University experiments

[Keith S]

You could even bring the boiler to pressure, turn off the burner, re-fill the tank and relight it, (being careful to move it to a different spot in case butane "pooled" while you were filling it). Then you could be sure of minimum butane "wasted" getting up to temperature.

[Matt]

You could even bring the boiler to pressure, turn off the burner, re-fill the tank and relight it, (being careful to move it to a different spot in case butane "pooled" while you were filling it). Then you could be sure of minimum butane "wasted" getting up to temperature.

The problem with that Keith is that it will be very difficult to work out exactly how much gas has been used during the test.  if we keep the amount of gas used to one fill of the tank, we will be able to work out how much energy we have available to us from the gas (weigh the loco before and after gas filling to get the mass of the gas, then multiply by the calorific value for butane to get the energy stored), as so simplify the efficiency calculations.

[Keith S]

That's true.

[Matt]

Progress Update. Most of the work done over the last week or so has been ordering thermometers for the testing, starting on the interim report, and, making sure everything is up to date on the paperwork. Also, we've got a couple more parts for the testing.



This is a standard Roundhouse boiler plug that's been modified in the University workshop. A 6mm deep 1/8th BSPT hole has been drilled and tapped in the top section of the plug, with a 3 mm smooth hole down through the rest of the plug.



(bit of a poor pic this one, but it shows the smaller hole.)

The 6mm hole to take a brass gland that will prevent steam leaking during the test. The thermocouple inside the boiler will be inserted through this, and the modified plug, into the boiler.







will report back when more progress has been made.

[Matt]

Hello All. The first proper tests were carried out today. I'm afraid that I didn't take any pictures, will hopefully get some taken next time I do testing on Monday. I did two full runs, recording the temperature for the fire tube, the interior of the boiler, and the exterior of the boiler, once each minute. During the tests, the temperature of the fire tube was maintained between 840 and 880 degrees Celsius. At working pressure, the temperature of the interior of the boiler was 139 degrees Celsius, and the exterior temperature averaged at 130 degrees Celsius.

I'll report back when I've done a calculation for the boiler efficiency, although one odd thing I did notice was that the temperature of the fire tube dropped when the gas flow rate was increased, and increased when the flow rate was cut.

[Matt]

I've finished the efficiency calculations for the first couple of tests. For the first test, I got an efficiency of 78.2%, although this was will an incorrect reading for the amount of gas used during the test (the amount was higher than recorded, as there was still some gas left in the tank from a previous trial test done earlier in the morning). For the second test, with the gas recorded correctly, the efficiency was 58%.

I have to say I am surprised that the efficiency was this high, although we were indoors in a moderately warm room.

[11thHour]

Matt

one odd thing I did notice was that the temperature of the fire tube dropped when the gas flow rate was increased, and increased when the flow rate was cut.

Intuitively (from a non engineer) at a certain point of gas flow, the faster the gas flows through the fire tube, the less is combusted and ends up out the end of the flue unburnt or is burnt towards the end of the tube. This unburnt gas displaces burning gas as well as cooling the fire tube by absorbing some of the heat.
You would need to devise a procedure to test the hypothesis.

Tim

[Matt]

Matt

one odd thing I did notice was that the temperature of the fire tube dropped when the gas flow rate was increased, and increased when the flow rate was cut.

Intuitively (from a non engineer) at a certain point of gas flow, the faster the gas flows through the fire tube, the less is combusted and ends up out the end of the flue unburnt or is burnt towards the end of the tube. This unburnt gas displaces burning gas as well as cooling the fire tube by absorbing some of the heat.
You would need to devise a procedure to test the hypothesis.

Tim

Well I could use a gas composition sensor to determine the composition of the hot gases at various flow rates. might be something to put down in my report for future testing.

[MDLR]

There is also the point that the slower the gas flow, the better it burns. If you turn the gas up higher, it doesn't have enough time to burn before it comes out of the chimney.

[GTB]

I have to say I am surprised that the efficiency was this high, although we were indoors in a moderately warm room.

So was I, as that is close to the efficiency quoted for a early 20th century loco boiler.

I went back and checked my calcs. and this time used the correct column in the steam table.....

I now get 46% efficiency for both a Roundhouse and an Argyll boiler. Near enough to your figure, as It was done in the home workshop, not a lab. Seems high though for a boiler with a single short flue and a very high exhaust temp.

Not sure why the flue temp. drops with increased gas flow, but the burner venturi will be drawing in more external air which has to be heated and the flame stoichiometry may change, which will effect the flame temp.

However, I get the same boiler efficiency at both high and low firing rates. Doubling the gas flow boils the same amount of water, but in half the time.

I can open the smokebox door on my locos and the flame does get longer as the gas flow increases, but the difference is only a mm or so and the flame colour doesn't change. There isn't any combustion going on away from the burner if it has lit properly.

Regards,
Graeme

[Matt]

I have to say I am surprised that the efficiency was this high, although we were indoors in a moderately warm room.

So was I, as that is close to the efficiency quoted for a early 20th century loco boiler.

I went back and checked my calcs. and this time used the correct column in the steam table.....

I now get 46% efficiency for both a Roundhouse and an Argyll boiler. Near enough to your figure, as It was done in the home workshop, not a lab. Seems high though for a boiler with a single short flue and a very high exhaust temp.

