Indicating Shelters

[Morello Cherry]

We've all seen the photos of engines with Indicating Shelters on the front:

 

 

 

However, I've never seen any photos of the inside of one. Has anyone ever seen the inside of one?

 

I'm guessing that in the bigger ones you could stand up. The Lord Nelson version looks pretty solid, the King looks like a garden shed from B&Q and the LB&SCR version looks like they knocked up this morning from some left over packing cases.  I guess that on the LB&SCR you just crouched down and held onto your hat.

 

This gives a good description but no image of the inside of shelter and I am struggling to visualise it from the description. https://www.railwaywondersoftheworld.com/testing-loco.html

 

Are there any accounts of working in an indicating shelter in any memoirs?

 

I'm assuming that the shelters were built as and when they were needed and then taken apart afterwards rather than permanent pieces of equipment that would be used on different engines over several years - in contrast to the dynamometer cars.

 

I would think working in an indicating shelter as being up there with any job in the pit at the end of the day as not a great deal of fun.

[kevinlms]

Perhaps the LBSCR version was more advanced and had a bench that the operator(s) could sit on and the others made them stand?

But yes, definitely not the most sought after job, especially as could be expected, the loco was working hard! The Dynometer Car would be a far more civilised place to work!

[Matt37268]

I very much doubt anyone who was tasked with being inside one of these would be concerned with taking a photo of the inside of one! 
I really would be happy to be proved wrong though. 😀

[Morello Cherry]

 

10 minutes ago, kevinlms said:

Perhaps the LBSCR version was more advanced and had a bench that the operator(s) could sit on and the others made them stand?

But yes, definitely not the most sought after job, especially as could be expected, the loco was working hard! The Dynometer Car would be a far more civilised place to work!

 

I did wonder if people in Brighton were very short :)

To the left (as you look at the photo) of the smokebox dart on the LBSCR photo there is a round black shape. Is that a bowler hat?

 

The line about a stout mackintosh and a shot of whisky being better for the crew than a cab springs to mind. (Was that Drummond?)

 

Something else I am not clear about, how would they have been fixed to the loco? I'm am guessing that their main purpose was largely to stop the people on the front from falling off. (Although thinking about locomotive engineers of the day it was probably to protect the recording equipment firstly and as an added bonus it stopped the crew falling off).

[Jeremy Cumberland]

I am trying to work out what the people inside the shelter actually do. Surely the measurements themselves are made by chart recorders, and I would have thought that the screens were there to protect the delicate instruments.

[LMS2968]

The measuring process wasn't continuous and triggered only when the people inside the shelters pushed the appropriate buttons.

[Mol_PMB]

I have definitely read about this in detail, I just need to try and recall where! I think it was in this book:

https://www.ebay.co.uk/itm/133471110631

Testing Times at Derby - A ' Privileged ' View of Steam by Alan Rimmer 2004 

 

 

The indicators created an x-y plot (piston position vs steam pressure) rather than a time-history plot, and therefore the paper needed changing every time a new reading was made (e.g. at a different cutoff or speed). There was also a need to change and clean the pens frequently, and ensure that a good plot had been achieved for each setting.

 

 

It was much easier when I did it, as we had the benefits of electronic sensors and data-logging and could watch the data coming in from the comfort of the buffet car behind!

 

[Mol_PMB]

I'll add a little more about how the traditional indicator itself worked.

I have used one of these myself, in a lab session during my engineering degree. That was on a single-cylinder gas engine rather than a steam engine, but the principles are the same.

The upper cylindrical thing is the recording drum. A piece of paper is wrapped round this (once) and held in place by the the pale-coloured clips on the side facing us.

The hook dangling down would be attached to the loco crosshead. As that (and the piston) move back and forth, they pull the cord wrapped around bottom of the recording drum causing it to rotate. Inside the drum is a powerful return spring to keep the string taut and return the drum to the starting point when the crosshead moves to front dead centre. It will be apparent that the circumference of the drum must be greater than the stroke of the piston!

 

A small-bore steam pipe is attached to a fitting on the end of the loco's cylinder and connected to the fitting at the bottom of the indicator machine. When there is pressure in the pipe, it pushes a small piston upwards, and therefore lifts the pen which is mounted on an arm attached to the piston. There is also a spring which acts to push the piston down. So as the steam pressure increases, it can compress the spring more and push the pen higher up. 

