P2 Locomotive Company and related matters

Rolling chassis by summer?

 

Which one?......... ......................

 

Sorry, I have been playing catch up on this thread. I nearly went to Doncaster, and I am gutted now that I changed my mind.
One question though, I have read that the foundation ring corners will be forged, not cast, to ensure that the crystal grain structure is optimised. Has that decision been changed?

Dave

 

Error there! Forged corners will be used on No. 2007, this is what David Elliott wrote in The Communication Cord #37, "Forged foundation ring corners. At the end of 2010, Tornado’s boiler suffered cracks to the foundation ring corners which necessitated removal of the boiler from the frames to renew them. The main cause was thought to be due to scale build up on the firebox causing it to become hotter than normal (scale is quite a good insulator). This caused greater than normal expansion and contraction in the firebox which in turn placed a greater cyclic stresses on the foundation ring corners resulting in fatigue cracking.

New corners were welded in and an improved water treatment regime has allowed the new corners to last longer, however there are some initial signs that cracks may be forming. When steel is cast the grain structure is fairly homogeneous, ie most of the crystal grains are of similar size and shape. When plate is rolled, it starts as a cast billet and is repeatedly run through rollers when red hot until it is the required thickness is achieved by which time it wider and a lot longer. This has the effect of stretching the grains in the steel in the direction it is rolled, which improves the fatigue strength in the along the axis of the grains. The foundation ring corners fitted to Tornado’s boiler are cut and
machined from boiler plate, however this results in the long grains in the steel being at right angles to the maximum stresses, which does not help in prolonging their life.

It is possible to control the direction of grain flow in steel by forging, and this is often used for manufacturing highly stressed components such as coupling and connecting rods. Forging involves hammering the hot steel to the required shape, and by careful control of the process the grain flow can be made to go in the optimum direction. After some searching, Brooks Forgings in the West Midlands have been contracted to produce tooling and make forged corners. As the tooling is a large part of the cost, we have
ordered twelve corners, four for Tornado’s boiler, four for No. 2007’s boiler and four spares. The new corners are being made with extra metal on their upper edges and ends so that when they are used as replacements, they can be welded to fresh metal on the firebox. On a new firebox they will be cut back to the original size."

Foxy

 

Thanks Foxy.

I was hoping my memory wasn't failing me - which at my age wouldn't be a surprise

Dave

 

This is interesting although I have some doubts. Thermal issues alone are not responsible for the cracking, as there will be considerable primary and secondary stresses in the foundation rings that are adding to the thermal issues. What was the result of the cyclic FEA done on the boiler at the design phase to identify high stress areas... You make my case for me by admitting the issue is still unresolved after a patch up.

Cast steel has a horrible grain structure with numerous flaws and very high internal crystalline dislocations and incursions. This is what makes cast iron so brittle. Its literally riddled with cracks.

Boiler plate or plate is homogeneous in all directions as specified by the standards it is rolled to. BS/EN/AISI and is rolled across all directions to create a homogeneous material. Directional plate steel does not exist for obvious reasons. Moreover, I have no idea how you would create it. Hot and cold rolling is a type of forging.

Although forging does help to align grain direction, its primary aim is to reduce grain dislocations and drive out inherent inter-molectular residual stresses. This is primarily what enhances material strength, not grain alignment.

"...however this results in the long grains in the steel being at right angles to the maximum stresses, which does not help in prolonging their life..." This is just not correct at all I'm afraid.

What you are getting mixed up are two separate topics. One being composite materials, where homogeneous and isotropic characteristics are defined by material properties and direction, and the other being the study of materials and grain boundary diffusion within those materials.

Upping the modulus of the rings is probably a solid enough approach, but without proper analysis its still just guess work. Moreover, there is a risk that the problem may end up elsewhere as those thermal and primary stresses still end up somewhere.

 

Surely it's slightly difficult to roll plate in all directions, a maximum of two I would have thought, unless you want a cube? By the description I would guess that Tornado's foundation ring is ultimately of channel form like an SR West Country, so to machine it out of plate will inherently cut across the direction of the grain structure.

I think you do cast iron a slight disservice. Yes, grey iron is poor in tension because the flakes of graphite leave sharp corners inside to act as stress raisers, without even starting to worry about solidification cracking. Spheroidal graphite iron is far less prone to the problem because the graphite form doesn't act as nearly so much of a stress raiser.

 

My understanding is that it is rolled at 90 degrees to itself, in x direction then y, and then finally in whatever direction the desired end geometry requires. Yes, I was a wee bit harsh on cast iron, and I do take your very well made point.

I'm not sure why some are getting the idea that steel is somehow like wood. I think it may be a terminology thing.

When we talk about grain structure, it is in reference to the Granular composition of steel, and that this is composed of grains of carbon, iron, manganese etc. It has grains, not a grain. Plural, not singular. It does not have grain running in an isotropic direction like wood does. Steel is homogenous, not isotropic.

To illustrate, steel looks like this.



Bear in mind, it looks like this in 3d, in all directions. That is just a 2d view.

So, when you cut it in any dimension, in any plane, you are cutting across this grain structure. There is no preferred direction in which to cut it, indeed that is close to the definition of what makes a homogenous material, homogenous. That it has the same material properties in all directions. Wood, being isotropic, does not. Again, steel is not like wood. At all.

