Traffic Flow - New Britain Gas Light Co

One of the major industries I'm looking forward to building is the New Britain Gas Light Co. I experimented with modifying the Walthers kit a while ago. I've since picked up several more, since I'd like to see if I can fit in two tanks, which will also serve as view blocks. I've also been picking up supplies to scratch build the supporting structures. I'd prefer that it look different from the stock Walthers kit (it will be right by the aisle), and the structures for the two tanks were quite different from each other and I'd like to capture that.

But I've continued to try to dig up more information about the industry itself, and decided to pick up Jeff Wilson's Industries Along the Tracks 4.

Although the whole book is interesting and well written, the information on the Coal Gas industry was the most in-depth.

There were actually two similar products produced and stored in tanks like these for gas lighting, heating and cooking. These were predecessors to, and entirely replaced by, natural gas. Underground pipes were run to homes and businesses to deliver the gas, which also made it easy to convert a city to the less expensive natural gas in the '60s.

I was familiar with Coal Gas, which is a flammable gas captured from burning coal. The biggest by-product of this is coke, and I didn't know what New Britain Gas Light Co did with this. Did they sell it locally and truck it to dealers? Did it get shipped out in hoppers to be used elsewhere? I wasn't sure, and hadn't found any info.

The second process is known as Water Gas, which is also known as the Lowe method. In this process, steam is injected through the burning coal. The resulting gas is less volatile than traditional coal gas, but it is then processed in a carberettor where vaporized oil is infused into the gas. This was actually the more common method of manufacturing gas. Most importantly for my research is that the fuel used to produce the gas could also be coke instead of coal.

In the book Jeff mentions Brown's Directory of American Gas Companies. As always, Hathitrust has several editions. The entries are financial in nature. I found that the 1917 edition had some key information not in the 1918 edition:

It notes that New Britain Gas Light Co uses the Coal and Lowe processes. Even better, it notes the annual production of 100,000,000 cu ft of Coal Gas and 161,000,000 cu ft of Water Gas. The population served in 1917 was ~43,000 (and even notes the number of ranges, water heaters, and arc lamps). 

Jeff gives some rough estimates of how much coal and oil is required for a given amount of gas. Based on those, I've come up with about 400 hoppers and 40 tank cars of crude oil needed annually, or 1-2 hoppers and 0-1 tank cars daily. 

Sure enough, the Sanborn Maps show all of the relevant structures which correspond with the descriptions in Jeff's book.

1950 Sanborn Map

Note the corrections with physical paper glued over the original.

1954 Sanborn Map

The Oil House has been replaced by horizontal tanks.

I'll need a vertical oil storage tank as well. There is quite a large amount of oil storage capacity on the property, so I may lean towards 1-2 tank cars/day.

1909 Sanborn Map

This is the closest I have to the 1917 date. Looking at the 1909 book, the population is lower, but the total capacity of the tanks is the same, at 950,0000 cu ft. By 1950 a 2,000,000 cu ft tank has been added, so the capacity is 3x the 1909 numbers. So perhaps 2-4 50-ton hoppers and 1-2 8,000 gallon tank cars daily would be a better average.

If I really feel that it's necessary, 1950s era copies of Brown's exist, but just aren't accessible to view digitally. But I'm comfortable with this rough measure for now.

Improving Intermountain AAR 70-ton Flat Cars

I started this post in June of 2019...Yep. Another flat car project.

The Intermountain AAR 70-ton flat cars are really well done models of an important prototype (including a New Haven one), and I have several of them. The only issue I have with them is that the laser-cut wood decking is a bit too thick, and tends to delaminate and curl. One option is to thin out the deck and glue it better. But I also wanted to experiment with doing a board by board deck, which should be a simple enough project with this model.

Despite the decking curling at the ends, it takes a bit of work to remove. It's installed using double-sided tape. I peeled them off, then scraped off any tape that was left.

I didn't have any 2" x 6" styrene strips at the time, so I started with the wood strips I had handy. I used my always on-hand Aileen's Tacky Glue to glue them in place. 

This process was working OK, but I have found that I really prefer working with styrene. In addition, the laser cut deck has boards of several different widths. I tried to find some prototype diagrams or information that would verify this, but I haven't been able to do so. So I decided to go with Intermountain's research and replicate that.

