by John McWhirter & George Huckaby
December 20, 2008

Background History:

The Market Street Elevated Passenger Railway was intended to be the backbone of Philadelphia's transit system when it was opened in 1907, and over 100 years later, it still is. When Thomas E. Mitten started numbering the streetcar routes in 1911, Route "1" was reserved for the Market Street line but that number was never assigned either by the Philadelphia Rapid Transit (PRT) or the Philadelphia Transportation Company (PTC). The line originally began at the Ferries (Delaware River) on an elevated structure, descended into a subway at Front Street and ran under Market Street from 2nd Street to 22nd street when it emerged on the east bank of the Schuylkill River. It crossed the river while climbing up to an elevated structure where it continued over Market Street from 30th Street to 63rd Street. The line began with only 40 cars, (1-40) built by the Pressed Steel Car Company of Pittsburgh, PA in 1906. They added another 40 cars (41-80) from the same company in 1907, 20 more (81-100) in 1909 and 35 (101-135) in 1911. J. G. Brill provided 30 similar cars in 1911 numbered 136 to 165, followed by another 50 (166-215) in 1913. There were minor differences between the Brill-built cars and the Pressed Steel-built cars but the cars were interchangeable and all these cars were classed by the Philadelphia Rapid Transit as Class A-8. They continued in service until 1960. Class A-8 would be retroactively reclassified M-1 by SEPTA even though these cars were long gone by the time SEPTA was created in 1968.

When the Market Street line was extended to Frankford by the city in 1922, Brill provided 100 cars (series 501-600) of a different design in 1922. They were not interchangeable with the A-8 cars. These cars, assigned class A-15, were owned by the City of Philadelphia and were also in service until 1960. SEPTA retroactively classified these cars as M-2. Although none of the A-8 cars survived, six of the A-15 cars did survive retirement in 1960 as work train cars (Cars 532, 551, 559, 583, 585, and 589), but SEPTA's short sighted management permitted none to go to museums.

The photo at left shows the first Brill-built A-8 car at the Brill Plant. In the upper right hand corner is one of the 1500 Nearside cars that would be delivered to the PRT between September 1911 and November 1913. All the A-8 cars were 49' 7" long. All A-8 cars and were essentially two motor cars with both motors on one of the two trucks.

Shown at left is car 547, one of the 100 class A-15 cars delivered in 1922 for the Frankford extension. These cars were 55' 0" long but were also two motor cars with both motors on one truck.

In 1960, the City of Philadelphia combined with the PTC to replace all the A-8 and A-15 cars with 270 Budd-built stainless steel cars, which would be assigned classes A-49, A-50 & A-51. By this time some of the A-8 cars were over 55 years old and were becoming maintenance headaches. One Saturday morning, the entire brake rigging fell from an eastbound train of A-8 cars just west of the 15th Street Station, derailing it, ripping up track and third rail and shutting down the line for an entire weekend.

Of the 270 cars, 46 (601-646) would be single units with operator's cabs at each end and would be classed as A-49. The remaining 224 cars would be married pairs with operator's cabs only at the ends of each two-car set. In each two-car set, one car would have the motor-generator for the pair and would be classed as A-50. These would all be odd numbered cars from 701 through 923. The even numbered cars would be classed as A-51 and would contain the air compressor for the pair. Each two-car set or married pair were actually "mixed" marriages because each odd-numbered car came with General Electric motors and equipment while it was permanently coupled to an even-numbered car, which had Westinghouse motors and equipment. The City of Philadelphia owned cars 601-615 (15), 639-646 (8), 701-722 (22), 745-749 (5), 755-838 (84) and 840 (1). The remaining 135 cars were owned by the PTC. 100 cars, 601-646 and 701-754 were designed for future installation of air conditioning but this was found to be less than economical when tried later. All cars were 55' long, the same as the A-15 cars, and had four distinct humps in the roof containing fans. These humps would cause the cars to be nicknamed "Almond Joys" (see below).

These cars were seen in some of the 'Rocky" movies and they lasted until 1998. SEPTA had reclassified these cars to M-3 by the time they were replaced by 220 cars from AdTranz (see below right). These cars consist of 110 married pairs of cars from 1001-1002 to 1219-1220. The M-4 cars, manufactured by AdTranz, were brought into service in 1997 to replace the M-3 cars. The M-3 cars, although a great improvement over the M-1 and M-2 cars had bad riding qualities almost from the beginning of service. The trucks selected for the cars were not the ones that Budd recommended, which resulted in some severe side-to-side shaking. One of the authors of this report has ridden on the M-1, M-2, M-3 and the M-4 cars.

