The Next Layer of Weathering

An important part of, well pretty much anything you learn, is figuring out how to recover from a mistake or when a problem arises. I wasn't completely happy with the last layer of weathering on this car, but I don't think it's easily removeable. If this was a model I cared about, how would I fix it?

I went back to oils. Since the tinted clear coat should have sealed the initial layer (which is pretty invisible now anyway), I should be able to do this without affecting it. Regardless of whether it's having the effect I wanted, I could at least test this theory...

I picked up a couple of other colors, and worked on a layer using primarily the raw umber as a sort of wash/filter to tint the grime, and also to add a bit of rain streaking from top to bottom.

More Weathering Experiments

 After the oils, I wanted to try to do a layer of soot and dust using the airbrush. Having never done this, I decided to start with tinting a clear coat with a little black. The idea being that not only will I add the layer of grunge that I want, but it will also seal the oils.

It took some work, and I did test it on a couple of spare models first. It was quite challenging to not get stripes. Since the coverage is light, when you do the next pass, the overlap is darker than the areas that don't overlap. It also basically obliterated the oil weathering I had done, probably being too heavy. It's still there, but not nearly as prominent as before.

I also think that this had a slightly different effect than I was hoping, although subtle. It almost looks too even, to my eye. Overall I think it will be fine, but needs something else to make it less uniform. But I either need a lot more practice, or to find a different approach to get that overall layer of soot and dust I'm looking for.

With the oils, I was ready to get started on the actual model immediately. I'm glad I didn't as I work through further processes. I think I need to weather another scrap model (or a few) with oils for practice, but also so I can compare them side-by-side as I try other approaches layer by layer.

This is a new paint, too. Dale and I stopped at Time Machine because I wanted to get some Vallejo Model Air black. They didn't have any, but they had a new rack with Mission Models paints, which neither of us had tried. Since we were going back to work on the layout, and Dale was going to show me some airbrushing tips, it seemed like a good opportunity to try them.

They were thicker than we expected, but airbrushed very well, with or without thinning. I think we both really liked how they handled. He did another coat on the X31a underframe and the color is good, not jet black, but much darker than the Vallejo color that I thought was a match for Grimy Black.

They were out of the gray primer, but I'm quite interested in trying it. It's a two-part mix and must be "activated" with up to 1/4 of their primer. I guess it's supposed to help it adhere better. They also recommend spraying at quite a low PSI - 10-15 PSI. So far I'm quite impressed.

Weathering Experiments

I'm making some modifications to one of the Rapido X31a box cars that recently arrived. Since I'm going through that process, I want to complete the car, weathering and all. To that end, I've grabbed a spare body (I removed the roof for another project) to attempt to develop a basic process.

Because the X31a cars were built 1934-6 and still in their delivery scheme, I'm using several of Jack Delano's excellent photos as inspiration. I believe these photos are c1942. I also found that there's a book of his railroad pictures, which I immediately added to my Christmas list.

The late '40s/early '50s is still the steam era, so the cars will still weather in a similar manner, so I think these are a good reference for any car that is still in the same paint scheme in my era.

While all of these are useful, I'm focusing on the Pennsy cars, like the X29 in the front.

There are two X31a box cars in this photo.

There's an X29 in the back row, left.

Key Observations

• The cars in this era don't show a lot of heavy areas of rust like modern cars do.
• There is a fairly consistent coating of dirt/soot on the dirty cars.
• There is a range of "dirty." The density of the coating of soot likely has to do with how many months/years a car has gone without a repaint.
• The roofs tend to be darker.
• Some of the roofs show paint failure.
• Some of the wood roofs and running boards show repairs.
• A few cars (very few) show "chalking" or streaking of the lettering, like the X29 in the third photo.
• Reefers tend to be cleaner. I know that PFE washed their cars, presumably others did as well.

Base Weathering with Oil Paints

Using this as a guideline, I started with painting along the rivets and other raised elements (including the door) with burnt umber oil paint. I really like working with this medium, and expect that it will form the base of most of my weathering. These photos show the process of working the oils.

You can apply it fairly heavily, since it's easy to remove.

After drying, I was seeing more streaks than I wanted.

I found that removing the excess on the panels with a paper towel was more consistent than a brush.

Here's a few photos on the layout after it had dried.

I tried a number of approaches, all of which worked pretty well. Wetting with (odorless) mineral spirits first, or not. Applying as a wash vs. a thicker application and taking away/thinning with a brush and mineral spirits, and similar variations.

I did find that I preferred upstrokes (from the bottom of the car) with fairly thick paint to apply the color, and down strokes with a damp brush to remove it. This was basically a brush that had been cleaned with mineral spirits and then dried with a paper towel.