What forging and rolling can do is align these more perfectly with each other, and drive out edge dislocations in the intermolecular structure, but you still end up with a homogenous material. That is, every boundary line between those grains has edge dislocations. It may be slightly stronger in one direction than the other, but it will be overall stronger in all directions as you have driven out intermolecular residual stresses. Cambridge have a good tutorial on this.

http://www.doitpoms.ac.uk/tlplib/atomic-scale-structure/defects.php

This is getting massively off topic. The point I made still stands. A thorough analysis of the stresses present in the foundation ring is the best way to resolve the issues with cracking. FEA would be the way to do this. However, seeing as the design is now virtually set, due to the commitment to have the two boilers largely the same, I appreciate that other solutions need to be proposed.

 

At risk of continuing the off-topic a little further, I think you've missed that forging and rolling also have the effect of changing the shape of the individual grains, therefore grains in plate will be longer in the plane of rolling than perpendicular to it (unless the material is heated above the recrystallisation temperature). Therefore, there are more grain boundaries per unit length across the plate than along it.

Since grain boundaries inhibit dislocation movement and crack propagation, the material does tend to be stronger and tougher when considering an applied load in the plane of rolling as opposed to one normal to the plane; plate does not have perfectly homogeneous properties. Achieving a fine grain structure by alloying or heat treatment has the same effect in all directions, up to a point.

Forging and pressing have the effect of forming this structure of elongated grains around the desired shape, such that a crack propagating from the surface has to cross more grain boundaries on its way into the component, thus slowing propagation.

 

How would Titanium perform on the foundation ring corners please?

 

Not well. Tornado and Prince of Wales has/have a fully welded boiler, so first you have the for all practical purposes impossible task of welding Ti to steel. Explosively bonded sandwiches are the usual preferred method if such a joint is required; you have a sheet of each metal (sometimes with an intermediate layer between) and drive them together extremely fast using explosives. They don't come apart. The relevant materials are then welded to each side of the sandwich (welding Ti is still not fun). I can't see how you could readily incorporate this type of joint into a boiler. Secondly the coefficient of thermal expansion is rather different for those two materials (steel being about 25% higher), so even if you could get a functional joint there'll be an immediate cause of additional stress when the thing heats up. It also has low thermal conductivity, so heat will tend to get concentrated inside the foundation ring; not good because the steel will be hotter than the bulk Ti and expanding more per unit temperature. The elastic modulus (i.e. how much it stretches when a load is applied) is also different for the two materials and sudden changes in stiffness usually cause stress concentrations. Something will start cracking, if not the Ti bit then probably the steel it's attached to. It's also horrifically expensive due to the cost of refining it from the (abundant) ore and processing the finished product.

Pick the right alloy and it would be a lovely material for connecting rods though...

 

Mmm, think of the reduction in hammer blow, reciprocating weight etc with all titanium rods and valve gear... That distinctive titanium colour might be quite cute too.

 

You are indeed correct, however we are not changing the material properties in any one direction substantially enough to allow us to drop the fact that it remains a homogenous material.

A good substitute material might be Inconel 625, widely used in oil and gas and more than capable of taking the heat and corrosion. It also has reasonable coeff's of expansion to allow it to live somewhat harmoniously with the surrounding steel. Best of all, it takes a weld, which is a key consideration here.

Ti connecting rods, now we're talking! A wee lick with some oxy/acc and you could have some serious aesthetics too! Edit, Jimc beat me to it!

But can you imagine the cost hahaha Nudging 6 figures a rod...????

 

Are you all barking up the wrong tree here? AFAIK a feature endemic to the German/European welded boiler is treatment of foundation ring corners as a short life consumable in a similar manner to how UK and Australian standards (and basic physics 8^) ) treat tubes and stays as short life consumables. The foundation ring corners in those boilers will always require reasonably frequent replacement because they are designed that way.

 

...moving on..... while most construction to date has been of a fairly traditional nature, Darlington Locomotive Works has now started to fabricate parts that are unique to the P2, particularly the complicated and rather beautiful egg-shaped smokebox, now fitted with a superbly machined smokebox door ring. You can catch the latest update on the P2 website here.

Foxy

 

Put it in for the Turner prize!

Dave

 

It would be a far more worthy winner than most of the crap I can recall winning it!

 

Necessary but rather unglamorous work continues at Darlington Locomotive Works. A whole host of components are being manufactured, machined and fitted, you can view some of this here. One of the challenges was producing half round beading for the cab.... a novel solution was found for that!


Meanwhile some pretty chunky frame stays are taking shape....

Foxy.

 

Great work. Coming along nicely.

 

All this is fine if maximum steam production is the only criterion, but it isn't. Most locos, and certainly these very large ones, did not work continuously at or even near their maximum power outputs, and there would be penalties had they done so. The LNER found this out early in WWII and ceased loading their big engines to 28 and more coaches.

It is essential that the engine makes sufficient steam to do its work. It is desirable that it has a reserve above this to meet unexpected contingencies. It is far from necessary that they could produce far more than even these contingencies would require, but the Stanier pacifics could do that with the existing system. Further raising outputit would achieve nothing except increased and more difficult maintenance. Increasing the steam production would need to go hand in hand with a means of using the excess; without that it becomes a wasted exercise.