Australian steam locos

[Gumboots]

Then there's the C38 Pacific. Designed in the late 1930's, but didn't hit the rails until the mid 1940's due to some silly mongrels wanting to start a war over something. The first ones were running around in grey primer at the time since paint and painters were in short supply. Capable of 90 mph when running light, but usual service speeds were 70-80.

The first five of these (3801-3805) had streamlined casings, which were pretty basic but which a lot of people liked. My mother remembers these when they were new, and says people would often check out the locomotive to make sure it was one of the cool ones before boarding the train. If it was one of the not-so-cool ones, people would sometimes wait for the next train if they weren't in a hurry. Yes, really.

Opinion these days is divided, with some people thinking the bare version looks better. It does look pretty good IMO. This is another loco it'd be easy to do both versions of, since the majority of the model is the same in both cases.

C38_Pacific_streamlined_basics.jpg

C38_Pacific_bare_basics.jpg

---

I also roughed out the loco and tender for a Midland Railway 4-2-2 Spinner. These first came out in 1887 and were a sort of "super Stirling". The Spinners were the ultimate development of the 4-2-2 type, and capable of 90 mph. The 4-2-2's seem strange now, but in their day they were popular because they made a lot of sense. Mechanically simple, with low internal friction, not much to go wrong with them, and quite powerful enough for the trains of the time. They only become obsolete around 1900 due to train weights rapidly increasing then.

Midland_Spinner_basics.jpg

---

Last but not least, the Chapelon 160 A1. I really wish every bloke and his dog would not keep calling their locomotives "A1" but we're stuck with it. Again.

Anyway, this one is the complete opposite of a Spinner. It's a 2-12-0 built for heavy freight work on steep grades, and was a compound locomotive with six (yes, six) cylinders driving three different axles. It was one of Chapelon's experimental locomotives, and is a very interesting beast in technical terms. The idea of course was to get massive grunt within the confines of the relevant loading gauge, along with economical use of fuel and water. AFAICT it worked just fine.

Chapelon_160_A1_basics.jpg

[Gumboots]

Ok, an update on Garratts.

Although I would like to have an AD60 in the game, based on my latest testing (see comments here) I've concluded it's just not practical. It's simply too long to get acceptable behaviour over uneven track. That means I'm crossing it off my list of locos to model, so the basic test model is not going to be finished.

The good news is that the SAR Class GL Garratt model is just, barely, short enough to have acceptable behaviour over uneven track. It was also gruntier IRL than the AD60. So, instead of finishing the AD60 model I'll finish the SAR Class GL when I get time. If I want a Garratt for Australian scenarios, the GL can stand it for the AD60, just as it will be able to stand in for the other big freight Garratts used in various places.

Smaller Garratts are no problem at all (apart from the time it takes to make them).

[mukpile123]

That's disappointing. I was getting all hyped up about the NSW AD60 and was looking forward to downloading. All well, I guess the SAR GL will have to do...

Actually, now that I've thought about it, what about the SAR GMAM?

• Overall it is shaped very similarly like the AD60 (although the wheel configuration is shorter at 4-8-2+2-8-4), so you might be able to work off your AD60 model rather than building it up from scratch
• Also, since it is only about 94 feet in the length (about 14 feet shorter than the AD60, which appears to be similar in length to the GL), it shouldn't look too funky when it rides over uneven ground
• Surprisingly, even though the GMAM weighed only 2/3rds as much as the AD60, it could still generate just as much tractive effort (about 60,000 lbf at 75% pressure, which was almost twice as impressive as the GL)!
• Plus, since this little powerhouse was manufactured during the 1950s and was in regular service until the end of the 1980s, it would be a fitting candidate for that Garratt grunt you've been looking for to have pull those super-heavy railcars that appear in the later half of the 20th century
• And finally, one more piece of trivia: over 100 locomotives where manufactured and put into service, which made it the most numerous Garratt class in the world!

Anyways, regardless of whether or not you feel comfortable enough to slice up your abandoned AD60 model and stich it back together into the GMAM is totally up to you. I'm just making sure that you knew about the GMAM's existence and it's potential as replacement for the AD60.

Either way, I love viewing all the hard work you've put into crafting all of these new engine models in this thread and I can't wait to play with them once you've published them (especially the Garratts)!

