Hello all. I’m excited to be apart of this space. I have an idea and need your smarts to help me work through this problem.
Problem
Coolant loss from a vehicle radiator can be catestrophic to a combustion engine. Currently my vehicle only monitors temperature and not coolant level.
Idea
Use a non-contact liquid level sensor to wrap around a coolant pipe (typically made of Ethylene Propylene).
Operation
These pipes are designed to run full of coolant. If there was coolant loss from the radiator system then the driver could be notified with a buzzer or flashing light.
My understanding
Liquid level sensors like the following are great when applied to a flat vertical surface with a distinct water level. I’m not clear on whether they are suited to my proposed application.
As a bit of a car enthusiast myself, I like the idea of this project! The only things that come to mind as ideas or constraints are:
Is it possible to just measure the coolant reservoir, rather than a pipe? or would this not respond fast enough?
Coolant pipes for cars are typically very thick for their diameter, so perhaps these capactive sensors may have trouble sensing the liquid inside… The only way to find out for sure would be to test one, I wish I could say I’ve seen this done before.
Thanks James, I’m excited to be involved in the forum. Are these sensors simple in operation? Are they prone faults or to degradation from heat, salt air, water? Common elements found under the bonnet of a 4x4.
Liquid level sensors are very simple, they will just act like switches that will output either a logic high or logic low signal when liquid is detected.
Flow rate sensors get quite a bit trickier as they have to be tracked over time to count pulses that correspond to the flow rate.
I can see those two sensors are rated IP62 and IP65 respectively, so neither is considered waterproof, and even if they were the IP rating system doesn’t consider salt water or other contaminants’ impact on the hardware so they may need external protection added.
Flow rate sensors would be no good in this application as stere is nil or very little flow until the thermostat opens. I think detecting the level (or lack of) of coolant in the reservoir would be the better proposition as James has suggested.
Cheers Bob
Must admit I went super basic - I put an aluminum pipe in the top radiator hose and welded a thread into it. I then drilled out a plastic bolt and put a metal conductor down the middle of it.
Once the glue was dry holding the metal rod in the bolt, the bolt was screwed into the thread.
A quick switched 12V, led and resistor to the metal conductor and so long as water is in the pipe, to earth the circuit, all is good. Once the light goes out, the top hose is dry.
Nowadays I’d do things a bit more intelligently, but the probe into the water worked a treat.
Another thought would be if you could introduce some kind of float switch into the expansion tank. To remove any sloshing around and putting the alarm on/off/on etc, perhaps you could measure how long the alarm was on for over a 15s period and determine the need to notify the driver accordingly?
Another thought… what’s the relationship between temperature and pressure in the water system when the water disappears? Perhaps consider measuring one or both of these instead.
The simplest way I’ve seen it done is by inserting a metal screw through the topside wall of the rubber radiator hose.
The head of the screw is connected to to a buzzer and the buzzer is connected to the positive battery terminal.
The buzzer is earthed out on an engine block head bolt or even a chassis bolt (earth point). A circuit is created through the coolant/water mix.
When the coolant drops below the tip of the metal screw, the circuit is broken and the buzzer sounds.
The issue I have, is I’m not sure how to create a good long term seal around the screw hole. It’s also not a simple installation. Not everyone is comfortable removing a coolant hose not drilling a hole in one.
How does induction work?
Could you wrap a piece of metallic material around the entire hose and then use induction in some way?
The issue with using the reservoir, is that older vehicles use the reservoir as more of an overflow. I’m not convinced that if the level dropped quickly from a hose break or head gasket issue, that water would be drawn from the reservoir.
Hi All
Interesting. I thought I would bring up quite an important point. You need to be careful when introducing bits of pipe or other parts into the cooling system. If your engine is of the alloy variety you cannot use any copper, like copper pipe etc. There is a reason that the water pipes and radiator bits are plastic or steel. If any copper is used electrolysis between dissimilar metals will cause a migration from the alloy to the copper particularly if any bit of battery voltage finds its way in somehow. If you had a copper radiator you would finish up with a fair bit of the engine blocking up the radiator not to mention holes in the alloy.
All OK if you are just a bit careful.
Cheers Bob
I have seen the float switch as a product for purchase. Like you said it can give false readings if it gets stuck or is sloshing around. Like you said, an electronic brain would be required to manage the false positives.
Mine didnt leak. I took a piece of correctly size aluminium tube and welded a boss into it and threaded it for an M6 (from memory) bolt. I then got an M6 plastic bolt and drilled down the centre of the bolt. A little thread tape to the M6 and silicone around the metal “rod” was good enough.
I do recall thinking that if the silicone didnt hold then I would need to get an M3 screw and tap the centre of the M6 bolts out to accomodate the metal screw down the centre, but I never needed to.
I dont have the vehicle any more otherwise Id take some photos.
This link is for a Landrover kit, but it shows a similar probe to the one I made… I’m sure you could buy the probe only.
Like was said above, a12V source into the pin and the circuit is made through the water and back to ground. Not sure if that’s an induction sensor or not!?!?!?
Hi Chris
I don’t think this is any sort of special “sensor” at all. It is just a conductor making contact with the water completing an electrical circuit to Ground, ie; the engine block. When the water disappears the circuit is broken or open and this is sensed somehow with some simple electronics to show or sound some sort of indicator. This depends somewhat on the conductivity of the water. If the proper coolant is installed this “water” will have additives to prevent freezing and also facilitate speedier warm up of the engine from cold. As I never had any reason to measure the conductivity or the resistance of that fluid I don’t know the answer to the finer details of the sensing electronics but I would think variations in the fluid characteristics would have to be considered. But apparently it has been done successfully.
The “sensor” or whatever you like to call it (probe would be a better term) could be a piece of wire, a screw or anything conductive but I think should be stainless steel to minimise corrosion and electrolysis. Or even the generation of oxygen and hydrogen. Remember that physics experiment with a container of water and a battery??
Metal bits exposed to this coolant seem to accumulate a light brown film over time. I don’t know what the conductivity or resistance is of this stuff but probably would have some effect eventually.
Cheers Bob
