SP 1941 Tunnel 26 Accident - The Official Interstate Commerce Commission Report

Concise and to the point, the facts included in the official accident investigation were recorded in just over 6 typed pages. A diagram, executive summary and cover sheet bring the total up to 10 pages. 

The website to find US railroad accident investigations 1911-1993 is linked at the bottom of this page. 

Here is the sheet with the vital data and the summary of what was determined.

Below is an engine of the same class as Engine 4193, leading a passenger train. 

SP Cab-in-Front 4194, Train 55 The Tehachapi, undated, location unknown.
from: The Central Pacific & Southern Pacific Railroads; Lucius Beebe; 1963; Howell-North.

*  *  *

The line diagram to the right represents an overhead view of the track plan. The headend of the train ran roughly to the midpoint of the 7366-foot Tunnel 26. There, an application of the brakes in emergency (after a knuckle break) gave the engineer, fireman, student fireman and headend brakemen few options.

Key points on the track diagram are linked to a track profile - the line diagram at the left. The engine stopped at the summit of Tunnel 26's 1% grade, just before the track transitioned to the slight descending grade leading to the west portal of Tunnel 26. (This is a very efficient method of presenting a lot of data.)

The box at the left shows the distances between the points listed.

*  *  *

Images from Today

Google Earth, date: 30 August 2024.

Above: The two distinctive 90-degree curves at the east end of the image are found on either side of very short Tunnel 28. The railroad was engineered to decrease the gradient by adding distance. After passing through a rock cut, the line begins to turn west as it passes through Tunnel 27. The track curves to line up with the 7366 feet of tangent track through Tunnel 26. At the left edge of the image, the thin grey line of track exits at the west portal. It can be seen between the multi-lane highway and the winding Santa Susana Pass Road - which originally traversed these mountains.

*  *  *

The grade to the east portal of Tunnel 26.

Below: In the lower right corner, the thin grey line of the railroad can be seen before it enters the rock cut (just below it is the smooth curve of an unnamed dirt road). The track passes through Tunnel 27. It then curves to the right to enter the eastern portal of Tunnel 26. It was on this track, and beyond inside Tunnel 26 that the Extra 4193 West got into serious trouble.

I don't know if the 1% grade is accurately portrayed in this image. The slow-moving train's flange resistance against the rails on the curves, combined with the gradient, challenged the motive power. On the night of the accident there were problems with driving wheel adhesion to the rail in this area and inside Tunnel 26.

East Portal of Tunnel 26 is at:  34°15'47.77"N 118°37'20.20"W
from: Google Earth, circa 2026.

*  *  *

Signals, West Portal of Tunnel 26 at Hasson, undated.
from: The Central Pacific & Southern Pacific Railroads; Lucius Beebe; 1963; Howell-North.

Extra 4193 West never did reach these ABS signals at the east end of Hasson siding.

The single absolute semaphore controlled eastbound traffic at Hasson (entering the west portal of Tunnel 26).

The double semaphore controlled westbound traffic at Hasson (on the mainline or into the siding).

*  *  *

This section provides a lot of civil engineering data, some of which I have illustrated with Google Earth images. 

There is no ventilation system for Tunnel 26. The wind through the tunnel in the report seems to correspond to the normal coastal wind patterns on land (i.e. The land is heated by the sun, warm air rises creating an area of low pressure, cooler air from the ocean is drawn into the area of low pressure. The cycle reverses at night). The normal wind currents between 24hr and 01hr - at the time of the accident - are not recorded. It seems logical that the presence of a train in the tunnel (with or without an engine) would obstruct most of the natural currents.

*  *  *

*  *  *

Technical Exhibits

The post is one of a series about the railroad line, the motive power and the circumstances surrounding this accident. The terms in the report will be used to identify these devices - most of which have been described at length in previous posts. While the oil firing devices are generally of the type used by the Southern Pacific circa 1910, they won't necessarily match those in use on the engine 4193.

Conceptually: Emergency drop valve. The image shows the fuel valve closed (Fig 4) and with the valve open and the 'spring compressed' (Fig 5). This device is intended to stop the flow of oil and extinguish the fire in the firebox in an emergency ... or during the normal course of extinguishing the fire at a terminal.

* *

Oil-regulating firing valve. 

