Section Menu

Railroad Home

Why I got Back into it     Inspiration: Don Adams     Details     Layout     Structure Models     Roster       Steam Locomotives       Diesel Locomotives         Alco FA2         Alco RS3         Alco S1         Baldwin AS-16         Baldwin VO 660         Baldwin VO 1000         EMD BL2         EMD F7         EMD GP7           GP7 #20           GP7 #22         GE 44 Tonner       Rolling Stock       MOW       Vehicles     WMRY References     WMRY Prototype Info     CSX Prototype Photos     MARC Prototype Photos     Moore-Keppel Prototype Photos     Workshop, Tools, CNC, Casting

Prototype Locomotive Manufacturer	EMD
Model Manufacturer	Life Like
Model Run	Proto 2000
Scale	HO
Date Acquired	Around October 2002 at Star Hobby

The EMD GP7 is a sharp looking diesel, and the Proto 2000 model of it is a nice, smooth runner.

Here is a picture of the model before I did any work on it.  It's a pretty sharp model right out of the box, but still has room for some added details.

Progress Made Following the Jim Six NYC GP7 Online Clinic

In March 2005, I followed along on Jim Six's online NYC GP7 clinic on YahooGroups. Since Jim was working on an NYC GP7, some of the details I added, and some of the steps I followed differ a bit from the online clinic. The information below includes some of the standard steps as well as some of my additional steps. For more information abou the clinic and for a parts list, contact Jim Six for his clinic CDs, or join in on the fun at YahooGroups NYC-GP7

Thanks again to Jim Six for conducting this great clinic.

Prototype Photos and Detail

GP7 20 Unknown photographer and date

Prototype photos of WM GP7s in original paint are difficult to find. Some time in 2006, the WMRHS will put out a BMX issue on GP7s, which will definitely help add to the available material. In the mean time, I used the scattered photos I could find, including the few that show up in the various WM Diesel and In Color books.

I also recommend the WMRHS and Bob's Photos as potential sources of good GP7 photos.

Some differences from the Life Like model include:

• Shorter porch on brakeman side
• Additional louver (4 instead of the 3 on the body) on brakeman side due to the shorter porch
• Different railing configuration due to the shorter porch (seen in prototype photo above)
• Smaller (1200 gallon) fuel tank
• Five-chime horns
• Poling Pockets

Replacement Stacks

Detail Photo taken by Pete Brown on March 11, 2005

Detail Photo taken by Pete Brown on March 11, 2005

Stack replacement is a standard part of the GP7 clinic. I followed along and sanded the stacks until the bases were approximately one scale inch in thickness. Removal of the existing stacks was detailed in the clinic steps.

Likewise, replacement of the incorrect sand filler hatches is also a standard part of the clinic. As an aside on the second photo, I have since gone back and removed that ugly mold parting line.

For information on parts and techniques for these steps, please see the GP7 clinic CD, or visit the YahooGroups clinic. Please note that Yahoo does not archive photos, so the CD is the only way you'll get Jim's clinic photos if you do not participate in the clinic in real time.

Replacement Louvers and Radiator Grill Covers

Detail Photo taken by Pete Brown on March 11, 2005

In the various photos of WM and NYC GP7s, one thing that constantly stuck out to me was the depth and visibility of the radiator grills and louvers. Those were typically a dusty gray color behind the grill cover, and tended to be quite visible in all the 3/4 views. I was unable to find scale grill material that suited this purpose 100%, but I ended up settling on the Highliners F3B chicken wire grill set. The GP7 uses 45 degree angle grills, but the chicken wire is at a slightly more slanted angle. However, short of etching your own, the highliners grills are the most delicate and scale-sized grills of any I could find. The downside is that the Highliners grills can be hard to come across, and cost upwards of $11 each. The F3B grills are more available at the time of this writing, and provide more grill stock.

This step is not a standard part of the clinic, but was something I wanted to try with the GP7.

