DIY floats

[GasSavers_maximilian]

OK, just measured everything. The tank is 5' long, 44" wide, and 27" high (that's only 1" over the range of that automotive sender, so a little modification should make it work). It's angled in two ways: 2.75 degrees sideways and 3 degrees lengthwise.

I guess my back is well and truly healed as I had no troubles down there at all. Been avoiding the crawlspace for more than two years. Need to do some plumbing so that's great news!

Right now I'm leaning towards Kyle's magnetic switch suggestion. Since the sensors and all the wires could be on the outside of the sealed tube it just strikes me as a very easy system to tweak. It's also cheap and gives me the excuse to order lots of electronics from the same place while I'm at it. Easy to add that flashing low indicator as well.

Thanks for all the great input. This has been quite fun.

[GasSavers_maximilian]

Just finished running the program to determine oil tank volume versus fluid height (accounting for its angling in two directions). I wound up reusing some code I wrote to convert polygonal models to voxels so it was quick to do. Even though I cranked up the resolution, the increase in computational power in the decade since I developed it kept it very fast. It's a real sign of my skewed skill set that I found it easier to write a program to generate the oil tank model than to use a modeler. Need to get a good CAD program at some point. Anyhow, here's the resultant graph. The total volume agreed to within about a hundredth of one percent of the theoretical tank volume, so that's a great sanity check. Since almost all the code had been tested thoroughly and I really used an overkill resolution I wasn't surprised it matched so closely. I included a linear estimate for comparison.



I wonder what would the curve would look like if I just used the two dimensional shape and scaled it to the tilted height? Sounds like a project for some other time, although it wouldn't be very hard to figure.

[GasSavers_maximilian]

Realized that graph doesn't show what ultimately matters: where to put the sensors.

[jeep45238]

Carb floats are cheap (4 or so) and are designed to live in gasoline, I'm sure they'll survive fine in an oil environment.

[GasSavers_maximilian]

It also has a hollow that might work for placing a magnet. I still hadn't settled on a float, so thank you!

[dkjones96]

Speaking of stupid strong hard drive magnets. How has this project been going?

[GasSavers_maximilian]

Been having a lot of trouble making it work consistently if I allow the float to rotate. The switches have little magnets in them so the poles must be interacting strangely. I'd have to keep the float from spinning, which would be harder to build. Or get different, more suitable magnets. I have a backup plan I want to try out, but haven't gotten around to it yet: micro switches with corks on them. Already have the switches so I just need to stop forgetting to pick up a cork. Wanted balsa wood but I can't get it locally.

[dkjones96]

You've been trying with PVC. What if you took something square, like rain gutter line that is thinner and an odd shape? It could even be non-magnetic tin.

You could then place a couple of magnets with the correct pole facing the right direction.

[GasSavers_maximilian]

Good ideas; I need to play with it some more. I'm taking the path of least resistance on this. If the micro switches are easier, I'll go that way. No need to worry about one being triggered at all times with that approach. Your idea to house the sensor outside the tank really opened up a lot of approaches. If I wanted to use this as a learning experience I'd go optical, since I've never done anything with phototransistors (I tossed a few onto my order just for the hell of it).

[theholycow]

Common rain gutter and downspout material is aluminum, and plastic is also available.