I built a simple jig to help keep things aligned using a couple of pieces of Masonite glued to a Masonite base. I have some passenger cars to scratchbuild/kitbash and this was built to prepare for that (coming soon!). 

I found that I needed to do a few pieces then glue them, otherwise they wouldn't stay in position for long. As you can see, I built them alongside the Intermountain deck after cutting all of the pieces using the Chopper.

I also ended up building it in two halves (unintentionally, actually, but found that it let me file to fit around the bolster and figure out what width the final piece needed to be). I also installed stakes in the stake pockets on one side, to help in alignment and trimming the stake pocket slots.

It's a good fit. 

I did finish the wood deck as well, even though I may not use it. I also made the end pieces, but I haven't cut out the curved area for the drop brake wheel or the deck stake pockets yet. I also didn't make nail holes yet. Instead, I'm going to have a friend make some castings. I'll modify the castings to make a final master to make several, since I'll need a few. If they come out well, In the meantime, I have a few more that need to have their decks removed.

More Potential Flat Cars

I've exhausted my stash of Tichy flats, so I just need to finish the ones I've started. In the meantime, I have come across a few additional prototypes if you're inclined. Most of these cars on any road were retired quickly post WWII, but I'm sure a lot ended up in MOW service. For those modeling the depression era and earlier, they would be worth considering.

NYC Lot 208-F

• NYC&HR 32500-33499, reassigned to NYC 49700-49799 starting in 1916;
• CI&S 74557-74856, reassigned to NYC 496557-496856 starting in 1916;
• CCC&StL 50000-50249
• MCRR 33000-33499

Most of the NYC flat cars are similar to the USRA design, with fishbelly side sills and a straight center sill. But this series of 41' cars closely resemble the Tichy cars. They have 11 stake pockets. Here's a picture that looks like it's one of these cars. 

NYC Lot 209-F

• NYC&HR 31900-31999, reassigned to NYC 476900-476999 starting 1916. 

I don't have a picture, but the diagram is quite similar. The biggest difference is that it's listed as a wooden flat car. Some received steel underframes c1924-6. I believe the originally had truss rods, but that would be doable.

BAR 71700-72099

Ray Breyer recently posted this on the Pre-Depression Era modeling group on Facebook. Like the NYC car it has 11 stake pockets per side. I think this is the 71700-72099 series of cars. Best I can tell, this was the only class with a steel underframe. The deck was 40', over the end 41'.

The others had steel center sills and truss rods. For example this one, also from Ray:

I don't have the dimensions, but it would be just as easy to scratch build these side and end sills with a straight center sill and the addition of truss rods. The slight "fishbelly" side sill is interesting on this prototype.

I won't be digging up the many truss rod cars, as they would almost certainly be out of service by my era, but essentially this design appears to have largely been a transitional one. It starts with shorter flat cars with straight center sills and truss rods. The next evolution was steel underframes with deep fishbelly center sills (even at the high angle, the center sill is visible in the photo of the NYC one). They appear to have been largely supplanted by the USRA design, but a few more were built into the mid-20's and had less substantial center sills. 

More Flat Car Decks

I wanted to try doing a deck that was less distressed, as it seems like the railroads maintained them better in my era. I also hadn't tried the AK Interactive paints that I had received for Christmas - Old and Weathered Wood Vol 2.

Although it's not evident in the photos, the first deck I did still has a slight sheen in person. A matte clearcoat might resolve this, but when looking at it, part of it seems to be the smooth finish of the plastic itself. So I lightly sanded the entire deck with a sanding stick to start. This process also highlighted the very subtle differences in board heights on the Tichy model that weren't evident to the naked eye. Nice touch!

Since this is lightly distressed, I simply used the razor saw to work on the board ends a little bit. Even that probably isn't that necessary. If I decide the light sanding isn't enough, it will be very easy to go back and distress it some more and refinish those parts.

The base coat again is a thinned light gray, with as many coats as looks "right" to me.

For the wash, this is the same German Camouflage Black-Brown as last time, but it's the regular paint instead of the wash. In other words, it's the same color, but thicker. This time I used a flow aid instead of just water and you can see it flows into the crevices better and more consistently. It doesn't have the mottled look that the tap water had.