Update on Grills

GP9RM Fret Photo taken by Jeff Briggs on March 12, 2005

Jeff Briggs pointed me to a fret he developed and which is currently sold (usually as part of a kit) by Kaslo Shops Distributing. On March 13, I ordered three sets. While this is not a GP7 fret (it is for a GP9RM), the grill detail is the same. I am really looking forward to using this, as it will make a huge difference in the level of detail on the model. Once they arrive from Canada (may take a month), I'll try them out and post the updates here. The photos here are photos Jeff sent to the GP7 group on yahoo.

GP9RM Fret Photo taken by Jeff Briggs on March 12, 2005

You can see that I will need to carve off the rivet detail and other raised edging on the GP7 model. This actually works out quite well, since the fret pieces will be more even and square than those that I carved up myself. I'll wait until the parts are in before carving off that detail, though :-)

GP9RM Fret Scan taken by Pete Brown on April 1, 2005

I received the fret on March 31, 2005. That was pretty quick delivery considering it came from Canada, and I paid by personal check. You can see a scan of the fret on the left. On that scan, I noted the pieces that I am likely to use. The fan grills themselves are unlikely to be used by me, but are of high quality. Only two are correct for a GP7. Instead, if I replace the fans, I will probably use the set of four that come with a Highliners kit, or I will use the Intermountain fans. TBD.

The items on the fret that interest me most are the long hood grills. Anything I use above and beyond that is just gravy.

Removing the Cast-On Detail

Detail Photo taken by Pete Brown on March 11, 2005

Here is a shot of the original grills on the model. While they definitely look nice, they really lack the depth seen in the various prototype photos.

Note also that the grills are hatched at 45 degree angles. As mentioned above, I could not find suitable material of fine enough detail to reproduce this exact hatching. By the time you read this, it is possible, though unlikely, that such material has become available. One of the main qualities you are looking for is for extremely fine hatching, to the point where it is almost invisible as on the prototype. The closest I came to that was the chicken-wire screening from Highliners. Other available etched chicken wire was not nearly as fine.

Detail Photo taken by Pete Brown on March 11, 2005

I tried two different techniques for removing the grills. The first is the one you see in the photo to the left. The second was more successful, so I'll describe that here.

First, drill holes lengthwise across the center from one end to the other. You want the holes to be as close to touching as possible, so you can later connect them with the blade. Once drilled, I connect them all together using the blade, forming a break down the center of the grill.

Second, I score along all the inside edges, as close to the edge as I can get. Given that there is raised detail here, that is fairly easy to do with a sharp blade. The scoring should be as deep as you can get. The holes you drilled in the middle provide a place for material to move so you don't distort good detail with your blade. Be sure to get into the corners as tightly as you can, and also be sure to have the perimeter completely scored. This will prevent damage later.

Detail Photo taken by Pete Brown on March 11, 2005

Eventually, the scores will cut all or most of the way through. Once that is done, start pushing the middle inwards using the knife blade. This causes the edges to break on the scores. Use needle-nose pliers (from the inside) to remove the broken pieces of grill.

Note that you are applying a good amount of pressure here, so you want to support the body as much as possible, to prevent breakage. I found that a roll of masking tape was a snug fit, perfect for supporting the pieces.

After you break out the grills, you clean up with your knife and some small files. When cleaning out the corners, use a small triangular file so you don't damage one of the other sides when working. A square file, unless kept perfectly square (assuming it is perfectly square to begin with) will cut into one or more of the sides, ruining the detail.

Replacement Louvers

Detail Photo taken by Pete Brown on March 11, 2005

Prior to installing the louvers, I completely covered the holes from the inside with .010" styrene. This provides a thin but usable backing to the louvers. Note that if you use styrene thicker than this, you may have trouble getting the body back on the chassis.

Rather than hunt through expensive commercial louvers, I found that styrene clapboard siding does a very good job of representing partially open louvers. If you want fully open louvers, you'll likely need to tear into some Cannon parts.

Detail Photo taken by Pete Brown on March 11, 2005

The clapboard I used is Evergreen #4041 .040" thick, .040" spacing clapboard. I cut a long strip of it, all the same width. I then filed the edges while test-fitting until I got a snug press fit into the chassis (note, don't put it in all the way while test fitting, or you won't get it back out!)

