The skill of the Engineman

Hi Tom
Thanks for your links to the Spirax videos...but video 3 clearly states, and shows, that the water level - as in the boiler and the gauge glass, will be increased by the formation of steam bubbles (- not any slight decrease of density).

 

Didn't the 9F hauled 'Windcutter' unfitted freights on the former Great Central route get up to around 50 mph?

 

Density of water at 100 degrees is just under 0.96 g/cm3 so if my sums are right increase in volume due to temp going to be a tad over 4% . So I submit that both steam bubbles (which will affect the effective water level eve though not present in the gauge) and expansion of temperature are responsible for the increased level in the gauge glass. A subtlety of no interest whatsoever to the practical engineman!

 

I think we might be at cross purposes, but the gauge glass is outside the boiler, free of the steam bubbles. So when you raise steam in the morning, the rise of water level in the gauge glass measured externally (but connected) to the boiler is simply caused by the fall in density of the water. At 200°C - approximately the temperature of boiling water at 200psi - the density is about 86% of the density at room temperature (which is a big difference, not slight), which causes the rise in water level at the gauge glass.

The foaming on the surface of boiling water takes place above that level, but you don't see it in the gauge glass. One reason why you need the steam takeoff in a boiler as high as possible above the nominal water level, which gets progressively harder to achieve as boilers get built out closer and closer to the loading gauge.

Tom

 

Free of the surface bubbles, for sure, but there are going to be bubbles of steam all over the firebox inner surface at least, which are going to raise the overall water level, and thus the level in the gauge.

 

The subject of boiler water and how it acts is one I'm particularly interested in. It is a fascinating subject. When you heat water in a vessel it expands as it becomes less dense. From room temperature at (say) 20°C up to boiling at 100°C is of the order of 4.2%, which isn't much. However, on our loco boiler we are looking at the gauge glass and that gauge glass is only looking at the top few inches of the water so the water in the glass will increase relatively significantly. This is best understood by considering a simple rectangular tank, thus:


It will be seen that the rise in water level in the gauge glass is quite significant. It is exactly the same with a boiler of more irregular shape.

As has been said by Tom, the water in the gauge glass is not boiling and essentially shows a steady level with no bubbles of steam. In the boiler, it is a different matter as the water is boiling. In a static loco boiler, the pressure head at any point on a horizontal plane is constant. Thus, the total amount of water above that plain must be constant. If you introduce a steam bubble the surface level above that bubble must be higher than where there is no bubble; as in the gauge glass. The following attempts to illustrate that:


To some extent, this works to the engineman's advantage when you mortgage your water as, when the loco is working, the level in the boiler is always higher than that shown in the gauge glass so there is always more water covering the crown than you think. However, the one thing that loco crew do not know is where the crown is. On UK locos there is generally no indication of that and it is a variable from type of loco to loco, known only to those in the drawing office. The footplateman ha thus to guess how low you can let the water fall in the glass and be safe. S160's are the only ones that I know of with any such indication. For those that think that, as long as there is water in the glass, you are OK, that is not always the case. For example, on the NYMR, on the 1 in 49 gradient, there is only a theoretical ½" of water cover at the front of the firebox when the water is at the bottom of the gauge glass. That's when those bubbles of steam causing a higher water level can become a life saver.

When a loco is sat, doing nothing with a fairly cool fire, the pressure may well be 220 psi but it will be in equilibrium and not making any steam. In this case, the water is not actually boiling and there will be no steam bubbles so the water in the gauge glass and in the boiler will be the same.

 

Thanks Steve
There are two points which I question though - although the volume of water increase is around 4.2%, the height increase in the gauge would only be around 1.4%. (Volume is the Cube of linear).
The boiler being hotter will increase in volume, but by a lesser amount ( ratio 1 to 4... 0 -100 deg C and 1 to 10.... 100 - 200deg C, from coefficients of expansion).
Also, even with a "cool" fire, the water will still be boiling to maintain the pressure so there will still be steam bubbles created in the water (albeit fewer than when steam is being used) which are being condensed in the cooler parts of the boiler.
Thus I still believe the fireman observing the gauge glass is seeing water level change far more due to steam bubble production than a change in density....either way he is interpreting the information with knowledge of the "road" and his drivers present and future needs, to an extent that we can only admire!

 

No ... Because the gauge is showing the total height gain of the whole column of water.

Thought experiment: Suppose you had a 1 metre long pole, fixed at the bottom, and you had only a narrow viewing window at the top with graduations in it, ten centimetres long with the top of the pole visible in the middle of that window.

Now imagine that the pole expands linearly along its length by 5%? Where is the top of the pole in your viewing window?

In essence that is what you are seeing. As it heats up, the whole mass of water expands, so you see the whole extent of that extension in the gauge glass, not just the expansion of the water in the glass itself. (A different thought experiment: imagine the gauge glass extended the whole height of the boiler form the foundation ring upwards. Now is it surprising that you see a big change in level in the glass when the water expands?)

(Bear in mind as well the "width" and length of the boiler, and the water in it, don't change -- so all the expansion goes in to increasing the height of the water, since that is the only unconstrained dimension.)

Tom

 

Thanks Tom
You are quite correct in your metre rod thought experiment...But if we consider a metre Cube of water expanding linearly by 5% ...its Volume has expanded 15% (1.05 x 1.05 x 1.05 = 1.1576). So, as I said, the 4.2% vol increase is only 1.4% linear change.
The boiler length does also change with temperature - as can be shown by the scuffed paint work on the firebox cladding sheets where they go into the cab frame....

 

The percentage change in boiler length is very small though in comparison with the length of the boiler - fractions of a percent increase.

To the level of accuracy we are taking of: to go from room temperature to 200C, the density decreases by 14%. If you imagine the boiler was a cuboid that changed negligible amount in length and width, all that 14% decrease in density has to come from an increase in the height of water, which you see in the glass. The complex shape of a boiler (basically a cylinder, but with uneven distribution of tubes; the firebox etc) it is more complex to model, but the basic phenomenon is a significant rise in water level as you raise pressure from zero to working pressure, caused by the drop in density. It's why boilers are filled up on disposal - if you left them at e.g. half glass, you'd have no visible water when you next came to light up.

Tom

 

Thanks Big Al and Tom....I think it all depends upon whether you are a glass half empty - or a glass half full!
But I would recommend the previously mentioned Spirax video Number 3 for a closer look....