Home | Store | About | Progress | TNP | Videos | Visit | Links Main Menu TMRC Store About TMRC Progress About the TNP Videos Visit the Clubroom Links

Trolleys at TMRC

Trolley enthusiasts have built streetcar systems at TMRC since 1954. The current layout continues this tradition, although with the simplest track plan it has ever had. Trolleys are powered by electricity provided by an overhead wire in the real world, and TMRC's trolley systems have always been built to operate in the same way.

We feel that running streetcars at TMRC is particularly appropriate in a city that still runs streetcars (well, Light Rail Vehicles) on the MBTA's Green Line, and will continue to run real streetcars on the Red Line's Ashmont-Mattapan "High Speed" Line.

See a photographic history of trolley service at TMRC.

MITCo Today at TMRC

MITCo is name of the trolley system at TMRC. MITCo is the abbreviation for Messachusetts Interurban Traction Company. (TMRC has a tradition of silly or punny names for things.) MITCo service has always served the Gifford City TNP passenger station, providing local service for passengers. The streetcars run in the middle of the street. You can see all of MITCo at our videos page.

MITCo Track Plan

The current MITCo track plan is the simplest one MITCo has ever had. It is designed to look like a double track mainline, with each end going off the edge of the layout. In fact, each end has a turning loop short of where it goes off the end of the layout, such that the operational track is what is known as a dogbone track. A dogbone has most of the track parallel along the length, with a loop formed at each end.

All the spurs off of this loop are only scenic, as they all go off the edge of the layout. They are not fully operational.

One loop is the downtown loop, which goes around a block of tall buildings. This loop is the only part of MITCo that was saved from the old Building 20 layout in 1997. The tall block of buildings are all the same height, in order to hide the fact that the block is hollow, and there is an access hole in the middle. This access hole is quite necessary in order to be able to reach the track and wire for maintenance.

The loop connects to double track through the main street of Gifford City, facing the window wall into the hall. It passes along the very tall wall of buildings, and past the Gifford City passenger station. From there, it goes around a broad curve overlooking the depressed tracks of the passenger station, past the Post Office.

On the inside of this curve is the four-track carbarn, which provides storage and service facilities for the streetcards. The carbarn was built by Alex Bardow in the late 1970's, and was moved from the Building 20 layout. The trackwork connecting it to the street tracks is more than a bit of a "show off" project. The angle of the switch ladder was chosen to cause overlaps of the frogs and mates of the switches. This trackwork is so arranged that cars can pull in off one track, and pull out into the other track.

The streetcar mainline then crosses a very temporary bridge over the Whattahack River. This may be replaced by a vertical lift bridge, or perhaps by a fixed span bridge.

From here, the nominal streetcar mainline passes Pettingill Circle, and proceeds off the edge of the layout. In reality, the track only takes the loop around Pettingill Circle, a narrow loop of street through an old-time business and residential district.

MITCo Design Ideas

The biggest change from the prior MITCo systems is that the main line is double-track from one end to the other. This was done since it really allows for a better looking urban street. It doesn't really make the street any wider, since you can model it with four lanes, two travel (shared with streetcars), and two parking. With a track down the middle of the stret, you need a travel lane on either side of the track, plus a parking lane on each side, so you really need five lanes curb to curb. The only downside of the double-track scheme is that you can't leave any automobiles on the travel lanes, since they would be in the way. However, the large amount of non-operational track in this layout allows leaving automobiles on the track in those areas.

MITCo Construction Details


All of MITCo's trackwork is in the street, using track that simulates prototype girder rail. Streetcar tracks in paving are normally laid using girder rail. This is a rail that has the head offset somewhat to the side, with a lip that comes up under one side to form a groove for the wheel flange, and a lip (like a guard rail) to form the far side of the groove. This lip is very important, as it prevents the paving (especially cobblestones and paving bricks) from creeping into the path of the wheel flange. Older wheels had rather brittle flanges, and they could be chipped by paving stones, which in turn caused a safety hazard.