The (customizable) drifting stop-pin (c) is designed so that when the handle is placed with part (d) at that the stop-pin (c), it will be held in that position. This maintains a low fire and is analogous to 'idle speed' on an internal combustion engine. Moving the firing handle to the stop pin (f) cuts off the supply of oil to the firebox.

*  *

The damper (e) admits air to the firebox. The damper control staff and the firing valve control staff descend through the cab floor ... with the oil regulating staff concentric with, and inside, the damper regulating staff. 

Generally, the damper should be opened more as the fire is increased (to meet the demand for more steam). A damper which is open more than necessary cools the fire and decreases steam production. A damper which is not opened enough leads to inefficient combustion and excessive smoke production which coats the flues with soot.

* If the engine is slipping, "the fireman must swing the oil regulator open, so as to give a good strong fire during slipping" (to avoid flue damage). 

*  *  *

Photos from a '4100 class' Cab-in-Front

Note: This implies any class from AC-4 (1928) to AC-8 (1939).

It is not necessarily identical to the cab of the 4139 (AC-8 class). 

The History of Western Railroads; Jane Eliot; 1985; Bison.

The view looking from the left cab wall, over the fireman's seat, toward the engineer's side of the cab. 

The firing valve can be seen above the damper control. (The open-closed positions may be reversed from the diagrams above. i.e. Here, the firing valve may be closed.)

The Worthington valve gives a readout on the boiler feed pump pressure.

In general, steam locomotives transmitted energy to operate pumps and other auxiliary appliances in the form of steam. Valves which sent steam down pipes were the common controls used by the engine crew. (Electricity on steam locomotives was an evolutionary 'add on' and electrical generation was generally performed by steam-driven turbo-generators.)

Control Labels, Left to Right

• Blower: Sends live steam up the smokestack to artificially draft the fire. An essential device for preventing dangerous draft reversals which result in combustion gases being blown into the cab.
• Oil (line) Heater: Heats the Heavy Bunker C Oil to 150-180 degrees Fahrenheit so it flows smoothly and atomizes correctly in the firebox.
• Blow Back: Used to direct high-pressure steam to clear obstructions in the fuel line: back to the tender tank, or forward to the burner.
• Atomizer: Supplies high pressure steam to the lip of the oil burner to shear the oil stream into a mist of fine droplets so it can burn. This is constantly adjusted at the same time as the oil regulator valve so the steam/oil always combine to produce an efficient fine mist. And, as mentioned above, the damper opening must also be adjusted to match the fire's needs.
• Tank Heater: Sends steam through heating coils in the tender so the oil temperature is maintained at 100-120 degrees F to ensure it flows well.
• Feedwater Heater: Uses waste steam from the cylinders and air pumps to raise the temperature of water being fed into the boiler to 200-230 degrees F. This improves thermal efficiency and prevents damage to the boiler from 'thermic shock'.

 

*  *  *

I Like Trains; ed: David P Morgan; 1980; Kalmbach.

Looking across the cab to the engineer, the fireman would see this. However, only an engineer posing for a Ward Kimball 'last run' story for Trains magazine would accept having a spotlight shining in his face (for the benefit of the photograph). You can imagine how dark the cab would be inside Tunnel 26 without this kind of unusual interior lighting. 

In the fireman's view, you can see two armoured cables leading to shielded lights over the gauges and the firing controls. On the engineer's side, a similar fixture is arranged over his gauge cluster.

*  *  *

The History of Western Railroads; Jane Eliot; 1985; Bison.

Presumably, in the same '4100-class' cab, we are looking at the engineer's forward-facing seat. We were just looking at the opposite side of the flat panel at the far right - which holds the fireman's gauges. Nowhere in these two cab views can I spot an obvious emergency cable control to shut the emergency drop valve to stop the flow of oil at the tender.

In the Discussion section of the report (farther below), the most junior person in the cab (the student fireman) will state that the engineer had difficulty closing the throttle throughout the trip. Many readers will know that when the driving wheels begin to slip, closing or decreasing the throttle quickly is necessary to stop the slip ... so the throttle can be opened slowly to re-gain traction before the train stops completely (i.e. the train 'stalls'). Dried sand can be blown down on the rail head in front of the driving wheels to help maintain or regain traction.