Once I was happy with the fit, I put some styrene cement inside the hole and on the backing piece, double-checked the orientation of the clapboard louvers and pressed the pieces in place. It helps to hold your finger against the inside backing styrene to ensure you don't press the louver pieces in too far.

Once the louvers have been painted the appropriate gray and dusted, I will cut and install the outside grills and styrene strips.

Louvers, Stacks and Sand Fillers in Place

Detail Photo taken by Pete Brown on March 11, 2005

The photo to the left shows the new stacks, louvers and sand filler hatches in place.

At this point, I am debating installing new fan covers. The Highliners fan grills look very nice, but do not fit the Proto 2000 fan castings directly. Instead, I would need to chop off the fan castings and replace them with the fan castings from a Highliners kit. This is a bit of a project, and something I may leave for an undec GP7 conversion in the future. However, since I now know that the P2K fans are a bit over scale, and that the fan grills themselves are not as see-through as the prototype, this might end up being one of those things that will bug me until I do it right :-)

Of course, between the stacks, fans, new side grills, and the new engine doors (see down below) we're venturing into "Should have started with an undectorated model" territory.

Porch Modifications

Detail Photo taken by Pete Brown on March 13, 2005

The Western Maryland used GP7s that had the same short porch on both sides of the long hood. The Life Like model is of a GP7 that has a long porch on the fireman's side (orientation: short hood forward). In order to make this a most distinctive WM model, I decided that shortening the porch on the fireman's side would be a good idea. Doing so requires a few significant changes to the model including shortening the side of the porch, laying down new walkway treads, and fixing the two long hood doors that were previously partially obscured by the porch. Of these all, the last change will be the most difficult.

Shortening the porch side was very simple. I took measurements off of the engineer's side and cut the fireman's side to match. I slightly sanded the edges to round them to be consistent with those on the factory side.

Detail Photo taken by Pete Brown on March 13, 2005

Once you do this, you'll notice that there is no tread detail in the area the porch used to occupy. Obviously, this needs to be fixed. I scraped and sanded the existing treads flat up to the next tread seam (done after this photo was taken). This way I can put down a complete sheet of tread material without having an unslightly seam halfway through it.

Detail Photo taken by Pete Brown on March 13, 2005

When it came time to pick tread stock, I noticed that nothing I had was an exact match. The cannon tread stock is very different, and the S&S safety tread sheets I have are slightly larger than the cast-on detail. Nevertheless, I decided to go with the S&S Hobby EZ Safety Tread stock.

I cut the tread stock to length, but slightly over-width, and ensured the end that would be at the seam was square. I then measured and notched the outsid side to clear the porch side. Once I was satisfied with the fit, I spread a thin layer of CA on the model and pressed the stock in place, doing my best to ensure the seam was square. 

Detail Photo taken by Pete Brown on March 13, 2005

Once the CA (ACC) dried, I took my bladed and trimmed off the excess and lightly sanded it to blend in with the edges. Upon looking at it, I see that I could have filed a bit more of the original plastic down to even up the heights more, but that is barely noticeable. Once this is painted black and weathered like the rest of the model, it will take a close eye looking straight down to see what was done.

Note that the tread extends all the way in to the edge of the plastic. This doesn't cause any real noticeable fit problem with the shell, but I may scrape off a bit of that inside tread anyway to ensure the shell has a tight fit.

Better Tread

Detail Photo taken by Pete Brown on April 4, 2005

The more I looked at that out-of-scale tread, the less I liked it. Once again, Jeff Briggs chimed-in with a better alternative: Plano makes very thin etched brass safety tread stock - party number 208. I ordered some of that and installed it on 4/4/2005.

The etched tread cuts easily with a regular hobby blade. Don't expect that blade to remain sharp afterwards, though, as you are cutting metal. I basically followed a similar pattern to the previous cut, only I cut it to the exact width before installing it on the walkway.