 

So when the loco is working, the indicator produces a plot with piston position on the x axis and steam pressure in the cylinder on the y axis, similar to that in my post above.

 

The operator in the shelter would have to move the pen to touch the paper when it was desired to take a reading, and then move it away again after a few cycles. 

 

The operator would then have to change the paper. There must have been some way to de-clutch the drive to the recording drum, as the loco didn't have to stop to change the paper on the drum.  

[Oldddudders]

10 hours ago, kevinlms said:

Perhaps the LBSCR version was more advanced and had a bench that the operator(s) could sit on and the others made them stand?

But yes, definitely not the most sought after job, especially as could be expected, the loco was working hard! The Dynometer Car would be a far more civilised place to work!

317 Gerald Loder spent much of its early career as a Portsmouth engine, so this was obviously a special trial on Forest Hill bank with a challenging load - B2s were not a success at their intended task of replacing Gladstones. Oddly, the LBSCR shelter seen here was in use on other locos, and never seemed to get any windows. Since I imagine that location was important for the chaps manning the kit to understand how hard the loco needed to work at any given point this meant standing up. A pic in Jeremy English's Atlantic book shows two bowler-hatted individuals posing behind the screen - on a static No 39 - which didn't even come up to chest height. A challenging experience even at 50 mph!

[Wickham Green too]

1 hour ago, Mol_PMB said:

...There must have been some way to de-clutch the drive to the recording drum, as the loco didn't have to stop to change the paper on the drum.  

I'd guess the thumbscrew to the right of the pulley would be tightened to grip the cord.

[Michael Hodgson]

1 hour ago, Wickham Green too said:

I'd guess the thumbscrew to the right of the pulley would be tightened to grip the cord.

It sounds as though thumb screws were needed to persuade the office junior to climb into the glorified packing case in front of the smoke box!

[Morello Cherry]

3 hours ago, Mol_PMB said:

I'll add a little more about how the traditional indicator itself worked.

I have used one of these myself, in a lab session during my engineering degree. That was on a single-cylinder gas engine rather than a steam engine, but the principles are the same.

The upper cylindrical thing is the recording drum. A piece of paper is wrapped round this (once) and held in place by the the pale-coloured clips on the side facing us.

The hook dangling down would be attached to the loco crosshead. As that (and the piston) move back and forth, they pull the cord wrapped around bottom of the recording drum causing it to rotate. Inside the drum is a powerful return spring to keep the string taut and return the drum to the starting point when the crosshead moves to front dead centre. It will be apparent that the circumference of the drum must be greater than the stroke of the piston!

 

A small-bore steam pipe is attached to a fitting on the end of the loco's cylinder and connected to the fitting at the bottom of the indicator machine. When there is pressure in the pipe, it pushes a small piston upwards, and therefore lifts the pen which is mounted on an arm attached to the piston. There is also a spring which acts to push the piston down. So as the steam pressure increases, it can compress the spring more and push the pen higher up. 

 

So when the loco is working, the indicator produces a plot with piston position on the x axis and steam pressure in the cylinder on the y axis, similar to that in my post above.

 

The operator in the shelter would have to move the pen to touch the paper when it was desired to take a reading, and then move it away again after a few cycles. 

 

The operator would then have to change the paper. There must have been some way to de-clutch the drive to the recording drum, as the loco didn't have to stop to change the paper on the drum.  

 

Thanks, I'm still trying to get a map of the layout in my mind.

 

Would it be one per cylinder?

 

Where would the drum and paper be - at the front or on the side?

 

Were there any other tests that could only be done from the indicating shelter?

 

I have just found this image which looks like the equipment was being set up:

 

 

Given how far down the side the shelter extends on the King I am assuming that the equipment is on the side.

 

 

I'm also trying to work out why the WR was still using indicating shelters in 1955 even though Rugby was up and running. But a much shorter set up down the side.

 

 

Still have no idea how it was fixed to the loco though.

Edited April 3 by Morello Cherry

[LMS2968]

Yes, there was one set of instruments per cylinder. The rollers at Rugby (and Swindon) measured performance at the wheel rim, not in the cylinders, so if you wanted to know what you were getting out of those, the indicator was the only way.

[pete_mcfarlane]

6 minutes ago, Morello Cherry said:

Still have no idea how it was fixed to the loco though.