Comparing the two cab views immediately above, you can see that the 'spotlight in face' engineer has an ergonomically superior throttle - a upper quadrant type whose long arc of travel allows more precise adjustment.

In contrast, the throttle in the cab immediately above is mounted at some height above the firebox door. The engineer (as we imagine him in his empty seat) must reach above his head on his left-hand side to operate the throttle. You can see the handle is just above the engineer's gauge cluster. I believe the cutoff (reverser) lever handle is located immediately beside where the engineer's head/left shoulder would be.

• There was adequate sand to support traction and prevent slipping.
• The oil supply was left 'on' slightly.
• The coupling failure was a new 'clean break' with no indication of earlier weakening. (Coupler knuckles are used like fuses ... they are engineered to break first - before serious damage is done to cars' draft gear.)
• Theoretically, if the crew had foreseen being stuck in a tunnel, four respirators could have been signed out in Los Angeles (total on hand: six). If a photo I have seen is correct, this was a full head-covering hood with a hose exiting the top, and with a small window in the front. These hoods take air from the main air brake reservoir. The reservoir pressure is stepped down and the air is filtered, for emergency breathing.

SP Cab-in-Front 4233, south of Alturas, California, 1955.
 from: The Central Pacific & Southern Pacific Railroads; Lucius Beebe; 1963; Howell-North.

The photo of a 1942 AC-10 class is shown above to provide a reminder of the size and power of these locomotives. It also demonstrates the towering power of the engine's exhaust. The headend brakeman can be seen through the front windshield on the fireman's side.

*  *  *

Before modern diesel-electric technology completely replaced steam locomotives, the few inches of slack between each set of car couplers was used as part of the most common tactic for starting a heavy train.  

... In 'taking slack' the engine reverses into the train, to eliminate the spaces between the couplers. Beginning to pull the train with the 'slack bunched', the engine theoretically starts just the first car in motion, then just the second car, and so on until the front of the train has gathered momentum so there is no danger of driving wheels slipping.

... The Extra 4193 West might have had 30-40 feet of total slack between all the couplers ... so, when starting after bunching slack on level track, the engine may have advanced 40 feet before the caboose started moving.

The crew on the engine were as follows (from newspaper reports):

• Engineer, age 46 ('the person in charge' of those on the engine).
• Headend Brakeman, age 52, technically 'train crew' under the conductor - not 'engine crew'.
• Fireman, age 24, hired just two months earlier. Directly responsible to the engineer.
• Student Fireman, age 24.

(The conductor age 61; the tailend brakeman, age 25; and the tailend brakeman who died - age unknown, are essentially out of the picture and near the tunnel's east portal. They are in or near the caboose. Technically, the conductor is in charge of the whole train. Hand/light signals will not work reliably along the side of the train in the tunnel.)

It is not clear why one would assign a (probably) probationary employee, on a full tonnage train, to train a student ... at night. It seems probable that the student fireman was present on that typical railroad procedure of that era: the 'unpaid trial trip' to learn the job and/or the railroad line. It is possible the student 'just showed up' - hoping for the usual unpaid training help from the working crew.

... A steam locomotive requires the fireman and engineer to know where a hot fire and extra steam will be needed - miles in advance. The fireman must know the characteristics of the road ahead and new firemen would start developing a 'firing plan' on the line where they worked to avoid the career-enduring nickname 'Coldwater'. Needless to say, the fireman must also know how all the unlabelled controls on his side of a steam locomotive cab are to be used.

To Preserve Life (with the benefit of hindsight)

There was only 2 feet of clearance between the side of the engine and the wall of the tunnel. The safety valve was venting the excess boiler pressure (scalding superheated steam) in a very confined area at 250 PSI. This pressure had been built up in order to pull the whole train up the 1% grade. The headend brakeman is excused for not wanting to uncouple the engine from the train at this moment. 

Probably, the 'slipping, then stalling in the tunnel, with the train brakes going into emergency' would have been a scene of incredible chaos, heat and noise in near total darkness. Communication would have been extremely difficult. The crew would have been aware that they were rapidly losing breathable air as the firebox pulled in oxygen and created oil smoke.

If we accept that it was impossible to uncouple the engine, the engineer or fireman could instead have extinguished the fire immediately, waited for the venting steam to subside, uncoupled the engine from the train, and the engine would still have had plenty of remaining boiler pressure to run itself out of Tunnel 26. 