Here it is after it has mostly dried. You'll get a feel over time for how it will change while drying. By working in thin layers, if it's not a strong enough effect, I can just add another layer.

I then did a thin coat of a very light brownish gray thinned and applied as a filter to alter the color of the entire deck towards a sandy brown.

Then, using several colors, I highlighted individual boards. It was applied wet, but quite lightly and then basically dry brushed across the board and off. This is to provide some of the variation in colors. I probably should have started with the gray and then mixed in a little of another color to shift it slightly, instead of it being so dramatic.

I mentioned in the last post that the amount of paint used is very little. Here's a look at my palette for both cars. I haven't bothered to clean it yet. In the center are the four colors above, wet because that's what I was working on when I took this picture. You can see that I used a very, very small amount of paint on both decks.

The next step is to blend it in with a filter.

Then the application of a buff filter, fairly heavy to alter the color and blend it more.

After that dried, the variations still seemed to severe.

So I took three grey and brown colors and did a much heavier coat, blending the colors to provide some variation, but less evident. The coat is thin enough that the more severe variation just bleeds through.

That looks much better. Although there is a subtle effect from the earlier highlighting, next time I'll probably just blend them when applying them. Then another wash of a blend of the light grey and black-brown.

To tie it together, I applied a coat of the light sienna pigment while it was just a little damp. The idea being that the paint will hold the pigment and I won't have to apply a binder.

At that point I checked it out next to the other deck, and on the layout.

I like it, but decided to do one more wash, using the flow aid, to highlight the details a little more.

I could have done better with distressing the edge of the deck in this location. Something to be cognizant of when working on actual models. That's why I'm practicing!

So here they are side-by-side. One thing I should point out is that the lighting, the camera, and the device you are looking at a picture alters the colors. While I'm very happy with how these are photographing, it isn't quite the same color as what I see in person. The pictures above are under the Barrina lights on the desk.

This is taken on the layout. Lighting is just the daylight fluorescent ceiling fixtures that aren't directly over the layout. 

This is in staging, with the Barrina led lights at 6500k. The lights are fairly close to the layout. As you can see, they look far more gray. 

Here they are outside in late afternoon sunlight.

Anyway, I'm very happy with how these are coming out, so I need to finish up those flat cars so I can get to work on the decks. It also may look like this is a long process, but the reality is that acrylics dry pretty quickly. So it's a quick coat, which takes maybe a couple of minutes tops, then wait 15 minutes or so. In the meantime I was working on other things. 

Storing Paint

As a side note, I got a storage case for Christmas designed for Vallejo style bottles. There are removable racks designed to hold the bottles upside down. However, several of my paints have been much thicker than they should be, even the washes as I've noted in the last couple of posts. While I've had this issue with other paints (especially Tru-Color), it has never been a problem with Vallejo, at least not for me in the past.

I noticed that a number of them still sounded very liquid, but didn't come out that way. I've come to the conclusion that while the "traditional" method of storing Floquil/Pollyscale paints upside down works well for those, it does not work well for these. I think the problem isn't that they are getting too thick. Like other paints, it separates over time, with the heavier parts (the pigment and acrylic binder?) settles out. But with the dropper bottles, this settles into the throat and is very hard to mix out once it does. 

So while I love the dropper bottles, I had never stored them upside down until the last year or so. It doesn't seem to be a good idea, and it's something that I won't be doing any more.

Flat Car Decks

Although none of the flat cars are quite ready for this, a recent post/video by Jason Hill on his blog about how he distresses, paints and weathers flat car decks inspired some practice. Jason's techniques with the razor saw were faster than my more methodical approach, and while most of the specifics are similar to what I do it's always nice to get a refresher in some of the techniques.

Harold Minkwitz's Pacific Coast Airline Railway site also has some great info and examples. Unfortunately it appears Harold passed last year, and the site may be going away. He had an article in the July 2006 RMC. I would post a link to the Wayback Machine at the Internet Archive, but it appears he blocked access for his site, which is too bad.