Unbolt the lamp irons from the footplate and use their fixing holes? 

 

 

The Southern one on the Nelson has windows at two heights, which suggests that the lower ones are for people sitting down. 

 

Another variation not pictured here was something like a sentry box with a door on either side of the smokebox. Since there were two of them, does this mean that there was a set of indicator equipment for each cylinder? And that the King and the Nelson have bigger shelters as (being 4 cylinder locos) there is an engineer and a set of kit for monitoring each cylinder?

 

This LMS one is interesting as it appears to have a removeable section in front of the smokebox door.

https://www.warwickshirerailways.com/lms/lnwrrm775a.htm

 

Presumably the 'garden shed' ones with full width planking at the front weren't in place for long, and had to be removed to open the smokebox door for cleaning purposes?

[Morello Cherry]

10 minutes ago, LMS2968 said:

Yes, there was one set of instruments per cylinder. The rollers at Rugby (and Swindon) measured performance at the wheel rim, not in the cylinders, so if you wanted to know what you were getting out of those, the indicator was the only way.

 

I am just wondering what all the pipework and stuff attached the cylinders was in this test:

 

http://www.traintesting.com/images/Bullied_Merchant_Navy_Class_35022_'Holland-America-Line'_on_Dyno_Test.jpg

 

I'd love to know what the things that look like a load of rockets on the side of the smoke box are for.

 

 

[Michael Hodgson]

37 minutes ago, Morello Cherry said:

 

Still have no idea how it was fixed to the loco though.

 

... by driving a GWR County Class 4-6-0 through a wooden chicken shed.

 

[Mol_PMB]

A single indicator measures one end of one cylinder. To get a complete picture you would need to instrument both ends of all cylinders, which for a King would mean 8 indicators. 

Other parameters we measured in our tests were the steam chest pressure and temperature (which also gave a measure of superheat), and the smokebox vacuum before and after the spark arrestor mesh. Depending on the technology used, some of those parameters might also be monitored locally by the attendant in the shelter.

 

I suspect, on the Bulleid photo at Rugby, the devices mounted on the walkway handrails may be water manometers for measuring smokebox vacuum at several locations.

 

I really would recommend that anyone interested in this should read Alan Rimmer's book 'Testing Times at Derby'.

Edited April 3 by Mol_PMB

[Wickham Green too]

3 minutes ago, Mol_PMB said:

.. To get a complete picture you would need to instrument both ends of all cylinders, which for a King would mean 8 indicators. ...

...or could you assume that the diagram for each outside ( or inside ) cylinder would be as close as dammit to its opposite number ??!? 

[Mol_PMB]

We all know that Assume makes an Ass of U and Me.

Variations in valve setting and the streamlining of the steam passages can give quite different results from different cylinders or even the two ends of the same cylinder.

The famous example is Gresley's conjugated valve gear, where the middle cylinder could end up doing much more than its fair share of the work.

[Morello Cherry]

5 hours ago, Mol_PMB said:

It was much easier when I did it, as we had the benefits of electronic sensors and data-logging and could watch the data coming in from the comfort of the buffet car behind!

 

 

You mean to say that they wouldn't let you fit a garden shed to the front of a double fairlie? (It would be a very tiny shed - and I guess being a double fairlie you'd need two).

 

25 minutes ago, Mol_PMB said:

A single indicator measures one end of one cylinder. To get a complete picture you would need to instrument both ends of all cylinders, which for a King would mean 8 indicators. 

Other parameters we measured in our tests were the steam chest pressure and temperature (which also gave a measure of superheat), and the smokebox vacuum before and after the spark arrestor mesh. Depending on the technology used, some of those parameters might also be monitored locally by the attendant in the shelter.

 

I suspect, on the Bulleid photo at Rugby, the devices mounted on the walkway handrails may be water manometers for measuring smokebox vacuum at several locations.

 

I really would recommend that anyone interested in this should read Alan Rimmer's book 'Testing Times at Derby'.

 

I will check out the book.

 

Just out of interest how would you measure smokebox vacuum before and after the spark arrestor from the indicator shelter?

 

[LMS2968]

20 minutes ago, Wickham Green too said:

...or could you assume that the diagram for each outside ( or inside ) cylinder would be as close as dammit to its opposite number ??!? 