Oh, and just for comparison, I'll just leave two pictures of the AD60 and the GMAM down below to compare features with each other (besides, guys like us can never see enough pictures of locomotives, now can we? )


New South Wales AD60 Class (4-8-4+4-8-4)


South African Railways Class GMAM (4-8-2+2-8-4)

[Gumboots]

I already knew about the GMAM, but it had less tractive effort than the GL. The GL was the gruntiest Garratt ever built, in terms of straight tractive effort rating. The GMAM and the AD60 were roughly equal for grunt, and the AD60 was faster, but neither came close to the GL for basic grunt. This is mostly because the GL had small drivers, which both limited its top speed and increased its tractive effort for a given cylinder size.

The AD60 test model is so basic that it's not really relevant to creating other models. The amount of detail it contains is so low that it's no drama to make others to the same level.

Anyway, if you want to play with the basic AD60 test model and see what you think of the behaviour over uneven track, go ahead. That's the best way to see how the length affects in-game visuals. I thought it was just too weird, but you may not.

One thing that had occurred to me was making it at a smaller scale to reduce the length. If it was done at 12" per RT3 unit, instead of the default 10" per RT3 unit, that would bring it down to under 100 units wheelbase. Problem is that then it would be out of scale compared to all other locos so I'd really have to re-do them all. And the cargo cars. Mind you, that's not all that difficult with the new Blender > .3dp script. However, it would be tedious to have to do the lot, and I'm finding it difficult to make enough time to do any of the models I want to do.

BTW, that shot of the AD60 doesn't show its normal livery. It was tarted up with that striping for a special occasion, but was usually plain unlined black.

[Gumboots]

I just realised I never put up the basic AD60 test model so that other people could try it. I'll package it up and upload it, so anyone who wants to can check out how it behaves.

Anyway, while thinking about this stuff again I had an idea. Part of the reason it needs such a long wheelbase is because the first wheels on the rear truck have to sit behind the ashpan. However, if we're going to get into making hypothetical but technically feasible Garratts to expand the steam roster in the 21st century, then there would be no need for an ashpan. Any 21st century steamer is not going to be burning coal. If it's liquid fueled instead there will be no ash, so no need for an ashpan. That means the wheelbase can be shortened while still keeping a 4-8-4+4-8-4 configuration.

That then led to more ideas. One obvious one is that if it's going to have pumped liquid fuel then it can be cab-forward, since you won't need a fireman sitting behind the firebox. The cab could go on the front truck, in front of the forward water tank. To keep good access to the cab it would then make sense to flip the motion, so that the front cylinders were towards the middle of the locomotive rather than at the ends. This would also shorten the steam pipes, which is a good thing for efficiency. The rear cylinders would be in the usual location, just to keep them away from the firebox.

AFAICT this should all be technically feasible, and should make for interesting units.

[Gumboots]

Ok, zipped version of the basic Garratt test model (shown in an earlier post) is attached to the bottom of this post.

It has just enough bits and pieces to get an idea of how the thing behaves. Skinning is ultra basic at the moment. Try it out if you want to see what it's like.

ETA: I'll rationalise and zip up the other two Garratts that are sort of halfway finished at the moment (SAR Class GL and SPR Class R2). Might as well.

I've also thrown some more detail at the D53 model. I know I'm supposed to be doing serious cargo cars, but this is just a bit of fun to get me enthused about RT3 stuff again.

D53_modelling.jpg

[Gumboots]

Latest box of tricks is the first locomotive ever in New South Wales. Goes by the name of "Number 1". No prizes for guessing why.

Blender model so far:

NSWGR_1.jpg

Actual locomotive:

NSWGR_1_irl.jpg

This one will come in looking smooth with a pretty normal poly count, due to being simple anyway.

[Gumboots]

Played with the SAR Class GL model a bit more too. I reckon I can bring that in with a vert+poly count under 3200. It's sitting on that at the moment, and needs a bit more detail added, but I've made no real attempt to save verts and polys yet so can easily knock some off without sacrificing looks.

Since the >4000 verts+polys of the experimental H10 double header was causing some hit to frame rate, I think if models can be kept down close to 3000 they should be ok. That would make them roughly halfway between the default locos and the H10 double header.

Anyway I'm getting quite a liking for the GL. I don't think I'd miss the AD60 much if I could have a GL instead.

SAR_Class_GL_front.jpg

SAR_Class_GL_rear.jpg

[Hawk]

For me, those screenshots are kind'a dark. Did you take them at night to highlight the lights?

[Gumboots]

Well yeah it's night time here, so I took them at night.

Anyway I was messing around with Blender's render settings to see if I could get something that more or less matched how RT3 will render stuff. If I can figure that out, it should be handy for skinning. Aint figured it out yet.

I ran those two shots through Photoshop and lightened them up a bit. Take another look.