It's been a long time since I've really painted anything like this. I did a quick coat on the crossing shanties, and did paint and weather the engine servicing pits. But painting is something I've been looking forward to getting back to for some time. I've painted thousands of fantasy miniatures along with some scenery back in the day. So I'm quite comfortable with brush painting with acrylics, but haven't done any in a long, long time. What I'll need to learn is how to use the airbrush, but that's for another day. This process is all about brush-painting with acrylics.

Here's a "before and after" look next to an unfinished Tichy model. I'm quite happy with how it turned out, here's an outline of the process I used.

Distressing the Deck

I'd already severely distressed the NH flat car deck, since it's representing a car near the end of its life. When I did the New Haven deck I used a razor saw, scalpel, and scriber to dig it up. The majority was with the scalpel. This is a shot in progress:

However, most of the photos I've seen from my era indicates that the decks didn't typically get that bad. Since I have extra Tichy decks, I used one to make what I would consider a "well-worn" deck that is still in regular service. So most of the work here was done with the razor saw. 

Painting and Weathering

My process is to work in thin layers to build the effect I'm looking for. This is an approach I'm very comfortable with from painting thousands of fantasy minis. When painting fantasy miniatures the basic approach is to paint from the inside out. That is, you start with any flesh and then "dress" them with paints. Other folks may have a better idea of how to get a similar finish by just painting it, but I work best in thin layers

The process at its most basic is a base coat, shadows and highlights. Reaper Miniatures has a line of paint that comes in groups of three colors: a base, a shadow, and a highlight. The classic method is to apply the base color first, then a wash of the shade to darken the areas that would be in shadow or collect dirt, etc., and then dry brushing the tint to add highlights. But for a more natural look, you'll probably have more than just these three layers and techniques.

For weathered wood, a white, grey, or light beige are all reasonable colors to start with. The deck itself is gray and could be used as the base color, but I wanted something a little lighter and less consistent.

I started with several thin coats of Vallejo Grey Primer. It doesn't need to be a primer, it just happened to be the light neutral grey that I had in reach. Applied much like painting with watercolors, I wet the model first, then applied a layer of thinned primer and let it dry. I did several layers until I had the coverage and color I was happy with.

I wasn't trying to get even coverage here, or I wouldn't have thinned it (or might have just airbrushed it). Instead, I'm trying to add variations in color and coverage over the plastic to provide the base.

I applied two different colors of a dark wash. This is to fill the spaces between the boards, along with the damage carved into them. In this case I used two because I felt the German Camoflauge Black Brown was too brown. Although this product was sold as a wash, it was very thick in this bottle, and I'm not sure the black mixed in sufficiently. No worries, though as a second wash with the Light Grey toned down the brown.

A wash, particularly with a light base and dark wash, immediately makes it "pop" and you can already see the texture much better. A very light gray or white base with a black wash is a common approach to simulate weathered wood and many modelers would stop at this point.

As that dried, you can see how some of the pigment settled out in a more mottled look, which I really like. This was a relatively heavy application, and there was enough liquid on the surface of the boards that I could tilt it different angles and let it flow and color some of the surface of the boards, and not just in the cracks.

I used straight tap water for this, which reacts differently than an actual thinner. I think the surface tension of the water tends to suck some of the pigment out of the cracks depending on how it dries. In addition, the pigment seems to separate sometimes into the mottles appearance. If you aren't careful, especially on smooth surfaces, you'll get an "edge" where the pigment collects as it dries.

Now for the highlight step. This is commonly done by dry brushing a lighter tint of the base color, along with other desired colors. Dry brushing is a great technique, but does have a specific look that isn't always the best approach, and is often quite noticeable in photos.

Another option would be blending. This is done using wetter and thinner (more transparent) layers of paint. each layer covering less area out of the shadows, and towards the highlights. You can use as many layers as you'd to get an even gradation that smoothly makes the transition from the darker areas to the highlights.

For this I did a bit of both. I felt that straight dry brushing would be too stark, so I started with that, and then went back with some isolated blending.

Another wash of a neutral color, helps blend everything together even more.

Some people would call this a filter or blending coat. What's the difference?

Where a wash is intended to flow primarily into the crevices and low spots to simulate collecting dirt and shadows, and emphasize the texture, a filter is to blend everything together to make it look more natural. The filter can be used to alter the color as well, but in this case I used a neutral gray because I wanted to smooth out the overall look, rather than adjust the color.

