Low Coolant Alarm (installed photos etc)

Defender200Tdi · 22nd June 2007, 09:25 AM

Some time ago I fitted a low coolant alarm to my Defender, and since my version is slightly different to some, I thought I’d share the ideas and reasoning.

I’ve always had an issue with trying to detect a low coolant state either in the thermostat housing or the top radiator hose. When the engine is running, these are the two worst places to be looking for low coolant level because of the operation of the water pump. In our Land Rover engines, the water pump draws coolant in through the bottom radiator hose and pushes the coolant from this hose up through the engine and out through the thermostat housing into the top radiator hose. If the coolant level was low it will show up in both the radiator and the expansion tank, before a low level was experienced in the thermostat housing or top radiator hose (whilst the water pump is operating). It would only be when the pump was stopped that the low coolant level would be equal in both the expansion tank, thermostat housing and radiator. This is in fact done deliberately by most manufacturers to ensure that the engine receives its full level of coolant whilst in operation, even if the system is a little low. The low level is carried in the radiator, not in the engine. Only when the level is so low that it falls below the level of the water pump (causing the pump to cavitate) will the engine not be full of coolant whilst the pump is operating.

As I said above, this is the system for the majority of engines. A very few have a reverse coolant flow where the thermostat is mounted below the water pump, and the pump pushes coolant down into the bottom radiator hose, up through the radiator and into the engine. In this system coolant level is ultra critical because any low level is carried in the engine (at cylinder head level) and the radiator will always remain full whilst the pump is in operation. The only example that comes to mind is the 12valve 2.4 Nissan engine used in the '90's in Pintaras and Ford Corsairs.

I decided that the expansion tank is the most convenient of the correct places to instal a sensor and initially went down the route of using one of those Range Rover caps that incorporated a level sensor. However, the Rangie cap sensor proved to be quite temperamental, so I searched for a more reliable solution. Enter the simple and effective float switch I discovered on www.4x4gadgets.com.au. This company sells the Little Black Box system which is very comprehensive, but I was only looking for the coolant level sensor. After a quick phone call I confirmed that the sensors were for sale separately (about $25 IIRC), and a plan was formed in my head.

Here’s the coolant level sensor, a straightforward level switch, easy to install with minimal chance of gremlins:

Installing the switch was easy, simply drill a 22mm hole in the side of the plastic expansion tank, screw in the sensor and wire it up:

Here it is installed in my 200tdi Defender:

Here’s one installed in a TD5:

And another one in a Series II Disco:

Next we come to the slightly more complicated part. The sensor is just a switch, and as it is it’d work fine in an on-road vehicle, but our vehicles may occasionally spend some time at a slight angle, or bouncing around a bit.

This would inevitably lead to false alarms as the coolant sloshes around in the tank giving the sensor a hard time. Obviously what is needed is some sort of time delay. At this point my fellow Land Rover suffering mate, Pete, came to the fore. He put together a simple and robust little circuit from fairly commonly available components that achieves our objective i.e., a short (adjustable) delay before the alarm is triggered to prevent false signals when off road.

Firstly, here’s a couple of photos of the completed circuit:

Now, here’s the circuit diagram:

Finally, here’s Pete’s description of the circuit for those of you who wish to replicate it:

This circuit is setup to provide an adjustable delay from the sensor located in the header tank when the level drops.
It works using a basic 555ic timer circuit. When the ignition source is initially turned on pin 2 is triggered via a short pulse from the C2 and R2 combination. This then allows C1 to charge via VR1 and R1. If the tank switch is closed (the tanks is full) C1 is held to ground and a small current flows through R1,VR1 and the tank switch.
This will remain like this until the tank switch is opened or the ignition is turned off.

In the event that the tank switch opens C1 will begin to charge through VR1 and R1. When C1 gets to 60% of the ignition voltage pin 3 on the 555ic will be grounded causing Rly1 to turn on and sound the alarm.

I also had to add a suppressor inline with the supply to prevent false alarms when heavy load equipment is started and stopped. This was typical fans and windscreen wipers.

My setup was set for about 8 seconds of delay. The components I used were what I had laying around but seem to give a good range and do not draw much current through the tank switch.

Component list

R1 – 33k
R2 – 33k
VR1 - 1M
C1 - 4.7 uf
C2 - Green cap unsure but not that critical
D1- IN4001
Rly1 - Standard 12volt relay

Now Pete is a member of this forum and goes by the username of Peterla, so if you have any particularly techo type questions, either PM him direct, or PM me and I’ll pass them on.

I guess the actual alarm is the final part of the puzzle. I simply used the output from the circuit to trigger a standard relay, which is in turn connected to a piezo buzzer and a light. The buzzer is tucked behind the dash, but is easily loud enough to be heard over the racket made by a 200tdi Defender at 110 kph. The light is mounted on the centre dash thus:

There you have it. Simple, reliable and effective and it's been running in my Defender for some time now without issue.

