer controlled systems, etc. The home mechanic could work on them because of their simplicity. What happens if your sonar sensor starts going nuts, how will you diagnose the problem, buy another one?

Just something to think about, simplicity is simple.

It seems like a hardware problem to me, but it could just be a design issue. Right now it errors out after every run. I can sometimes get it to fill with water once, but after that it will always think it's full of water until I mess around with it. Here is a picture of my current "dirty" water sensor.

My CG just finished baking and gave me the 3 beep error, I took this out, cleaned it, put it back in, and now I get the 1 beep "dirty or missing sensor" error. And I've been trying both sensors that came with the unit. What really boggles my mind is how the unit still attempts to dry itself after having this 3-beep/clog error. If the unit really is clogged and full of water why the hell would you want to run the heater on it for 30 minutes and still have it error out and think its full of water. The error handling is what really makes me hate this machine. That and the whole leave us a message and we'll maybe call you back sometime in the next 36 hours.

I would love it if they just had a simple ball float to detect water levels. You could even semi-accurately detect the actual level if you wanted to go crazy with it (aside from the current accuracy of full or not full). I've had zero ball floats fail on every toilet I've ever owned. I've also never had to clean one or had one falsely report that the tank was full when it was empty.

Yea your sensor doesn't really look dirty. But, if you take the sensor out, clean it, and the unit does it's first fill properly then stops filling for the rest of the cycle, you can assume that the sensor is involved. If it was entirely hardware, doing a sensor clean should have no effect on the results, but you say it does. It could be a combo of both, as well.

I found I had to clen the top and the Led and IR LED last time to end the beeping. Seems some stuff floats in the air and gets up there. Right now, I went three weeks and was greeted by yelling cats this morning with the CG doing the 3 beeps. I have to get to that in a minute. I don't have the parts right now to build the circuit but it is top on the list. Hardest part is the stainless steel screws. I just can't find them.