No. Uneven steam distribution wasn't uncommon. Even if the valves were evenly set for both ends of the cylinder, which they rarely were, especially for a tender engine expected to normally work chimney first, there were other factors such as the reduced piston area of the back of the cylinder caused by the presence of the piston rod and angularity due to the slight forward position of the piston when the connecting rods and crank were at a true ninety degrees.

[Mol_PMB]

10 minutes ago, Morello Cherry said:

 

You mean to say that they wouldn't let you fit a garden shed to the front of a double fairlie? (It would be a very tiny shed - and I guess being a double fairlie you'd need two).

 

 

I will check out the book.

 

Just out of interest how would you measure smokebox vacuum before and after the spark arrestor from the indicator shelter?

 

 

Owing to limitations of instrumentation, we did only measure one end of one cylinder, and duly discovered that it was doing less than 1/8 of the total work! Here's the pressure transducer:

Should have measured more...

Still, the main aim of our tests was draughting rather than valve setting.

 

The traditional way of measuring smokebox vacuum is with a water manometer (U-tube partly filled with water), one end connected to a pipe within the desired part of the smokebox, the other end open to atmosphere. The difference in height of the water in the two sides of the U is a measure of vacuum. This could be mounted on the side of the smokebox and data recorded manually in the indicator shelter. Alternatively, with a much longer pipe, it could be positioned in the loco cab or even in the dynamometer car.

 

We used electronic pressure transducers, which are attached to the ends of those angled pipes in the photo below. Having the long pipes helped to keep the transducers from getting too hot!

 

 

However, the traditional manometer doesn't tell the full story on smokebox vacuum because the inertia of the water tends to average out the reading to a near-constant value.

Here's a time-history of smokebox vacuum measured with an electronic transducer, showing how it fluctuates (black line):

Each of the big downward spikes represents a 'chuff', in this case about 12 chuffs per second, or 3 revolutions per second, about 17mph on the Melvin.

The chuffs aren't all quite equal, suggesting slightly imperfect valve setting (though it sounded fine, and I've known a LOT worse!)

Between chuffs the smokebox vacuum drops back towards zero (top of the graph).

 

[Wickham Green too]

4 hours ago, Mol_PMB said:

We all know that Assume makes an Ass of U and Me.

Variations in valve setting and the streamlining of the steam passages can give quite different results from different cylinders or even the two ends of the same cylinder.

The famous example is Gresley's conjugated valve gear, where the middle cylinder could end up doing much more than its fair share of the work.

4 hours ago, LMS2968 said:

No. Uneven steam distribution wasn't uncommon. Even if the valves were evenly set for both ends of the cylinder, which they rarely were, especially for a tender engine expected to normally work chimney first, there were other factors such as the reduced piston area of the back of the cylinder caused by the presence of the piston rod and angularity due to the slight forward position of the piston when the connecting rods and crank were at a true ninety degrees.

Yes, Yes, Yes ........... which is why I made no mention of different cylinder ends nor speculate about three-cylinder locos which are more or less guaranteed to give different results - and I was assuming ( yes, that word again ) that valve events would be carefully set before each test .................................. unless the test was specifically to suss out any pre-existing inequality, of course.

[Nick Holliday]

On 03/04/2024 at 11:29, Oldddudders said:

317 Gerald Loder spent much of its early career as a Portsmouth engine, so this was obviously a special trial on Forest Hill bank with a challenging load - B2s were not a success at their intended task of replacing Gladstones. Oddly, the LBSCR shelter seen here was in use on other locos, and never seemed to get any windows. Since I imagine that location was important for the chaps manning the kit to understand how hard the loco needed to work at any given point this meant standing up. A pic in Jeremy English's Atlantic book shows two bowler-hatted individuals posing behind the screen - on a static No 39 - which didn't even come up to chest height. A challenging experience even at 50 mph!

No. 317 was the fourth B2 to be rebuilt with a C3 boiler and other improvements to become B2X. As the first conversions were not quite as successful as was hoped, various tweaks would have been tried, and this run was probably testing the efficacy of the latest attempt. The loco still has its brass number plate, which helps to fix the date. Ex-Gerald Loder was actually allocated to Brighton shed in 1906, before being rebuilt in 1908, and was at St. Leonard’s in 1922.

[rodent279]

Do we know whether the shelters were painted, and if so, what colour? Would they match the loco colour, or would they just be a nondescript brown, grey or black?

And have any survived?