Paul

**waynep** · 22nd June 2007, 10:26 AM

That's a very good solution for DIYer's. Couple of questions ;

1. would a solid state sensor as used in the Engine Saver unit be able to be used in this design too ? Would there be any advantage in solid state over an electro-mechanical sensor switch ? I assume the solid state sensor merely changes its resistance to earth when in/out of water.

2. you say ( I think ) that the first indication of low coolant level is detctable in the expansion tank and the radiator. As far as I can see the EngineSaver design puts the sensor in the top radiator hose which essentially monitors the level in the top tank of the radiator anyway. Would this not therefore be an acceptable place for it ?

Just askng because I have an Enginsaver in the top hose and if there is an issue with it actually detecting a real low coolant problem in landies I'd like to know.

Defender200Tdi · 23rd June 2007, 11:49 AM

Originally Posted by waynep

That's a very good solution for DIYer's. Couple of questions ;

1. would a solid state sensor as used in the Engine Saver unit be able to be used in this design too ? Would there be any advantage in solid state over an electro-mechanical sensor switch ? I assume the solid state sensor merely changes its resistance to earth when in/out of water.

2. you say ( I think ) that the first indication of low coolant level is detctable in the expansion tank and the radiator. As far as I can see the EngineSaver design puts the sensor in the top radiator hose which essentially monitors the level in the top tank of the radiator anyway. Would this not therefore be an acceptable place for it ?

Just askng because I have an Enginsaver in the top hose and if there is an issue with it actually detecting a real low coolant problem in landies I'd like to know.

1. It depends on the signal from the solid state sensor, but my initial thoughts are yes, it should work fine. I can't really thinnk of any advantage of one type of sensor over the other.

2. No, the top radiator hose will still have water flowing through it, even when the radiator is half empty. Think of the radiator as a tank being filled by your water pump (on the engine) via the top radiator hose. Or another way, imainge a fish aquarium with a filter pump running. You can have a full flow of coolant through the top hose when the engine is running but a low level in the radiator. The coolant flow will only be reflected in the top hose when the level has dropped so low that the pump begins to fail. However, if the engine was switched off, then yes, the low level would be reflected in the top hose too.

Paul

**Dave110HardTop** · 29th June 2007, 12:31 PM

Originally Posted by Defender200Tdi

2. No, the top radiator hose will still have water flowing through it, even when the radiator is half empty. Think of the radiator as a tank being filled by your water pump (on the engine) via the top radiator hose. Or another way, imainge a fish aquarium with a filter pump running. You can have a full flow of coolant through the top hose when the engine is running but a low level in the radiator. The coolant flow will only be reflected in the top hose when the level has dropped so low that the pump begins to fail. However, if the engine was switched off, then yes, the low level would be reflected in the top hose too.

Paul

You know, after reading this I'm starting to think 2 sensors, one in the top hose, and one in the tank might be the safest option...
cheers
Dave

**matbor** · 29th June 2007, 12:46 PM

Originally Posted by Defender200Tdi

1. It depends on the signal from the solid state sensor, but my initial thoughts are yes, it should work fine. I can't really thinnk of any advantage of one type of sensor over the other.

2. No, the top radiator hose will still have water flowing through it, even when the radiator is half empty. Think of the radiator as a tank being filled by your water pump (on the engine) via the top radiator hose. Or another way, imainge a fish aquarium with a filter pump running. You can have a full flow of coolant through the top hose when the engine is running but a low level in the radiator. The coolant flow will only be reflected in the top hose when the level has dropped so low that the pump begins to fail. However, if the engine was switched off, then yes, the low level would be reflected in the top hose too.

Paul

An example of the engine saver working, from someone else that has one installed :

My own alarm went off yesterday as I had lost about 60ml of coolant due to a loose hose clamp.

It varies with exactly where the probe is positioned, but they are very sensitive because as soon as a small amount is lost an air bubble is formedin the hose.

My own went off as I applied the brakes and stopped as I accelerated again.

Tightened the clamp, topped up and all is well again.

**Pedro_The_Swift** · 29th June 2007, 01:54 PM

Originally Posted by waynep

2. you say ( I think ) that the first indication of low coolant level is detectable in the expansion tank and the radiator.

