top of page

Sump Pump Pro

Updated: Feb 16, 2022

It all started when our basement flooded. We had the standard hardware store moisture alarm which alerts you when it's already too late to avoid the problem. I was determined not to let it happen again, so the first step was to find the cause. In my case it happened in the spring when the snow started to thaw, but the ground was still frozen. The line leading from the house to the storm sewer froze solid. When the surface snow started melting and the pump started running, the output of the pump could not travel underground, so the pipe at the side of the house just overflowed, sending the water down the side of the foundation, back into the drain tile under that basement and back into the sump.


Eventually, the pump was running almost continuously. It overheated and the thermal fuse cut the power to the pump, allowing all the water to overflow the sump pit and send water all around our nicely finished basement. What a nightmare.


I eventually found a mechanical way to divert the water away from the foundation, but that required human intervention and most home owners don't realize how important sloping the the ground away from the foundation really is. I wanted a technology solution and I found my inspiration from a Make Magazine article about a guy that wanted a text message when his laundry was done. He used a simple current sensor that you past the cord from an appliance through and by measuring the magnetic field cased by current flowing through the cord, you can tell if an appliance is running or not.


I deduced that if I could tell when my sump sump was running, I could measure the number of times it cycled each hour. There would be some baseline value (probably different for every house) but by charting the changes from the norm, I could tell when the pump was cycling faster and send an alert when the pump was running too often. I started with a rough prototype (see below).

I was able to locate an appropriate sensor and by plugging the pump into one of the cables and the other cable into the wall. I was able to wirelessly get the alerts I was looking for. In hindsight, at this point I should have partnered with someone to take this commercial. I had proven the concept, there was clear commercial value and there was a mountain of work that would need to be done to make it a viable product. But I didn't. I kept working on the project by myself and inched it forward, learning all the way, but not moving ahead fast enough to make it a product. Lesson learned.


Without the benefit of the hindsight, I worked through several tough problems. The first was a simple package design. I spent a lot of time looking for an appropriate enclosure. I found the enclosure below and made the second prototype. It was hand soldered and ugly inside, but the outside was clean enough and I started to add important features.

I added a piezo buzzer alarm so if you are in the house and not near your phone, you still here the alarm. The software became quite sophisticated. I had to do some work to get the duration that the pump ran, and properly calculate run frequency. In the end I collected Run Frequency (cycles/hour), Average Run Duration (seconds), Total Run Hours (hrs), Pump Current Draw (amps), among other things. I had alarm limits on all these values. If the pump ran too often, ran too long at one time, had run too many hours and drew too much current (failed bearing) you would know about it.


I started to design an app, but my skills at cross platform app development were limited and doing this at night after a long day of work, I had limited patience for learning something as challenging as Mac xcode and Android's current version of app tools.


I set the project aside for more than a year and I made one more run at this. I found a sexy new enclosure and packed all the component inside. I even went so far as to build customer circuit boards and order enough parts to build 10 of them.

In the end, I had taken so long on my own that the Particle Photon board I designed it for was being discontinued and it would require a new board design and maybe a new enclosure. I finally put the project to rest when I discovered a similar product in Family Handyman magazine that had made it to market.

I still have all my designs and prototypes, and a nagging feeling that I should just make it open source and let other enterprising soles take it from where I left off. I learned a LOT from this project and I still steal bits from the code for other projects.

0 comments

Recent Posts

See All

Comments

Rated 0 out of 5 stars.
No ratings yet

Add a rating

© 2021 bobgeenen.com

bottom of page