The current MITCo track is built using manufactured girder rail. TMRC's supply was bought from Richard Orr, who designed the rail, and funded having it rolled by RailCraft (now Micro-Engineering). This rail is still available, from Custom Traxx.

The Orr girder rail provides a flangeway that is fully conformant with the NMRA standards. The head of the rail is the same size as current code 100 rail, and the rail is the same height (100/1000 of an inch) as code 100 rail.

The trick with the Orr girder rail is that the cross-section is extremely asymmetric. If you try and curve it sideways by hand, it will also curve vertically. It also loves to kink when bent. Orr makes a special jig with rollers shaped to match the side profile of the rail, which allow smoothly curving the rail with almost no vertical curve.

The current track is laid on top of 1/4" by 1-1/4" pine lattice stock. This is nailed and glued to the 3/4" plywood roadbed. No ties are used, the rail is spiked directly to the lattice stock.

Note that the spacing of the two tracks is quite close. They are closer than the NMRA standards, at 11-1/2 scale feet center-to-center. Note that the spacing increases dramatically around curves, so that streetcars will not hit each other there. The spacing was tested with a model of a Washington, Baltimore, and Annapolis articulated interuban car, which is quite large. Hopefully they will also clear a Pacific Electric "Blimp", but we havent tested this. (Not that we'd move the track now, but we may yet have to move a trolley pole or two.)

All of the switches are "single point" which is traditional for trolley street trackwork. In a single point switch, there is only a moving point in one rail. On the other side, there is a "mate", which just has grooves in both directions. The single point guides the car in the correct direction by pushing against the wheel flange. Normally, the point is on the inside of the curve, and the mate on the outside.

The switches are built from a combination of Orr girder rail, and from our earlier technology of using a code 100 running rail with a code 70 soldered sideways into it to make the flangeway. This happens to work very well for making the closure rails, point, and mate portions of the switch.


The current paving around the track is all patching plaster, colored with powdered carbon black. Curbs are formed by more lattice stock, which gives the street a 1/10" crown. Wiring staples (from a T-25 staple gun) into the plywood provide an anchor to secure the plaster to the layout. The plaster is mixed to a moderate consistency, and laid from curb to curb, up to the rail heads. The plaster is worked as little as possible while wet, since we don't want the work hardening of a classic wall plaster job. When the plaster is set just hard enough, paint scrapers are used to do a final smoothing, and to ensure that the plaster is all a few thousandths of an inch below the top surfaces of the rail.

When the plaster work is done with final shaping, there is a very important final step. The plaster is sealed with shellac diluted with alcohol. This is abosolutely essential, since otherwise the plaster continues to release a fine dust, which is very abrasive and gets into the gears of the streetcars. So long as the shellac is properly diluted, it doesn't make the streets shiny, it just soaks into the plaster, makes it hard, and makes it a good bit darker. Sometimes it makes the color a bit more uneven, but real streets are uneven as well.


The sidewalks are all made from sheet styrene, about 0.060 inches thick. It is all cut to shape, and then scribed for expansion joints. Andy Miller is our expert sidewalk painter, airbrushing them a light grey, and they overspraying with a spatter of brown and darker grey. They are glued down over the lattice stock that forms the sides of the road, overlapping slightly onto the paving, making smooth curves.

Trolley Wire

The overhead wire is in the last stages of completion. The main loop has been closed, but there are some adjustments needed at the backside of the loop, as one frog causes consistent dewirements.

The wire is hung 3 inches above the track, which is 1/4" higher than the NMRA standard. This was more important in the old trolley system, which operated interchange freight, but has been continued in the current system. It does make it easier to reach a hand and arm under the wire for cleaning operations.

The trolley poles are mostly 1/8" diameter brass rod, which can be bought cheaply as brazing rod. Some older poles are steel, and there is a group made of bronze. (The bronze is close to impossible to solder to, not a good choice.)