Having said that, best I can tell there isn't a lot of agreement exactly what a filter or wash is. I've always understood a filter to have a similar effect as a camera filter would, altering the entire photo or, in the case of painting, area that it's applied.

For many a filter is an oil or enamel based wash. AK Interactive's filters are enamel based, intended to be used over their acrylics. Citadel Colours (Games Workshop) used to have a line of inks. I still have some. They were similar to a wash, but the color was more intense. I haven't used AK Interactive's filters, but they may be like that.

I'll stick to the term wash for a thin coat intended to flow into the crevices, primarily to make them more evident, and a filter as a (usually) thin semi-transparent coat to alter/blend the colors underneath them. They may be a very similar consistency, it's really the intent or application that differs for me, so it lets you know what I'm trying to accomplish in that step. Anyway, moving on...

A few more shots of what it looks like at this stage. I like the way it came out better in photos than in person, but I'm quite happy in both cases.

My goal is for individual boards to look different, rather than a smear of color across several boards, yet for them to look appropriately similar, since they weather at a similar rate. I chose not to do any single or multiple replacement boards that are significantly newer than the rest of the deck.

You can continue this process of layering as much as you'd like. One thing to be careful of is to not make things too dark. Paint buildup isn't an issue since they are very thin layers. For this deck, I decided to stick with one layer of each technique, just to see where the fasted approach would land. In addition, it lets me see whether there are areas or effects that I feel are lacking. It can be difficult to tell if the initial layers are providing the desired effects until you get to a blending wash like this.

Sitting on the layout under the (current) layout lighting:

I've wanted to try this Tamiya weathering product that I've had for a while and the colors seemed perfect for this. In person the deck had a little bit of a gloss and I thought a light coat of "dust" would help. Although you could compare these to Pan Pastels, they really seem more like a make-up product, and come with a make-up applicator. A paint brush was insufficient to transfer the weathering medium, which means it was also tough to get a very light application. I treated it like dry brushing, but it still applied heavier (or not at all) than I intended.

I also did a final weathering with a light coat of Vallejo pigments. These will probably wear off over time, since I didn't seal it.

Overall I think Tamiya weathering blended things together perhaps more than I would have liked, but that is probably due to the nature of the medium. The Vallejo pigment would probably have been sufficient on its own. But overall I'm very happy with the results. It does simulate the effect of a coat of dust/dirt quite well, perhaps a little bit at the expense of the individual character of the individual boards.

Paints

As you can see in the pictures, I primarily used Vallejo paints here. I used Vallejo, Citadel (Games Workshop), and Reaper paints among others for painting minis, and am very comfortable with how they handle. Vallejo was my go-to until I purchased an entire set of Reaper paints. Once I switched back to building the railroad and needed to free up the space, I sold the Reaper paints. It's not that they won't work for this purpose, but they are produced in colors more appropriate for fantasy minis. 

I preferred the Reaper and Vallejo paints in part because of the quality, but also because of the dropper bottles. I'm usually working with very small amounts of paint, and it also allows easy measurement for mixing up other colors. A lot of people (like in Jason's video) use inexpensive craft paints for this sort of application, and those work just fine. While Vallejo is quite a bit more expensive, in the small quantities I use they last a long time, so they work for me.

Since I was never a Floquil/Polly Scale user, and had quickly switched from enamels to acrylics in junior high when painting minis, it was an easy decision to switch back to Vallejo for modeling. While they don't have pre-mixed railroad colors, because of their long-time support of military modeling they have great options for everything else. Joe Fugate's Guide to Acrylics has formulas to match Vallejo paints to many of the old Floquil colors. Micro-Marks' MicroLux brand of paints is also produced by Vallejo.

One thing I did like better about Reaper paints is that they had agitators (severed heads of minis) in the bottle. In Joe's guide to acrylics, he decants everything to dropper bottles and adds a couple of nuts, which actually work better than the severed heads since they don't fall into the throat of the bottle. I haven't added the nuts to my paints yet, but I did add them to the dropper bottles I used for brands that don't come packed in them. I do have lots of other brands, which will likely pop up in time.