Steam Launch Water Gauge High Reading

[electrosteam]

Hi all,
A friend of mine has constructed the hull, engine and propane fired boiler for a 22 ft (about) steam launch with a 3" x 4" cylinder.
The first firing of the boiler in the workshop took place a couple of days ago with a closed system (no steam drawn, no water added).
There were the usual anomalies and leaks etc, but the test went very well except for the odd behavior of the water gauge.
The heat-up time to nominal pressure of 125 psi is incredibly fast.
Boiler is intended to be run closed-loop with engine exhaust condensed, oil separated, and shaft driven water pump used to feed boiler.

The gauge started cold at mid-gauge, and went out the top of the gauge as the pressure built up.
Gauge returned to middle on cool-down.

The boiler is a vertical drum with peripheral vertical water tubes inside a cylindrical housing, no fire tubes.
Drum 6" diameter, 18" length.
Tubes 12 in number, 1/2" diameter, about 12" vertical leg, 16" diameter over the legs.
Gauge mounted with 1/2" tubes from drum, top has a half loop about 4" high, reduced to 3/8" at the gauge.
Capacity reported as 8 litres.
Conventional circular propane burner.

Any suggestions as to what the problem is, and how it could be corrected, or at least mitigated.

John

[pault]

Hi John

Since you had a closed off boiler i.e. no steam drawn off, this does not seem that surprising, as when the water is heated it expands so the glass reading will increase, in rough terms from the numbers given the 8 litres will become about 8.3 litres when the boiler is in steam.
Does the gauge glass have a blow down, if so was it used and did the water return to it’s level in a normal time. Just to be on the safe side I would check all the waterways/valves are free from obstruction/fully open. Can you post a picture of the gauge assy, it is a little difficult to picture from your description.

Regards
Paul

[electrosteam]

Paul,
Spoke to the owner, I should be able to get a photo later in the week.
It seems hard to believe water expansion could do this, the copper expanding would make a correspondingly larger volume to be filled.
There are no isolating cocks for the water gauge, but there is a blow-down.
Exercising the blow-down produced an enormous amount of water/steam, but the gauge returned to the over-full reading.
John.

[Shawki Shlemon]

As I mentioned to you before , I suspect that a a blockage in the upper connection between the gauge and the boiler will create uneven pressure on the water and push it up in the glass . That is only a suspicion worth checking and if I am wrong then we must look elsewhere .

[houstonceng]

On a Maid of Kent filled to half water gauge, the level will rise to over the top nut if left in the sun for an hour. So no problem going high when fired.

[bhk]

Hi all,

Just for the benefit of those that have not been taken through blowing down a gauge glass and why:

Blowing down of the gauge glass should be undertaken when:

*you have reason to believe the gauge glass is not reading correctly

*As a matter of procedure on raising steam, before proceeding to the track or in this case water

The procedure is as follows:

Close steam cock
Close water cock
Open drain

(You should now see the water drain from the glass)

With the drain open, open the steam cock and check drain, you should see it blowing down, close the steam cock and repeat with water cock.

The point of this test is prove both steam and water cocks are free of blockages.

Now close the drain cock, open water cock. You now see the gauge glass fill completely, now open the steam cock and watch to see the water in the gauge glass drop to the correct level.

If the water disappears out of the glass at this point it indicates that the water level is below the range of your glass tube but is still above the lowest part of the boiler fitting for the water cock.

The importance of blowing down your gauge glass correctly and proving both steam and water cocks to be clear is quite simple. Is either cock is blocked it will cause the water level in the gauge glass to read high.

With regards to this particular topic. I believe it's quite simply the boiler is being overfilled before firing.

Cheers
Sean

[electrosteam]

Sean,
Thank you for that detailed procedure for blowing down the gauge, I will pass it on to the owner (and record it for myself).
It should be relatively easy to add isolating cocks.

As for the amount of fill when cold, I have suggested that the owner become a member of a steamlaunch forum and to post this same question there.
They will have the experience of boilers of this type.

I am going to research the physical data references for thermal expansion values and steam tables to see if I can make some sense of the change in water gauge level being only a static expansion phenomenon.
Is it common for small locomotive boilers to exhibit the effect reported by houstonceng, that of the gauge reading higher as the ambient increases ?

John.

[bhk]

Jul 21, 2014 23:04:51 GMT electrosteam said:

Sean,
Thank you for that detailed procedure for blowing down the gauge, I will pass it on to the owner (and record it for myself).
It should be relatively easy to add isolating cocks.

As for the amount of fill when cold, I have suggested that the owner become a member of a steamlaunch forum and to post this same question there.
They will have the experience of boilers of this type.

I am going to research the physical data references for thermal expansion values and steam tables to see if I can make some sense of the change in water gauge level being only a static expansion phenomenon.
Is it common for small locomotive boilers to exhibit the effect reported by houstonceng, that of the gauge reading higher as the ambient increases ?

John.

Hi John,

Very welcome, I would certainly add the isolating cocks as not only does it allow you to correctly blow down the gauge glass it also allows you to isolate the glass in event of breakage.

the effect reported by houstonceng, is common in both miniature and full size.

When operating boilers its normal practice to fill the water level to a quarter of the glass when cold, this is of course only if the gauge glass is fitted at the correct level at boiler construction.