... Then ... the unknown reason for the emergency brake application could have been determined and corrected. This would probably have blocked the main line for a couple of hours. The engineer might have faced discipline and demerit points for poor train handling ... or ... for accepting an engine with mechanical defects at the roundhouse. 

... However, the local railroad culture probably would not have much patience for an engineer who delayed trains because he was always throwing hissy fits about a sticky throttle or a leaky cylinder cock. (Childhood story books never accurately present the difficult realities of being an engineer.)

The mature headend brakeman might have been serene about the possibility of losing his job for 'abandoning' the train. He did save his own life and that of the student. 

Each of the two senior men on the train was associated with the death of a more junior employee under their immediate supervision. Engineer: fireman. Conductor: tailend brakeman sent to the headend.

*  *  *

Possible cylinder cock leaking. 
from: The Central Pacific & Southern Pacific Railroads; Lucius Beebe; 1963; Howell-North.

Without realizing what I was doing, I may have provided an example of the kind of defect that the report describes at the 'Left, Number 1 Engine' cylinder cock. This detail comes from the previous train image above. The report perhaps suggests that the driving wheels of the 'Number 2 Engine' lost traction and slipped because of the film of water laid in its path by the leaking Number 1 Engine cylinder cock in the confines of the tunnel.

When I was trying to learn more details about this accident and did not realize that I could find and read the Official ICC Investigation, I found a message board conversation where someone wrote something like 'Yeah, I heard the slipping/the accident was caused by steam in the tunnel coming from a leaky petcock. In the tunnel the wall directed the steam/water onto the rail.' 

... I looked at the employee timetables and I have a different theory about the slipping which I hope to review and explain (as a possibility) in the next and final piece about this event. 

*  *  *

Locomotive Record Cards for the 4193

Form 4395; California State Railroad Museum; archive.org.
https://archive.org/details/form-4395-4150-4304/page/n89/mode/1up

The damage from the accident seems to be recorded on the 'detail side' of the card, seen below. The date to look for is 17 December 1941.

On the card above, 'Vacated' is the Southern Pacific term for taking a steam locomotive off the roster: 27 September 1956. 

It is interesting from a corporate accounting perspective to see that the Ownership (above) was recorded as Southern Pacific Railroad, however, the SP subsidiary company El Paso and Southwestern Railroad became the engine's owner on 1 January 1955.

Cylinder cocks: Air, Stm (above)

Cylinder cocks exist to drain water from the cylinders ... through a small port in the cylinders' bottoms. Water is not compressible so an excess amount of condensed water in a cylinder can cause bent driving rods, or it can blow the cast iron head of the cylinder off.

The SP 4193 was originally fitted with manually-operated compressed-air-activated cylinder cocks. While starting (for example), the petcocks were opened when the engineer used his remote-control air valve and line to allow the cylinder pressure to blow the water out of the port at the bottoms of the cylinders. 

Rolly once told me: 'You only have to open them for a second!' ... to blow out the water - otherwise you're wasting steam. 

When you watch the North American restored corporate steam locomotive videos, you'll observe that they'll often open the petcocks for multiple cylinder cycles to create huge crowd-pleasing clouds of steam as the engine departs. Just like the black smoke created for Lucius Beebe and the photographers of that era, these petcock steam displays are wasteful theatrics.

The switch to steam-activated cylinder cocks eliminated the challenge of routing long, compressed air lines to the cylinders (especially on the Cab-in-Front engines). Compressed air lines become contaminated with condensation and on some of the territory operated by the SP, this condensation would freeze and block the line - risking the previously mentioned damage to the engine.

In addition to always using hot, live steam lines ... the steam cocks were also designed to operate automatically, removing any human error from the protection of equipment. 

... When the steam pressure in the special device's differential piston was overcome by excessive pressure in the cylinder, the pop valve opened automatically to expel the water.

From the cards, I could not determine which type of cylinder cocks the 4193 had at the time of the accident.

Form 4395; California State Railroad Museum; archive.org.
https://archive.org/details/form-4395-4150-4304/page/n89/mode/1up

*  *  *

For more accident investigations,

you can set up specific searches at the link below.

United States Department of Transportation

Investigations of Railroad Accidents 1911-1993