Not sure why the flue temp. drops with increased gas flow, but the burner venturi will be drawing in more external air which has to be heated and the flame stoichiometry may change, which will effect the flame temp.

However, I get the same boiler efficiency at both high and low firing rates. Doubling the gas flow boils the same amount of water, but in half the time.

I can open the smokebox door on my locos and the flame does get longer as the gas flow increases, but the difference is only a mm or so and the flame colour doesn't change. There isn't any combustion going on away from the burner if it has lit properly.

Regards,
Graeme

Yeah, that idea about increased air flow was what I first thought when I got the readings. I might have a look into it if I've got some more time later on after I've got the testing I need to do done.

[Keith S]

Where are you measuring the temp in the flue tube? It may be the temperature does not decrease but the point at which it is hottest moves about. In the engines at my work we measure what is called "inter-turbine temperature" at a fixed point in the engine, (between the turbines) and the engine has different temperature limits for different power settings. This is not due to cooling or the metallurgical limits somehow changing, but because the "flame front" is at a different spot at different speeds. The maximum temperature for starting the engine is 1100 degrees for no more than two seconds, yet the maximum temperature for full power is 725 degrees for no more than two seconds. This is because when the engine is running at full tilt, if it's showing 725 at the probes, it's going to be 1100 degrees somewhere further down the engine. (Also there's blade stretch and all that but you get the idea)

Maybe when the gas is turned up the "flame front" in your flue moves further down and away from the probe.

Another thing may be that the mixture is changing. Perhaps the fuel/air ratio is more "chemically correct" at one setting, and the jet is not entraining air as efficiently at a higher setting, resulting in a richer mixture that runs cooler.

[Matt]

i'm not too sure how far up the fire tube the tip of the probe was during testing. when I adjusted the gas flow to reduce the temperature, it would settle at a new point and then started to rise again after a few minutes. presuably, if you theory is correct, when the gas flow increases, the point at which the highest temperature is moves down the tube due to the increase in gas flow, and as the pressure drops in the gas tank, it moves back up the tube.

[Matt]

Progress update. managed to get 3 tests in today. Would have managed to get a 4th in but i had to go to an interview at the squadron. below are a few pics I took.



A bit blurry this one, just shows the external thermo couple being put on. the fire tube thermocouple was inserted down through the chimney.



fire tube thermocouple fitted, and smoke box back on.



loco om the weighing scales. the loco was weighed before and after being filled with water, and then with gas, so the total mass of water/gas could be calculated.



loco on the testing rig, with all thermocouples fitted. the temperature values were recorded from the meter on the left.

Finally, a quick video showing one of the tests.

http://www.youtube.com/watch?v=j3Y8lv8Y ... e=youtu.be

will report back when more progress is made.

[Matt]

Finished off the tests today. last 3 tests were with the loco stationary, to see if there is a difference in efficiency when the loco is in motion and stationary. I got some pretty good endurance from the gas tank during the tests today (the last test lasted over 50 min until the gas finally ran out!). will post results when I've finished the analysis.

[Matt]

Completed the calculations for 4 more of the tests. for the two remaining running tests that worked properly, I got efficiencies of 60% and 50%. for two of the stationary tests, I got efficencies of 48% and 52%. will report back with the final set of results some I've got them calculated.

[Matt]

okay, i had to re-do the calculations for the efficiency values on the computer, as I've been using the wrong values for specific enthalpy at working pressure. apart from the faulty result, all my values from the 7 other tests fall between 52% and 66% (quite a bit of variation, but still pretty good results, giving the number of variables). My supervisor is happy with the results I've got, and I'm now ready to starting writing up the final report.

below are a few photos and another video from the last day of testing.



















http://www.youtube.com/watch?v=_CHWf4ZW0VI

[Keith S]

This whole thing reminds me of an article in "Eagle Annual" one year that featured "the design and building of a modern railway locomotive" . It had what I feel must have been an A3 or something going through the design and testing, in comic-book style. One panel had a picture of the loco all wired up with testing gear. Yours looks like that. Cool experiment.

[Matt]

This whole thing reminds me of an article in "Eagle Annual" one year that featured "the design and building of a modern railway locomotive" . It had what I feel must have been an A3 or something going through the design and testing, in comic-book style. One panel had a picture of the loco all wired up with testing gear. Yours looks like that. Cool experiment.

Thanks. Hopefully Roundhouse will be able to make some use of it when I send it off. I've got a coupe of ideas to improve the efficiency. On one of the model engineering forums, one of the guys who owns a model steam boat insulated his boiler with kaowool paper. I've done a bit of research, there's a company that produces it in thicknesses down to 0.5mm, and they have bases in Leeds and Sheffield. might be something to look into for future testing.

[Keith S]

It would be interesting for sure. I wonder if changing the efficiency of the boiler on a Lady Anne or Billy or similar would necessitate a re-design of the fuel tank. Roundhouse like their engines to be set up so the gas normally runs out before the water does. If a better boiler used less gas for a given volume of steam, I suppose you would save gas and if so they might have to make the tank smaller. Of course a smaller tank would look nicer. I'd someday like to make a gas tank for my engine that fits into the dummy side tank so I can have less clutter in the cab.