miniature boiler sight glass low water alarm.

the old fart

Senior Member
Hi Guys,

I need to detect the water level in a sight glass for a model steam boat.

I could use a probe fitted into the tank, but would prefer to clip 'something' onto the sight glass to detect water low level.

The sight glass is 5mm o/d.

Any idea what the 'something' could be?

thanks
TOF
 

rq3

Senior Member
My first thought was optical (LED and phototransistor), available as a slotted unit. Might be an issue with clear glass and clear water. Can you lightly dye the water?

Second thought, a small magnet on a small cork floating in the tube, with a Hall effect sensor or maybe even reed switch?

Third thought, metal foil around tube at the appropriate location, attached to a Picaxe "touch" pin.
 

the old fart

Senior Member
Hi rq3,
thanks for your input.
water needs to be clean and pure, the steam engine is over a grand.
a cork, sight glass is 5mm outside diameter.

led/photo transistor could work, I'm hoping that someone has already done some experimentation.
tof
 

AllyCat

Senior Member
Hi,

If you shine light (or IR) from a LED through the glass (refractive index say 1.5) then about 5% (IIRC) will be reflected off the inner surface when air is present. That should fall to approximately a half when water (refractive index 1.3) is present, which you may be able to detect reliably. It's the basic method used by motor vehicle automatic windscreen wiper controllers, but they are probably able to achieve better control of the optical beam(s).

If you can shine the light at an angle then the amount reflected will increase, but 5mm is a rather small window, how thick is the glass? Some LEDs have a very narrow beam width (just a few degrees) which may help with both sensitivity and beam control.

Cheers, Alan.
 

the old fart

Senior Member
Hi Alan,

Not sure on glass thickness.

I have some I/R transmitters and receivers, 3mm diameter. (used same on model railway a long time ago).

It's a starting point for me to explore.

TOF
 

mikeyBoo

Senior Member
A plastic sphere with a magnet inside is my favorite (from industrial days). Can be picked up by hall effects. However, sounds like you don't have much room to work with. Maybe those little plastic BB things (can't remember what they're called).
Let us know how you solve the problem, sounds interesting.
 

elanman99

Senior Member
I have no experience relevant to detecting the presence of water or air in a glass tube that is only 5mm OD but its sounds quite a challenge.

When I say the question I wondered where a Picaxe was involved but I now realise that once you find a way of detecting the water you will find that quite a bit of digital logic may be needed.

I presume you are trying to make sure that the water is always above a certain level and not actually trying to measure the water level?

I think that if its a float based method it will need say, three, across-tube beams, so its possible to determine which way the float has moved. With only one beam the float could be above or below the beam.

Ian
 

mikeyBoo

Senior Member
You need a lo-level probe to latch-in your fill valve. The hi-level probe will unlatch it (cut if off).
Three pieces of thin stainless wire will work. The water conducts between the probes (same principle as a touch switch). Seems like I remember the Picaxe used as inputs are pretty hi-impedance, so you probably won't need an op-amp. If my memory is correct, you just tie the hi & lo probes to high (+5) each via 10 Meg resistors. The third probe connects to Common (aka ground). When the water is across the common to hi or lo probe, it pulls the input lo.
Tried to explain without drawing a picture, hope this makes sense for you.
 

rq3

Senior Member
Hi Alan,

Not sure on glass thickness.

I have some I/R transmitters and receivers, 3mm diameter. (used same on model railway a long time ago).

It's a starting point for me to explore.

TOF
Or merge the two ideas? Float a black plastic bead on the water. That should make optical detection pretty darn easy. With two detectors you could easily determine if the water level was falling and needed topping up, or was rising and was full enough.
 

the old fart

Senior Member
Thanks for all the input,

I'll try an IR detection through the glass tube less destuction to brass tank.

Only need low level, then alarm out via picaxe, possible some lound annoying 'tune', and a bright led.

If all else fails, then a probe in the tank, and use the 'touch' command.

Tank capacity is about a pint.
TOF
 

techElder

Well-known member
It makes a difference whether you have access to both sides (180 deg. apart) of the sight glass. A close up photo of the arrangement would help.
 

AllyCat

Senior Member
Hi,

I'll try an IR detection through the glass tube less destuction to brass tank.
A problem with IR is that you can't directly "see" what's happening. But many digital cameras are sensitive to IR, which may help. Ceratainly it would be much easier if you have access to the inside of the tank, but to expand on my suggestion in #4. :

Light will be also reflected from the front face of the glass so you will need a barrier betweeen the LED and photosensor. But simple plastic tubes (e.g. 3mm black sleeving) touching the front face may be all that's needed. Optical systems often use transparent "glue", but that may modify the beam width of the LED/sensor, adding to the unpredictability.

