Oil tank Level sensor?

Hi I want to make an oil level sensor for my Heating oil tank and connect it to my PC through an arduino, can anyone point me in the right direction please?

Connecting to the PC should be easy but I cant see the best way to detect the oil levels. I assume most water level sensors would not be suitable?

Mike :-?

Yes, oil is non-conductive so any sensor based on electrical conductivity probably won't work. You still can use float switch or measure the 'head' pressure with a sensitive pressure sensor connected to a bottom pipe tap. Ultrasonic sensors looking at the top level if there is a mounting arrangement that can look down at the liquid level. And last, you can weight the whole tank with load cell sensors and calculate level from that.

That's all that comes to mind for the moment.

Lefty

Thanks a pressure sensor sounds good.

Do you know of one that would mount at the tank outlet?

Have had a google but cant find any.

Mike

You could make a capacitive fuel gauge 0-5V http://www.rst-engr.com/kitplanes/index.htm it's a ultralight aeroplanes page look for entries: Capacitive Fuel Gauge (Part I) -- 2 minutes Capacitive Fuel Gauge (Part II) -- 2 minutes Capacitive Fuel Gauge (Part III) -- 2 minutes June, july, august 2000

The other approach is to use pressure meter: it could be as simple as encasing a simple altitude meter like HP01D http://www.futurlec.com/Pressure_Sensors.shtml inside an airtight can (metal can or plastic) but beeing able to be squeezed a bit, the sensor is placed on the bottom of the tank, the weight of the liquid will squeeze the canister and build the pressure on the sensor. To be sure that the sensor can take the difference in pressure, you suck a little air out of the canister before sealing it. Or when mounted on an outlet pipe at the bottom witth a full tank, then you adjust the pressure of the canister to be at approx 90%max rated for the sensor.

David

The other approach is to use pressure meter: it could be as simple as encasing a simple altitude meter like HP01D http://www.futurlec.com/Pressure_Sensors.shtml

That one won't work, that's an absolute pressure sensor so it would measure the atmospheric pressure plus the oil depth pressure - errors of the order of 1/2 a metre in oil depth as a weather system passed over. You would need a differential pressure sensor - there must be a standard industrial part... Don't see how to connect it unless the tank's totally empty though.

I love that Kitplanes fuel gauge. I didn't realize that the site was still there with the plans. Jim Weir is a great electroniker. I miss reading his columns. Glad to see there are others who have gotten ideas from his work.

I do recall while working in a oil refinery that there was a metal probe level sensor called a capacitance probe that would change capacitance as the liquid level changed and it would work in hydrocarbons such as fuel oil. The probe would see a change of capacitance between it's probe and the ground shell of the tank as the liquid covered or uncovered more of the probe, would change the dialetric strength of the 'capacitor'. The external electronics would use the capacitance to control the frequency of a oscillator and then a frequency to voltage converter would output the DC measurement voltage (or current in their case, as all measurements used the 4-20ma DC standard).

Not cheap as most industrial instrumentation carries so many safety ratings and built to heavy duty requirments, but the concept is simple and could be done DIY I bet.

Lefty

Those capatance probes... My Process Control teacher told us that the pressure probes use piezo-crystalls but not as a "voltage thingy" but as a resistive sensors, the pressure only gives momentarely voltage but the resistance alters with pressure on the crystal so with active current running through you can mesure continually, there should be easy way to test this using piezo-buzzer ;)

@MarkT ...and about the absolute pressure sensor, you just keep one on the top of the tank as a referance = 0 altitude and then you have a dp-sensor. (better yet, on the outside of the tank in the same height as the other sensor) ;)

David

Hey, not have any experience in this topic but if they say that the capacitance changes with a probe, (just a thought) how about trying the capacitive touch sensor library that they have (http://www.arduino.cc/playground/Main/CapSense) as it gives resonably large numbers (hopefully accurate)..

Hope this help at all

:)

the cap-sense might work with a metal rod wrapped with one layer of ptfe-tape (thread tape) and metal tape on top and you have a long capacitor to use with the cap-sense library

David

Here's an article from nuts and volts showing how to do it using a basic stamp, should be easy to convert to Arduino:

http://www.parallax.com/Portals/0/Downloads/docs/cols/nv/vol1/col/nv27.pdf

Mike
Those solution all sound good, my suggestion for a KISS solution would be ultra sonic sensor from the top of the tank aimed onto the level of the fluid…

Depending on the shape of the tank and the distance the oil level drops, gives you the range of the sensor to use.

(finally a question I can offer an answer, rather than keep asking for help!) ;D

@ArduinoM: Be very careful about using any sort of tape in an oil tank. If the oil causes it to partly break down, or breaks down the adhesive, it will be a real b*tch to clean scraps out of the tank. Even worse if scraps get sucked down the outlet pipe.

Also while that construction method would create a capacitor, the oil would be on the OUTSIDE of the capacitor. It needs to be between the two electrodes in order to substantially affect the capacitance.

Ultrasound is probably easiest. Pressure is the most common method for industrial instrumentation, requires access to the bottom of the tank. Is this an above-ground tank or buried?

Your're right ;) lose my objectives sometimes ;) but the method in the http://www.parallax.com/Portals/0/Downloads/docs/cols/nv/vol1/col/nv27.pdf should work like a charm ;) it's what I probably remembered from somwhere but a bit distorted :) Cause I have seen this kind of sensor.

David

Also while that construction method would create a capacitor, the oil would be on the OUTSIDE of the capacitor. It needs to be between the two electrodes in order to substantially affect the capacitance.

In the level capacitance probes I was talking about, the probe was a single conductor probe. It uses the walls of the vessel as the other 'plate' and the liquid acted as the dialectic between the probe and the wall of the vessel. It did require the vessel to be metal, it wouldn't work with a plastic tank. The probe was as I recall made with a single conductor wire with a Teflon coating. It was truly a capacitance type sensor as calibrating the electronic sending unit involved disconnecting the sensor from the electronics and temporarily connecting know valve capacitors to the input of the sender unit to simulate various level conditions.

PS: Drexelbrook was one brand name as I recall. Lefty

With the ultrasound solution I referred to earlier, there is a unit in the UK that has the sensor, tx link and receiver for around £50 called a Watchman, designed for monitoring the level of oil tanks. It might be less work overall to open up the receiver and interface that to the PC if you don't want it standalone?