I wanted there to be more dirt under the rivets than on top. I used a very fine brush to work on other fine details to ensure they received as much dirt as I wanted. Thinned paint, but not to the point of a wash, on a very fine brush worked well for along the panel lines.

A generous wash in large depressions, such as the door corrugations, did a nice job of providing a random application or dirt. Once dry I used a brush or paper towel to remove excess on the corrugations themselves, particularly along the top.

I wanted a very thin coating to dull the bright white of the lettering.

So far I'm quite happy with the approach, and I can see how it would be easy to do a number of cars at once time. 

More Flat Car Stuff

I've actually been working on other things besides flat cars, or at least working on working on other things such as getting prepped for installing the layout lighting. (It's gotten more complicated than anticipated. Which I should have anticipated).

But I have found a few other prototypes that would be easy to make from the Tichy car. 

NYO&W built fifteen 39' foot, 40-ton flat cars in 1911, which looks like the deeper center sill.

Here's an obscure one - LOP&G RR - the Live Oak, Perry & Gulf Railroad received 30-ton flat cars in 1914. It has the shallower center sill. 

The US Navy also received some 30-ton flat cars in 1914, and were 35'6" over the end sills. 

What I find interesting is that these were built in the same year, both by Magor, but the Navy cars had straight center sills of 15" steel channel (the side sills were 12" channel) and the LOP&G car had 10" channel side sills with a fishbelly underframe. 

The Seaboard F-5 46000-46931 class is another one, I don't have a photo available for that class, though.

There were a whole bunch of truss rod flat cars that could also be done that I haven't really explored as none would be in service in my era. 

These are also 50'-52' flat cars that resemble extended versions of the Tichy model. Jason Hill has some photos of a couple of kitbashes he's been working on with for SP 50'-ish flat cars, including the F-50-10,-11, -12 and -13 classes, and the F-50-14. He hasn't covered how he made the models yet, but his modeling is excellent so I'll be keeping an eye on that as another potential project.

More Flat Car Decks

Weathering flat car decks has turned out to be something very easy to do with 5-10 minutes of free time here or there. So I've continued to experiment with different mediums. The techniques are pretty much the same as what I've been doing, although the order often varies. In many cases this is because I get to a point, don't like what I see yet, so just keep adding more layers.

I did several Proto 2000 50-ton AAR Flat Cars:

I continue to make a point to take a picture when I complete each step with the materials I used so I can go back to them if I'd like. For something like this it's an imprecise process, so this isn't a "one drop of this, three of that" thing.

Here are the three P2k cars along with the original two experimental decks outside:

The three P2k decks, also outside:

Although I used a different number of steps, and different colors for these, the general approach was the same, and the final effect is quite similar for all of them. On one hand I'd like a bit more variation, but the reality is that only a few flat cars will be on the layout at a given time, and typically not side-by-side, so I don't think it will be that noticeable.

I'm also still on the fence about how much I should distress the decks. Based on photos from the era, it seems that the railroads maintained them better than they did in later years. I also have very, very few color photos from my era. The bottom line is, I'm still undecided on how much I think these decks should vary. They do look more uniform in the photos than in person.

Wood Decks

I've also weathered a laser cut wood deck. This is for a Chad Boas kit for an L&N flat car. This kit consists only of the main casting, the laser cut deck, and a piece of paper with few sentences on how to build it, in this case with Tichy stake pockets. I'm not aware of any decals for these, however. In the many levels of resin kits available, this is as basic as it gets. Having said that, they are very workable castings for prototypes you otherwise cannot get. So I have quite a few of Chad's castings and would recommend them.

I used the scalpel to rough things up and dig out some chunks, and started with washes of the Tamiya Panel Line Accent Color. The basic process is really one of staining rather than painting. I did also use the Vallejo Black Wash (75.518). I found that sandwiching the deck between two kit boxes while wet was sufficient to keep it flat.

Here's the deck in progress between the two original tests I did to compare:

The main difference is that when it looked like things were a bit dark, I sanded the deck using variable pressure so it wasn't uniform, but overall quite lightly. I repeated this process, staining, sanding, distressing, until it's gotten to this point:

The staining/sanding approach worked out a lot of the natural brown tone of the wood, which I like. The look is subtly different than the other approach. Because the wash works more as a stain, the grain (from the scalpel) is more visible. It's also easier to distress than the plastic.

The only real "challenge" with working with wood is that it does have a variable reaction to the stain. For example, where I had dug into the deck, the inner (or lower? Might by a plywood) layer didn't accept the stain much at all. As the wash dried, I added more to it several times to overcome the reddish-brown that was showing through in those large gashes.