Cheers
Sean

[pault]

Hi John
Yep the water will rise in the boiler as it heats up mainly due to the fact that the water has a much higher coefficient of expansion to copper and you are looking at linear expansion against volumetric expansion.
The main change which has effect on the copper of the boiler is linear expansion which in the case of copper the linear coefficient of expansion is 17, 10-6/K at 20 °C, effectively the rate of increase in length of the material of the boiler. The main change in the water in the boiler is volumetric expansion, the coefficient of volumetric expansion for water is 207, 10-6/K at 20 °C
Hope this makes sense
Regards
Paul

[ejparrott]

I only ever fill a boiler to 1/4 glass before steaming, maybe a 1/3, depends whether I'm paying attention. Left the Scot full the last fornight as it's been out so often, filled the glass to the top nut before pulling the fire, once cold it was down to half a glass, which then returns to a full glass once at pressure.

[jackrae]

Also be aware that the water level will rise when you start to consume steam. The drop in boiler pressure means the water boil rate increases and the increased bubble volume raises the water level. This is a classic control problem in automatic boiler water level control - what looks like a rise in water level actually necessitates an increase in water feed rate.

[electrosteam]

The owner is currently partly dis-mantling the boiler casing to re-work the pilot arrangement and gas regulator.
We have got independent advice in line with that provided here, to only partially fill the boiler when cold, and then be ready to adjust when steam is drawn.
I have confirmed that isolating cocks are fitted to both legs of the gauge.
It will be a couple of weeks to the next firing, and I will report the result of that exercise when it occurs.
Thanks to all for the advice and suggestions.
John

[steam4ian]

"Also be aware that the water level will rise when you start to consume steam. The drop in boiler pressure means the water boil rate increases and the increased bubble volume raises the water level. This is a classic control problem in automatic boiler water level control - what looks like a rise in water level actually necessitates an increase in water feed rate."

Read more: modeleng.proboards.com/thread/9508/steam-launch-water-gauge-reading#ixzz38KtTlvhP

which is why in commercial systems with small drums and high steam rates feed forward control based on steam mass flow is used to control feed water flow with drum level used as a reset.

On locos without much steam space water level rise on opening the regulator and drop when shutting off can be a trap. You shut off the regulator when going over a peak and the water level fall coincides with the water running forward in the boiler; an oops situation can quickly develop.

[electrosteam]

We have fired the boiler for the second time with several more problems arising.

1. The builder has got the gauge water leg isolating cock in the wrong place, to be fixed before the next test.

2. I mentioned before that the gauge has a substantial half loop in the steam leg - is this OK ?
Presumably it is easy to get good gauge operation by blowing it down if there is any doubt.

3. The draft makes pilot burner operation questionable.
- A safety cut-off valve is fitted with a temperature sensitive probe monitoring pilot operation.
- There is quite a natural draft up the long stack with a cold boiler, much stronger when the boiler warms up.
- Initial pilot ignition with a barbeque lighter requires a damper across the stack to eliminate the natural draft.
- When the main burners come on, the temperature probe activates the safety shut-down.
- It is difficult to see the pilot against a background of main burners, but it seems as though the pilot is extinguished by the draft.
- There is no burner jet directed at the temperature probe.
- The burner/pilot/valve arrangement is what the supplier provides asian restaurants for wok cooking.
- There is quite a delay from main burner valve opening to flame ignition, with a disconcerting "whoomp" that has dislodged some insulation.

Perhaps there needs to be a supplementary pilot jet maintaining heat in the temperature probe when the main burners are operating, and assisting in main burner ignition to reduce the "whoomp" effect.

We cannot think of any easy way of shielding the pilot from the draft that allows easy lighting and visual confirmation of flame presence, and can then ignite the main burners.

We can fit stack restrictions and/or a controlled damper, but how can we determine the correct draft with main burners operating ?

I am finding this boiler interesting and educational, and look forward for any comments or advice from you all,
John.

[jackrae]

With reference to point (2). If the "substantial half loop" could permit the possibility of condensate to collect and so prevent steam getting to the top of the sight glass then this is NOT ok.

[electrosteam]

Jack,
Good point, but the half loop is mounted upwards, so condensate should not be a problem.
The loop was included to introduce some flexibility between the boiler gauge legs and the rigid flat plate glass gauge.
For an operator used to tiny cylindrical gauge glasses on ride-on locomotives, this gauge glass is a truly magnificent specimen.

The builder and I are planning to visit the burner supplier to describe the situation and to investigate the practicality of changing to a larger pilot, or adding a supplementary pilot.

John

[ejparrott]

Its all interesting stuff. A few pictures would be a great help though.

[jackrae]

John,

If the loop rises from the gauge glass then drops back to the boiler tapping there is still a problem. Consider the situation where the boiler level is at the upper tapping. In the gauge glass there is water at the correct level and there is steam or air in the upper loop. When the boiler level rises above the upper tapping, the steam/air remains trapped in the loop so the gauge level rises slightly as the hydrostatic pressure compresses this "gas" but the indicated level is NOT that in the boiler. In effect the boiler can be over-filled and the gauge glass gives the impression that the level is high but not overly so.

In the ideal situation gauge tappings should connect horizontally to the vessel with neither rising or dropping loops in both upper and lower connections.

[electrosteam]

Jack,
another good point, would it be better if the loop was in the water leg ?

It is interesting that these steam launches seem to have a very loose consensus on build arrangements, and virtually no actual directives, standards or inspection authority.
A fellow loco man commented to me that the local miniature locomotive boiler code people did have discussions with the launch people some time ago, but nothing came of it because steam launch boilers were larger than the existing loco code range of application and they did not want to write a new code just for launches.
The launch boilers are also too small to get any interest from the Government authorities.

John

[jackrae]

Still a problem, if the boiler tapping is higher than the bottom of the gauge glass, because water trapped in a lower loop would maintain a level in the glass giving the impression that there was water in the boiler. Horizontal tappings are the only "safe" way