The overhead wire is a very fine 0.01" diameter copper-plated steel, which is number 30 AWG. This wire is 33% smaller than the number 26 trolley wire commonly used in HO scale, so it looks much nicer. It is also extremely strong, and does not stretch with age.

The wire is held in place by trolley ears, which are soldered to the wire, and then connected to pullof and span wires which connect to the trolley poles. Most of these ears are old Suydam ears, the newer ones are from Precision Scale Company. The spacing between the pulloff ears on curves is based on prototype information from the 1920's, which is much closer together than many trolley modelers use.

One disadvantage of this steel wire is it's high electrical resistance. This requires that many connections be made from it to the a sturdy electrical bus under the layout, otherwise there will be slow spots on the layout. Over 50 trolley poles are soldered to a network of 12 gauge Teflon-insulated white/black candy-cane wire.


The track for MITCo is divided into conventinal track blocks, although there are currently no block switches. All the blocks are currently bussed together, and wired temporarily to a simple power pack. Since the overhead wire is not yet complete, the trolleys are currently running in two-rail mode. This is the way MITCo systems have always been built, the two-rail operation allows thorough testing of the track before hanging the wire, as it is exceedingly difficult to make any adjustments to the track with the wire and paving in the way.

Once the overhead wire is complete, then operation will switch to drawing power from the overhead wire, just like the prototype. The power comes from the overhead wire, goes down the trolley pole into the streetcar, through the motors, and on to the wheels and rails back to the power pack. Just like the prototype, the trolley pole makes ocassional blue sparks against the trolley wire.

Considerations are being given to installing DCC on MITCo, it's quite popular with trolley modelers, because the trolleys need to be able to get quite close to each other. We cannot use the standard computerized progressive cab control system the mainline uses, because it's not compatible with overhead wire operation, especially with the wire as a "common rail" system.

MITCo Bad Examples

There are some aspects of the current MITCo which we would not recommend other modelers try and copy. The biggest one is that some of the track is simply too far from the edge of the layout. Hanging wire over the track between the passenger station tracks and the tall wall of buildings requires squatting on the passenger yard. (There is the alternative of removing the great wall of buildings, but that is major project.)

Some of the wiggles in the track around Gifford City station came out a bit silly. They would have made more sense had there been platforms in the street, but the station wound up closer to the street than was expected.

We would suggest that anyone considering trying to use the same type of wire for the trolley wire note that if a hanger is pulled off the wire, it removes the copper plating as well. Also, using too much heat and active flux can remove the copper plating. We have not yet found any way to solder directly to the steel wire, whatever alloy it is will not tin. Generally, if a spot on the wire is so damaged, either we have to relocate the hanger, or we have to replace a section of the wire. This problem is a strong motivation for using closely-spaced trolley ears, so that the side tension on each one is minimized. This greaty reduces the chance of them pulling off.

Some of the overhead wires go for extremely long distances without any switch pans. This leads to having to manage very long lengths of wire during installation, since there is no easy way to make a splice. I did make some splice hangers by mounting a Pacific Scale trolley hanger in a vise on a Dremel drill press, and drilling two number 80 holes up through the hanger at an angle away from the center. These can be used like a prototype wire splice. It would be nicer not to need these, but I wound up using three because of places where I lost the tinning on the wire.

TMRC Trolley Trackwork Evolution

Various methods have been used to simulate girder rail in building the MITCo tracks through the years. Most prototype streetcar track has used girder rail.

H. Clark Frazier used two code 100 rails side by side, soldered together at the bottom. This resulted in a flageway (the groove) that was reasonably consistent with the NMRA trackwork standards. But the two wide railheads on each side made for track that looked far heaver (and shinier) than the real thing. This technology also depended on HO scale rail cross-sections that are different from what is available today, those rails had narrower bases than today's rail. Using this technique today would result in a very wide flangeway.