There are also often sections or grain that is harder than the surrounding wood and doesn't accept the stain well. The imperfection spans several "boards" because the deck is a single piece, and partially cut to look like separate boards. You can see two diagonal lines across several of the boards. I was able to address most of the places this was evident with distressing, sanding, etc. But not entirely. Imperfections are good, because that's normal, but these look unnatural the way to cut across several boards, especially as two parallel lines. In the photos above you can pick up a "U" shaped imperfection as well.

The picture is a little blurry because I had to take it at an angle to accentuate the effect. I think it is more pronounced in person because it's not just that they didn't accept the stain as well, but they reflect the light differently. Picky perhaps, but it's noticeable to me and I'll try to minimize the issue.

I can't say that I really prefer working with one over the other as I've been quite happy with the finished effect in either case. They just require a slightly different approach.

Weathering Experiments - Oils

I stopped by the local Blick's for some supplies and picked up a tube of Winton Burnt Umber oil paint since I've never used oil paints before. Wow, what a pleasure to work with. 

Because they are so slow in drying, you can continue to make adjustments for hours (literally). Too heavy? A bit of mineral spirits and you can brush it off. It lets you work over a wider area to ensure that you like the way it looks as a whole. 

The pictures themselves only tell a very small part of the story - that I like the final results I can get.

I painted over this entire side with un-thinned paint, then wiped it off with a paper towel to fill in the lines and also leave a slight layer of "dirt" on the entire side. You can see the un-weathered section to the left for comparison.

I also used a slightly damp brush from top to bottom to spread the dirt.

I did a lot of experimenting along the side sill. Mostly to get a feel for working with the medium. While the finished result isn't all that different from this post, or this one, it was the process that was different. Because I could rework it I had a lot more control over where the paint would go. I had started very heavy, then brought it way down. 

This is with a (very) light dusting of Vallejo pigments. So the pictures show that I can get similar results with these as I have with acrylics (no real surprise there, as it's still just using paints).

These aren't intended to be finished, I just wanted to see how working with oils really was (for some reason it seemed intimidating to me or for "advanced" modelers only). The reality is, its the most forgiving approach I've used to date. It won't be the only technique I use, but I can see myself using it on most weathering jobs.

Especially where I'll be adding details using only paint. It's easy to use an acrylic or enamel filter to dirty up the side, or a wash to add the dirt around rivets or in the crevices, etc. I did like the way this flowed in that process, and the ease of removing it when not happy with the results. But for details like rust spots on a car side for example, where you aren't using a physical aspect of the model as a guide for the paint, using oils lets you experiment without any concern because if it doesn't come out how you like it, you can remove it entirely - hours later.

When weathering the flat car decks and the RS-2, the process was largely "random." Using washes, filters, drybrushing and some blending. But the finished weathering is directed, but imprecise, taking advantage of the way the paint flows around the details, things like the way water as a thinner causes the paint to dry in different ways than a flow improver or solvent, etc.

But there have been times, like the RS-2, where that approach wasn't entirely satisfactory to me. Where I'd like to be able to precisely paint what I'd like. And also remove what I don't like. Although that can be done with any type of paint, the removal process is much harder with acrylics because they dry very quickly.

Since I have an extra RS-2 cab, I did some more experimentation. Note that the paint I'm using on my palette is the same that I started with yesterday. It's still workable.

Here's a heavy application. Something that would be tricky with acrylics or even enamels.

Using mineral spirits to spread and thin it more into a filter.

Another view. Very streaky at this point.

After more working, particularly with a dry brush that is removing and blending the color.

I also used a paper towel to stipple the color while removing it.

I wanted more around the louvres and doors, so adding more.

Then using the dry brush to shape, remove, and blend.

This looks much closer to what I was trying to achieve on the completed models. The photos in my era show the RS-2s are quite clean, so I just wanted some subtle emphasis around the doors and rivets, and where dirt might collect. Definitely an approach I will be using for weathering locomotives in the future. 

I did the other side with Tamiya Panel Line Accent Color (Dark Gray).

The effect was good, especially between the doors and on the louvers. It was much harder to control, otherwise, since it's a wash. A mix of techniques and mediums seems to be best, which isn't a surprise.

For reference, here's what my completed RS-2 looks like - and the effect is a bit heavier in person than in photos:

I certainly wouldn't consider it "bad" weathering. Especially for the first locomotive that I've ever weathered. But I think I definitely could have done better if I had used oils at the time. I also hadn't discovered how differently the washes/filters work with tap water (which is what I used) and other thinning options. So it's always a learning process, and I highly recommend learning new techniques and using new materials, just so you have more options to find the best approach for a specific project.