The new tread looks so good, it makes the rest of the tread on this model look out of scale and overly chunky. Unfortunately, I draw the line on this model at replacing all the tread. filing all of those walkways flat isn't my idea of fun :-)

Replacement Grab Irons and Lift Rings

I replaced all of the grab irons and lift rings on the model with metal Detail Associates equivalents. For part numbers, see Jim Six's clinic CD.

New Grabs and Lift Rings The short hood Looks like a submarine from this angle. Photo taken by Pete Brown on March 19, 2005

To install the grabs, I taped a piece of evergreen .030" thick strip to the body, in between the holes. Then, for all but the top and roof grabs, I pushed the grabs into the existing holes and then used Cyanopoxy from the inside.

For the top and roof grabs, due to their inaccessibility from the inside, I clipped them short, dabbed them in Cyanopoxy, and installed them from the outside, still using the .030" spacer.

Still left to do is to drill holes for and install the NBW castings above each of the grabs.

Not only are these new grabs a lot more durable than the plastic Life Like grabs, but I personally think they look better as well. The durability is the most important part, though, as the factory grabs are too easily smushed or pushed out of the way. I've see lots of magazine closeups lately with smushed grab irons. Too bad, really.

Lift Rings Photo taken by Pete Brown on March 19, 2005

The lift rings were replaced for the same reason. On the long hood, the lift rings follow the pattern identified from several WM photographs. These tended to vary quite a bit, but I found a pattern that fit reasonably well. It matches no one photograph, unfortunately, as there were no roof shots anywhere. Please ignore the scratches on the hood for now. I did a bit of a botch-up on the paint job for the top, as I was not aware of how Model Flex behaved when brushed on. I am used to thicker acrylics. Since this is a learning locomotive, I plan to hide some of that with weathering if possible.

On the short hood, there was no room to double up the front lift ring on the Engineer side, as the grab iron interferes with it. I may still try to squeeze it in, however.

Vent and Door Replacement due to Porch Modifications

Because I shortened the porch, the cast-on doors and vents near that porch were no longer accurate. To see for yourself, take a look at the photo further up this page with the roll of blue tape propping up the body. You can see the incorrect doors clearly in that shot. 

To fix those, I cast a copy of the doors on the opposite side (which are 100% correct for the doors on this side, including the locations of the latches, according to my reference photos) and adhered them in place with Cyanopoxy.

Building the Mold Walls

Step 1: Mold Walls Photo taken by Pete Brown on March 19, 2005

The first step in making the mold was to form mold walls. Since I was molding a detail off of a larger piece, I could not use my usual methods (legos, containers etc.) to get a nice even and level mold. Instead, I use Alumilite clean clay (you can't use regular modeling clay, or your RTV may not set - a real mess) and built a dam around the edges of the detail. The clay cleaned off later with soap and water and a soft toothbrush.

You need to mash the clay into the detail enough to ensure that none of the RTV rubber leake under it and drains out of the area you are casting. I used a pallet knife to assist with pulling the clay back from essential details while smoothing it onto safe areas.

For a good casting, you'll want the mold walls to be a bit higher than you see in the photo here. I later added more clay on top. Another good alternative would be to set a styrene box into the clay and use that for the mold walls. You can make the box out of strips of .020 or thicker styrene sheet.

Whatever you do, just keep in mind that there is a delicate model under there. You want to avoid putting too much stress on the model sides. When pressing the clay in place, always support the model from the inside of that wall.

Pouring the RTV Mold

Step 2: Pour RTV Photo taken by Pete Brown on March 19, 2005

Once I had the clay dam in place, I mixed the Alumilte Quick Set RTV and poured it onto the details. Before the pour, however, I used an acid brush to stipple some of the rtv on to the details to ensure full coverage.

The Quick-Set RTV I use sets in four hours if mixed correctly. Also, while it requires no de-gassing (I do not have a vacuum chamber), I found that agitating it after pouring helps to both remove the bubbles and to fill in the details. To make this easier for future pours and larger molds, I plan to build a small agitation table using a cheapo random orbital sander as the source of the vibration. Of course, de-gassing is the way to go, but the agitation table would be useful even if the RTV was de-gassed.

