Detection of boiler water levels

I have been thinking about this topic and tried out some ideas:


The Lykamobile series 2 (“Lyka2”) provides for the driver to monitor boiler water levels using the sight glass, in common with established practice for steam engines, steam cars and locomotives.  Modern static steam generating facilities are required to use automatic means for monitoring and responding to boiler water levels.  It is a sensible argument that mobile steam generation should do the same (but don’t push for this as a new regulation to haunt existing steam locomotive users). 

As a desirable objective, the various means of detecting boiler water levels should be considered:

-        - Conductive detection:

-        - Physical level detection;

-        - Capacitive detection;

-        - Optical detection.

Conclusions on the above methods in the context of Lykamobile construction have been subject to my preliminary observations.

1.      Conductive

The original Lykamobile used a set of conductive probes in a separate vertical copper tube, in the position normally used for the optical sight tube, to determine boiler water levels at pre-set points.  The plumbing is an extra for Lyka2.  The electronics is not complex, but requires some expertise for design and assembly.  Some users have reported using spark plugs as suitable detectors, as they isolate the detection probe and provide for a seal against pressure.  I had some left over from the car I maintained 50 years ago!  At room temperature and pressure, their resistance measured >20Mohms when dry and ~400kohms when immersed in water. A simple detection circuit could be arranged to drive the burner relay and a warning light.  I have no information on the electrical characteristics of water and steam at 200+ psi, so cannot assess if my assumptions would be valid.

2.      Physical

a.      Pressure

Some modern systems use a differential pressure monitor to measure water height – I have not investigated this option – it looks a bit complicated for an ”Add-on”

b.     Flotation

It should be possible to attach a magnet to a float in the sight glass (or a copper tube in parallel) , and then use a reed switch to detect the position of the magnet -  no extra plumbing is needed for the sight glass.    I am not sure about the material for the float.  Many plastics sink; others melt at the temperature reached by water boiling at 200+ psi.  When I put expanded polystyrene in my cooker at 220deg C, it started to melt and shrink, but remained buoyant.  Polypropylene (the lid on my Swarfega container) just floats, and seems to have a tolerable melting point.  I would worry about putting plastics in the sight tube;   we could end up with gloop to block the waterways, or re-inventing the 1960s-style lava-lamp but for high pressure boiling water.  Perhaps the plastic could be contained in a metal cap from a water bottle (e.g. tonic – let’s all have a gin and tonic for experimental purposes).  This looks like a possibility.

3.      Capacitive

Some modern systems rely on the capacitive properties of water. I have not investigated. It seems to involve the use of an a.c. detector voltage.

4.      Optical.

Both refractive index and opacity vary between water and steam.  These could be used to trigger a response when a certain water level is reached in the sight tube.


I have found that a glass container will not trigger a response from my battery operated soap dispenser, but when filled with water, the same glass container will set off the trigger.  I can only conclude that the detector and electronics respond to water as if it were opaque – quite possible if the source were an infra-red LED.  To buy and re-work such a dispenser (from Reckitt) might be a solution.  There would be no need to re-work the Lyka2 pressure pipe-work.

If a black and white pattern is placed behind a half-full bottle, then the refractive index of water causes the pattern to move (as in a magnifying glass); careful alignment of an optical detector might be used to measure water level.