Fresnel's Equations indicate that about 4% of the light will be reflected at a glass-air (or air-glass) interface when there is "Normal incidence" (at right angles to the surface - ignore all the maths before that) and less than 1% at a glass-water interface, which could be easy to discriminate. BUT we need to consider what happens to the other 99% of the light. If the inside glass surface or the water is "dirty" then light will be scattered backwards, which might be similar to (or more than) that reflected back by the glass-air interface. :(

However, you have said that the water is clean, so where does the light go? If the inside of the tank is "bright" brass, then light shining directly in through the window may be refelected straight back, so you probably need to angle the LED and sensor slightly. That's where the beam widths could become important and a fair amount of experimentation may be required. Good luck.

Cheers, Alan.
 

the old fart

Senior Member
Hi,



A problem with IR is that you can't directly "see" what's happening. But many digital cameras are sensitive to IR, which may help. Ceratainly it would be much easier if you have access to the inside of the tank, but to expand on my suggestion in #4. :

Light will be also reflected from the front face of the glass so you will need a barrier betweeen the LED and photosensor. But simple plastic tubes (e.g. 3mm black sleeving) touching the front face may be all that's needed. Optical systems often use transparent "glue", but that may modify the beam width of the LED/sensor, adding to the unpredictability.

Fresnel's Equations indicate that about 4% of the light will be reflected at a glass-air (or air-glass) interface when there is "Normal incidence" (at right angles to the surface - ignore all the maths before that) and less than 1% at a glass-water interface, which could be easy to discriminate. BUT we need to consider what happens to the other 99% of the light. If the inside glass surface or the water is "dirty" then light will be scattered backwards, which might be similar to (or more than) that reflected back by the glass-air interface. :(

However, you have said that the water is clean, so where does the light go? If the inside of the tank is "bright" brass, then light shining directly in through the window may be refelected straight back, so you probably need to angle the LED and sensor slightly. That's where the beam widths could become important and a fair amount of experimentation may be required. Good luck.

Cheers, Alan.

Hi Alan,

Sight glass tube is external to the tank. no light shining into tank. and its 5mm diameter.
will make a clip on, black plastic, bracket which will hold the 3mm i.r diodes.
lots of experimenting before I admit defeat and drill holes in the tank for probes.
TOF
 

AllyCat

Senior Member
Hi,

... no light shining into tank.
The "light" that I was referring to is the (IR) light that your LED is shining into the sight glass tube. We have to hope that 99% (literally) of that gets "lost" inside the tank and doesn't come out again. ;)

Cheers, Alan.
 

Dartmoor

Member
I assume that the reference to "tank" means the boiler, rather than a water tank to supply water to the boiler?
There are several options that I have seen work:
#1 a metal (foil) around the sight glass that (I believe) is used to monitor capacitance. Maxitrak used to sell a sensor like this?
#2 metal probes (as suggested in post #8). These can be made with a copper/stainless steel bolt through a ptfe bush. To avoid drilling holes, these could possibly replace the gauge glass using the two bushes that already exist in the boiler? Three indications - level is below bottom sensor, between sensors, or above top sensor. There was a design for this in Model Engineer in the 1970's.
#3 temp sensors fitted as probes. I have not seen this done but should work in theory. Temp of the steam will be higher than temp of the water. I have seen a manual version of this in a full size boat built by two blind engineers, where a copper rod with studs was fitted parallel to the gauge glass. They could tell the water level by touching the studs & feeling the temp!
Have fun!
 

sages

Member
Measure the boiler temperature and when it's run out of water the temperature will rise rapidly :cool:
Or measure the rpm and throttle setting, rpm getting less and no throttle change probably means no water left. :eek:
 

AllyCat

Senior Member
Hi,

Ah, a picture is worth a thousand words. I had assumed the "sight glass" was a "port hole" window into the tank itself. :confused:

A tube like that may act like a (cylindrical) "convex lens" with a varying focal length dependent on whether it contains water or not. So a LED and sensor on opposite sides may detect sufficient variation in the light transmission (particularly as the meniscus passes through). But, with such an arrangement open to ambient illumination, you may need to modulate the light beam.

A PICaxe "touch" input might be able to detect the water as a change in the dielectric between two "capacitor" plates. But that needs a detailed understanding of how the touch circuits (and water as a dielectric) behave in practice.

Cheers, Alan.
 

the old fart

Senior Member
Update,

finaly got hold of a sight glass.
using 3mm I/R transmitter and a 3mm I/R receiver, mounted in a black plastic holder a reading can be obtained in an 08M2.

transmitter and reciever mounte at 90 deg to each other, tip of led's at right angle to the sight glass.
Tried led's at 180 deg each, but didn't work.

readadc10 ,no water 502. with water 820.

transmitter has 180 ohm resistor
receiver has 20K ohm resistor,
adc measured at 20k - receiver connection.

simple led from c.4 for alarm, for now.

tof
 
Top