First up
Nice job Paul

secondly,
a question from the more engineering challenged amoungst us,,,

the expansion tank---
I always thought it was just that,
a place for fluid under pressure to escape to,,
and when cool it transfers back into the system,,

but for this system to work, the expansion tank fluid has to "fill" any space left by escaped fluid, which in turn will stop the temp rising untill no expansion tank fluid is left, THEN the sensor will operate,,

how am I going so far

Defender200Tdi · 29th June 2007, 06:38 PM

A lot of people confuse the Land Rover system with the old type expansion tanks used back in the 70s & 80s. Those systems never had the expansion tank as part of the pressurised system, they simply caught the overflow from the radiator via the pressure release built into the radiator cap. This flowed out into the expansion tank through a thin tube, the end of which was supposed to be below the coolant reservoir in the expansion tank. The radiator in these systems was designed to be completey full i.e., no air gap whatsoever. When the system heated up and expanded, some displaced coolant pushed past the pressure release in the cap and out into the expansion tank. Later, when the system cooled down and contracted, a vacuum was created in the closed system and the radiator was forced to draw the coolent back from the expansion tank via a one way valve in the radiator cap. This is really just a coolant recovery system that relies on the integrity of the main cooling system to work. If you hole a radiator, split a hose or open the main part of the system up in any way, it will cease to draw coolant back through the one way valve in the radiator cap, and instead draw air in through the easiest route. The level of coolant in the expansion tank is no indication of the level of coolant in the main system.

Our Land Rover system is quite different. The expansion tank is part of the main pressurised system, and whilst it contains the air gap for the system, it always reflects the level of coolant in the main system. Consider the drawing below:

In the above image you can see how the expansion tank fits into the system. Coolant actually flows into this tank via the branch from the bottom radiator hose. Coolant from this hose flows up the branch into the bottom of the tank, but only when you are filling up the system or when the system is heating up and the expansion needs to force the coolant out somewhere. Since the air gap is in the expansion tank, it's the air that's easiest to compress (compared to the coolant or metal of your engine), so the air compresses slightly and coolant pushes up throught the bottom hose.

You can also see that there is a second hose running from the top of the expansion tank to the top of the radiator (ignore the Y piece for the 300tdi shown for the time being). These top and bottom hoses enable the expansion tank to level out with the coolant in the radiator (and indeed the rest of the coolant level throughout the engine when the pump isn't running) - simple gravity will see to that.

Now, what happens when you start the engine? The water pump is mounted on the engine about half way up. Provided the coolant level is above half in the radiator, thanks to gravity we know the pump will also be below coolant level, so the pump will start drawing coolant in from the radiator, through the bottom hose, push the coolant up through the engine and eventually out past the thermostat and into the top radiator hose, where it will then spill into the top of the radiator. No coolant flows through the expansion tank when the engine is running, except for the "expansion".

The important thing to consider is what happens when the engine is running and you loose some coolant (split a hose etc). Like I said in the post above, think of the radiator like a tank being filled by a pump via the top hose. If the coolant level drops, the water pump will still draw a full flow of coolant via the bottom hose (until the level gets VERY low), and push a full flow of coolant up through the engine and out through the top hose. The system is still a closed system though, so the leaking coolant will have to be replaced from somewhere, and the first way it is replaced is with small vapour bubbles from the boiling coolant, that are now generated by the reduced pressure in the system. These bubbles flow through with the coolant until they reach the radiator, where they will collect as an air gap at the top which then flows through that top little hose to the expansion tank as gravity equalises the coolant level in the expansion tank and the radiator via the bottom hose. Phew!! If you're still reading I'm about to get to the point.

We now follow how the expansion tank reflects the level of the radiator, and how the radiator (and hence expansion tank) carry the loss of coolant. We can also see how (other than the initial vapour bubbles) anything between the water pump and the final spill from the top hose into the radiator, will be blissfully unaware of any loss of coolant. Obviously, once the coolant level in the radiator gets so low that it is below the level of the water pump, gravity will stop the pump being immersed in coolant, the pump will cavitate and cease to work and the whole system will need a re-boot.

One final point: the expansion tank seam is mounted level with the top of the radiator, so that the expansion tank is the one carrying the air gap, not the radiator or any part of the engine (ignoring the first fill air locks etc).

Hope that helped.

Paul

**97discotdi** · 29th June 2007, 07:00 PM

I am wondering what was the problems you encountered using the RR expansion tank level switch. The reason I ask is I have been using the RR switch for about 4yrs and untill now have had no problems. Your level switch looks pretty good as it is installed quite low in the tank. I also used the piezio buzzer, too easy to miss seing a light and the missus will soon get annoyed with the noise and pull over.

**Pedro_The_Swift** · 29th June 2007, 07:08 PM

Great response Paul,,
makes it seem a very sensible arrangement.

the big question is,,,,

how do I get one,,,,

Defender200Tdi · 29th June 2007, 08:08 PM

Originally Posted by 97discotdi

I am wondering what was the problems you encountered using the RR expansion tank level switch.

Just inconsistent operation from the ones I tried. Sometimes they'd work, sometimes they'd give off false alarms. Perhaps I was just unlucky...

Paul

Thread: Low Coolant Alarm (installed photos etc)

Thread Tools

Low Coolant Alarm (installed photos etc)

Bookmarks

Bookmarks

Posting Permissions