Extracting and Cleaning the Mold

Step 3: Extract Mold Photo taken by Pete Brown on March 19, 2005

The finished mold was extracted around four and a half hours later. You can see the incredible capacity for detail inherent in silicone RTV rubber. Notice that it even picked up the the painted-on stripe!

Normally I would do my best to get a flat and level mold. This makes casting easier later. However, the unique molding technique on this model made that very difficult.

Do not clean the mold with any water or other liquid. Water and uncured resin don't play well together. Instead, only clean the mold with your pallet knife and your x-acto knife. You want to ensure there is no clay on the top, and there is no meniscus of rubber that would prevent the casting from sitting flat on a piece of glass on either side. As a matter of practice, I always use my knife to trim away all four edges of the casting side as well as those on the opposite side.

Mixing and Pouring the Alumilite Black Resin, and Cleaning the Casting

Step 4a: Casting With Flash Photo taken by Pete Brown on March 20, 2005

Step 4b: Casting Cleaned Photo taken by Pete Brown on March 20, 2005

In the first photo, you can see the squeeze out caused when I pressed the small round mirror on to the mold, after pouring the resin. In the second photo, you see the casting removed and, for the most part, cleaned up. The casting is only .004 of an inch thick, thinner than even our thinnest Evergreen styrene sheet.

I used a brand new x-acto blade to trim the flash from the sides, and to clean out around the hinges. I have to stress that only a brand new blade will work here, as the casting is simply so thin that using a slightly dull blade will risk damaging it.

Because I used a glass mirror as the backing for the casting, the back of the casting was perfectly smooth. To provide some tooth to ensure it adhered well to the shell, I lightly sanded the back using 1500 grit sandpaper. I suspect the end result was only .0035" thick or so.

Removing the Old Cast-On Doors

Step 5: Remove Old Doors Photo taken by Pete Brown on March 20, 2005

The next step was to remove the existing cast-on doors. I did this using a chisel blade, a sharp #10 blade, and some 400 grit sandpaper. I did my best to leave the Western Maryland lettering intact, as I was unsure about my ability to find a perfect match for any damaged parts of the "W".

The goal here was to get the door area chiseled down to the thickness of the body minus a tiny bit. I did not bother to smooth the area with anything beyond the 400 grit sandpaper.

The main reason the doors ended up sitting a bit proud is that I did not want to get too close to the right and damage the lettering or other doors. On an undecorated model, one can more easily carve in that area without worrying out ruining lettering.

Installing the New Doors

Step 6: Install Doors Photo taken by Pete Brown on March 20, 2005

Finally, I installed the replacement doors using Cyanopoxy spread in a thin layer with a pin. The installation of the door is pretty much a one-shot deal. If you refrain from sanding the back of the casting, you may have a bit more play, but it will not adhere as well to the body. Unfortunately I installed it a hair up too high, but hopefully it will not interfere with the etched grills I intend to later install.

The doors look like they sit a bit proud of the surface, and they do, but only by about .004" or so (measured with calipers). Once I weather the model and paint the stripe, however, this will not be very noticeable at all. In fact, even now, I can only notice this in the photos :-) On the next one I convert, I will ensure I sand the casting even thinner, and make double-sure I carve away a bit more of the shell to allow the doors to sit in closer.

You can also see a lot more detail in a large photo than you can with your naked eye. In the photo above, I noticed that I neglected to remove some flash from the bottom of the doors. However, I doubt that will be very noticeable once painted.

The other thing I noticed is the recessed area for the old steps. I didn't realize that until I looked at the photo. In future models, I will fill that before installing the new doors. On this model, I still plan to fill it, but it will be slightly more difficult with the doors in place.

Conclusion

Overall, I think this door casting idea worked out far better than I could have hoped. For one, it showed that the Alumilite black casting resin can hold up quite well in very thin castings. I was previously afraid I would need to cut the entire model side out in order to install new doors.

I would not hesitate to use this technique for similar situations on future models.

More photos coming as the model progresses!

If you're a model railroader, a WM fan, or just like this set of pages, please be sure to drop me a note in my guestbook.