Dana Pettingill's track used two code 70 rails side by side. There were some spikes in the gap between the rails. They were also soldered together, somewhat, but the gap was rather wide to do this particularly well. It was also rather easy to damage the rail when cleaning out excess solder after the soldering process. Also, probably from the temptation of making soldering the rails together easier, the guard rails were often too close to the running rail, violating the NMRA guaging standards. This "check guage" error would often cause the wheelsets on the trolley cars to drift to being wide-gauge, and could lead to derailments.

John Shriver's 1980's trackwork used code 100 rail as the running rail. The guard rail was formed by putting a peice of code 70 rail sideways against the code 100 rail, with the head of the code 70 rail butted into the side of web of the code 100 rail. They are then soldered together using a big honking 100 watt soldering iron. This technique is reasonably easy on straight track. It is quite difficult on curved track, as the code 70 track has to be laboriously curved in a vertical direction, smoothly. The cross-section of railroad track (real and model) is designed to be make this difficult -- this is the direction that track is supposed to be strong in. This system results in very attractive rail, as the lip on the inside of the rail is very thin. There are also no check gauge problems. But there are problems with the flangeway getting shallower than the NMRA standard, if the code 70 rail is allowed to sit a little too high when soldering. Even when executed correctly, the flangeway is just deep enough for a NMRA RP-25 flange on the wheels, equipment with deeper US or European flanges tries to run on the flanges, and this works horribly, since there's no way to keep the bottom of the flangeway clean.

Trolley Paving Evolution

The paving used by Frazier and Pettingill was Savogran Wood Putty, which was TMRC's standard scenery material in those days. TMRC bought by the 100 pound tub. Wood Putty unfortunately was a bad choice for this, as it is designed to expand as it hardens. This caused it to mess up the rail gauge as it hardened, generally resulting in slightly wide-gauge track. This exacerbated MITCo's check gauge problems. The expansion also made it swell up over the top of the rail head, which left the streetcars wheels often riding on paving instead of track, which makes for poor electrical connection.

Frazier's wood putty was natural color, and painted black. As attempts were made to sand it back down even with the rail heads, white areas were exposed.

Pettingill's wood putty was colored by mixing in carbon black, eliminating the "white" problem. Unfortunately, wood putty mixed with carbon black hardens much more quickly, and winds up even harder than wood putty normally dries. This made sanding or scraping it out of the way of the rails very difficult.

Current trackwork has all been paved with plaster mixed with carbon black, sealed with diluted shellac.

Overhead Wire Evolution

The original overhead wire of MITCo was the standard 26 gauge phosphor-bronze wire, probably bought from Suydam, who also provided the hangers. It was hung to operate with trolley poles only. The wire is soldered to a small groove in the bottom of the trolley hangers, which are then connected to span and pulloff wires that eventually connect to trolley poles.

In the early 1970's, John Purbrick bought a huge spool of 30 gauge copper-plated steel wire from Eli Heffron's electronic surplus store in Cambridge near MIT. This wire is much smaller than the 26 gauge phosphor-bronze wire, and is extremely strong, and never stretches over time. The downsides of this wire is its high resistance (many poles must be fed electrically under the layout), and the fact that the copper plating can be pulled off the steel, resulting in a spot on the wire that cannot be soldered to for repairing the pulled hanger. We have since determined that this wire is probably MIG welding wire.

When Dana S. Emery started working on the overhead, he converted sections of it to this new trolley wire. He also attempted to make the overhead wire allow universal use by both trolley poles and pantographs. Unfortunately, the result was wire that worked horribly for both.

In 1976 John Shriver and Edwin Kreoker started replacing much of the trolley wire, all with the new wire. The compromise operation was removed, and pole-only configuration ruled.

Tech Model Railroad Club of MIT
MIT Room N52-118
265 Massachusetts Avenue
Cambridge, MA 02139

+1 617 253-3269
Email: tmrc-web@mit.edu