The HO scale model:

The HO scale model of the A-50/51 cars consists of a very well detailed resin casting mounted on an aluminum floor which has been designed to accept a Bowser 125120 mechanism. The shells we obtained in 2007 were nicely painted but they were neither numbered nor contained PTC, City of Philadelphia or SEPTA heralds. There were no known source of such decals at the time.

The first problem encountered was the inability of the powered car to take any radius less than 12". The second problem was the inability of the powered car to adequately pull the unpowered car. When these problems were first discovered and privately related to the manufacturer, Imperial Hobby Productions (IHP), we received a much less than professional response, so the cars were put aside for over 18 months.

When the club decided to look for other than New York City prototype subway cars to use on their increasingly popular subway module, the club pulled these cars off the shelf and began to tinker with them again. The powered car was disassembled and the normal method developed by the club to allow Bowser powered cars to negotiate smaller radius curves was employed. It was at that point that we discovered there were two more problems to deal with. Initially it was discovered that, in a rather serious engineering lapse, IHP placed body mounted floor mounts in a location that restricted the swing of the power truck to much less than that allowed by the aluminum floor. In fact, three (indicated by the brown arrows in the photo below) of the four floor mounts were placed in locations that either impaired the swing of the power truck or the trailing truck. There were two other mounts, which were useless as they were placed above underbody equipment which made it impossible to insert screws to attach the chassis to the mounts.

The other problem was due to another error by IHP in designing the floor for the mounting of the power truck bolster under the floor instead of on top of the chassis floor (as is the normal mode for the Bowser traction mechanism) and then not have the floor milled under both motor supports to permit the correct angle of the drive shaft between the power truck and the motor. The photos below left of the 125163 Bowser floor for the Custom Traxx/Miniatures by Eric CLRV when contrasted with the "Almond Joy" aluminum floor in the below right photo clearly show this awkward angle. The angle causes the drive to be much noisier than is normal from a Bowser drive.

Having solved the minimum radius problem with the removal of the mounts and creation of new ones where appropriate, the club decided that the best way to eliminate the second problem was to power the second car. So one of the authors of this article contacted the supplier IHP to see if a second aluminum floor for the powered car could be obtained. The initial response from IHP was that the powered floors were not currently available but, in a totally unexpected but surprising event, a second very positive response contained an offer to swap the entire unpowered car chassis for a powered car chassis. Since this would be extremely favorable to the club, the unpowered chassis was returned to IHP on November 6, 2008 and the powered floor arrived on November 17th. Installation began immediately. All of the solutions applied with the first floor and body shell were applied to this floor and shell also.

Hoping to operate the train at the December 6-7, 2008 Great Train Expo in Ventura, CA, work proceeded on the powered car. Unlike the models, the prototype cars were married pairs and were permanently coupled with a drawbar. So we fashioned a drawbar from Evergreen StripStyrene #224 1/8" hollow styrene tubing so that all wiring between cars could be sent through it. Radial couplers were installed at the outer ends of the married pairs. The photo below shows the fabricated drawbar containing the two wires connecting the wheels of both units electrically. The two additional wires were installed after this photo was taken. After a little handling of the two cars, it was concluded that fashioning a drawbar in brass would be more sturdy so an identical drawbar was made from brass tubing and flat stock.

After extensive testing of these floors, it became obvious that we would have to lower the motors to have any hope of decent operation. At this point, we had two choices, mill out the floor for both Bowser #1279 motors mounts or mount the floor directly to the aluminum floor with Aquarium sealer and mill out the floor to clear the A-line 20040 flywheel. We opted for the second path. The milled out areas on both powered floors are shown below:

Both floors were then reassembled with the motors attached with Aquarium sealer and the flywheels over the milled areas. The drive-lines from both A-Line 20040 flywheels kits were reinstalled along with the four wires between the units. Two wires (red and black) connect the wheels on each unit, giving the total combination 16-wheel electrical pick-up. Two other wires (orange and gray) carry the motor output from the currently planned single decoder and motor in the first car, 849, to the motor in the second car, 876. The almost finished chassis' are shown below:

These chassis were taken to the Southern California Traction Club test track and they ran much quieter and negotiated the 9" radius curves with no growling noises and derailments, so the 10" radius curves in the subway module should cause no concern. Since Bowser, Custom Traxx and the Southern California Traction Club work closely together on new HO scale traction products, it was decided to mill a third floor to take the Bowser motor with supports and provide this floor to Bowser for future reference should they manufacture more of these floors or any floors with the bolster mounted from below.

Since these subway cars have no interior lights, we decided to make the dummy headlight and taillights at the operating end of in each car functional. At the same time, we installed a Train control Systems (TCS) M1 w/BEMF decoder in the first car, 876, to operate the headlights and taillights in that car. For both the headlights and taillights, we chose to install 1.5 volt lights in the ends. We drilled out the lights starting with a very small drill, #72, and then slightly larger ones culminating with a #55 drill. We then obtained Miniatronics red and clear 1.5v 30mA bulbs, painted the bulb black except for the lens end. We also painted the interior of the holes drilled also with black paint, allowed the paint to thoroughly dry, inserted the bulbs and secured them with a moveable glue. Then the issue of the correct amount of resistance to be placed in series with these bulbs arose. Most of the literature that we read gave us the formula for one bulb but we were going to have two bulbs in parallel. Our calculations showed that the correct resistance was about 208 ohms. Two of these resistors, one for the headlights and one for the taillights would be required for each car.

Another method to discover the correct resistance is to use a Resistance Substitution Box. We have at our disposal an old Heathkit RS-1. We started inserting resistance at 1000 ohms, then removing resistance a little at a time. The lights illuminated at 220 ohms.

At this time, it is planned to install a function decoder in the second car, 849, to operate headlight and taillights in that car but this will be done at a later time. Meanwhile the taillights in car 849 were connected temporarily. The lights were installed along with all the connections but not the function decoder. Car 876 is shown below with the headlights (left) and taillights (right) operating.


We decided to model married pair 849-876, the odd couple created after car 875 was destroyed in the now famous York-Dauphin wreck in December 1961. This wreck, which occurred just after the Almond Joys were placed in service, resulted in the first fatality in the lines history. Car 850 was damaged beyond repair when a tower operator opened a switch at Bridge street while the train was entering the station and impaled the car into the center island platform. All married pairs were sequentially numbered when delivered such as 701-702, 849-850 and 923-924. We obtained some numbers from a Microscale lettering set and applied the numbers. We also assigned this set as a "B" train. These rush hour skip certain stops with the "B" train normally by-passing six stations. Additionally, we were privileged to use these cars to test the new wheels intended for the new Bowser San Francisco 'F-Line' PCC cars due next year. Car 876 with the test wheels is shown in the next photo:

So after some work and modifications, the club had a pair of 1960-1990 era Philadelphia subway cars for their subway module. Because of the reliability of the Bowser drives and the plentiful supply of parts, the club looks forward to operating this set of cars for years to come. Because a three-car train would look much better on the SCTC layout and actually represent Sunday level service on the line, we decided to obtain two shells and kit bash into one of the A-49 cars with operator cabs at both ends. IHP was contacted and agreed to furnish all necessary parts for a very reasonable price. This will take some time to complete so the club will be operating the 876-849 two-car combination for some time to come.

After all of our work, we really wish that traction model makers would remember that the NMRA has standards on most of the aspects of trolley/traction/subway models in HO and O scale, and then make an attempt to follow them as much as possible. Bachmann made a valiant attempt and Bowser will also. These standards includes minimum radius and radial coupler standards in addition to many others. These are available on-line to everyone so there is no excuse for any manufacturer, especially those with active web sites, not to know and attempt to comply as much as possible if not totally. Models which are sold not-painted but ready-to-run are intended for true traction modelers, and most of these modelers operate on 6 to 9 inch radius curves, so it seems foolish to put models on the market that have to be modified so much just to be able to run on the modules of clubs like the East Penn Traction Club or the Southern California Traction Club. Most credible traction brass model builders got that right years ago so today's manufacturers should also "get with it". And it might be nice to thoroughly test powered models to make sure that they run as expected. It also seems incomplete in the current environment to sell unpainted models without information on prototype paint schemes and decals, or at least providing a source for same, especially if such is available on the open market.

At this time, we have assembled several IHP traction kits and have received comments and reports from several other modelers, who have also assembled IHP products, including members of the East Penn Traction Club. The current consensus is that at this time IHP provides excellently detailed resin shells (some of the best available) as evident in their recent PCC-II, Single End Kawasaki LRT (second run) and the Almond-Joy models that we have experienced. But it is also felt that some of the complementary items, such as the floors and trolley poles that we have experienced, range from barely acceptable to poor. IHP would be well-advised to concentrate on developing, producing and marketing their excellent resin shells but should defer to or team with someone else to create and produce these complementary products, especially the floors. Almost every IHP-designed floor that we have tried to use had some major operational problem due to some problem which would have been discovered during an adequate testing program and which required a lot of work to correct. The more effort that a modeler puts into the assembly of an IHP kit, the better is the result. But a lot of this effort is required due to shortcomings built in to the kit due to lapses in system engineering lapses and errors in system integration. Despite these problems, a talented and skilled modeler can obtain a really nice model.