How To Check Journals

The subject of checking journals came up recently, and in particular whether you should allow the spring-loaded lids to slam shut.  I was told many years ago by several people that Henry Cordell in particular, the master mechanic of the North Shore Line, was very adamant about not allowing the lids to slam.  And I was taught to let them down gently.  It's not hard to do:

So this was the Car Dept. official policy, I believe.  But I'm afraid I don't recall exactly what the bad effects of allowing the lids to slam shut might be.  Constructive comments are welcome!

Buzzer Wiring

I'd gotten some questions about buzzer circuits, and finally dug out my wiring diagram -- from 45 years ago.  This was drawn for car 309, but I believe the circuit is the same for any CA&E car, except that the interrupter may be at the #2 end, and of course for the steel cars the buzzer is connected not to line 7 in the control circuit, but the buzzer jumpers.  The buzzers will have a typical value of about 500 ohms.

Reset Switches

This is your place to find answers to technical questions about electric cars of all sorts.  So....

The question came up recently on how the reset function works on our interurban cars, whether North Shore or CA&E.  Most of them have a reset/power switch ("canopy switch") located above the controller.  When in the center position, the control system is off; moving the handle to the right turns the controller on.  When moved to the left, it will reset the overload relay if tripped, and there's a spring return so the switch cannot be left in the reset position.  And you can reset only by turning the control system off.

Note that this reset switch should reset the overload relays on all cars in the train.  The way that works depends upon the particular type of control system.  Below is a simplified HL diagram for Westinghouse.  

In the center at top, the reset switch applies power either to the controller, or to line 8.  And line 8 is part of the control jumper, so it connects to all cars in the train.  Lower down and to the left, we see the reset coil for the overload relay, which is connected to line 8.  So that's pretty simple.

For cars with General Electric MU control, such as the CA&E steel cars, it's a little more complicated.  Here is the simplified diagram for C-165.  

In the upper right-hand corner, we see that the reset switch supplies 600V to either the controller or line 3.  And looking to the left, you can see that line 3 has connections to two circuits.  Contactor 4 is energized on all points, so whenever the motors are powered, line 3 will energize contactor 1 and provide some acceleration, its usual function.  But if the traction power is off, line 3 instead takes the lower path, and if the overload relay has tripped, the reset coil will be energized.  Again, this will affect all cars in the train.

Not all GE MU controls included this reset function, by the way.  The CA&E wood cars do not have overload relays, except for cars 319-321.  And there the reset switch is located inside the cabinet behind the motorman.  Now we have never found a wiring diagram for the 319-321, but I suspect that the reset function was not wired into the control circuits in the same way as the C-165.  And if I have some spare time, it shouldn't take long to verify that.  So if that's the case, even if the 319-321 were all in the same train, the overload relays would have to be reset from each car.

Any questions?

Switches in the Street

A couple of weeks ago I got an inquiry from some historians in Oneonta, N.Y. who were researching the upstate interurbans, and wanted to know how switches in street trackage worked.  I finally remembered to take some pictures, so this is for anyone who is not familiar with street trackage.

We have two switches of this sort in front of Barn 4, which we have been using for the past 50 years.  They are not actually paved over, but I think you can see that they easily could be paved without affecting the operation in any way.

First of all, on the inside-of-the-curve rail, we have this movable point in a large casting.  A small rubber block holds the point in the desired position.

To move the point, we keep a switch iron hanging on a nearby pole.  In a city, each car would have one for the motorman to use.  You pry up the rubber block, lever the point over, and put the block back in.  It's that easy.

On the other rail, there's a casting with no moving parts.  A single point is sufficient to guide the wheels in the desired direction.

And the complete switch looks like this:  

Finally, here's some of our street trackage, not yet connected to the general system.  Eventually we will probably have some switches in the street like the ones above.

I also mentioned to them that in later years some cities, such as Chicago, had automatic switches that the motorman could control electrically, but I'm not sure exactly when these came into widespread use.  I'm hoping some of our knowledgeable readers could supply this information.   Thanks!

CA&E Headlight Circuit

The CA&E had its own system for portable headlights, which cannot be interchanged with anything else at IRM.  The main parts of the circuit on each car are the headlight switch and fuse in the electrical cabinet, the resistor box under the car, and a DPDT switch at each end in the ceiling above the motorman.  The headlight has two leads, connected across the bulb.  One plug is white and the other is black, and these go into two receptacles in the car, labeled B and D (for bright and dim).

You will observe that the direction of current is reversed when switching from bright to dim.  The reason for this has to do with arc headlights, explained below.  It makes no difference with incandescent bulbs, of course.

However, the headlights were later modified by adding a small resistor to the headlight, which grounds through the case.  This has the effect of making "dim" somewhat dimmer.  But this resistor can be removed or disconnected, so that the headlight does not ground through the case in either position, and so that it doesn't matter if the headlight leads are switched.  This change thus makes the system safer.  Some but not all of our available CA&E headlights have been thus modified.  In any case, as a general safety precaution all crews should be careful not to handle a headlight or even open the train door when the lamp is on.  On the North Shore cars this is especially important.

The DPDT switch is enclosed in a metal casting that looks like this:

In this picture the switch is in the center off position.  Inside the casting:

I believe most of the wood cars originally had the receptacles for the headlight plugs located one on either side of the train door, as seen here.  In our collection, the 308 and 309 are still this way.

But later, blocks with two receptacles were mounted to one side, and this is the way all of the steel cars are arranged.

In fact, we have a couple of spare blocks from Wheaton in our collection.

The main cause of system failure, other than a bulb burning out, is the failure of one or more resistors in the box under the car.

 

Typical Results for 600V Trolley Voltage

	Dim	Bright
Voltage across bulb	60V	100V
Voltage across box	540V	500V
Current	2.7A	4.0A
Resistance across box	200Ω	125Ω
Resistance per tube	40Ω	42Ω
Power dissipated per tube	300W	650W
Power dissipated in bulb	160W	400W

Note: Five tubes in series when dim, three tubes when bright.

Of course, these values may vary depending upon the particular headlight bulb.

CARBON ARC HEADLIGHTS

Carbon arc lamps operate by starting a spark between two carbon electrodes.  The spark continues by allowing a current between the electrodes due to the generation of a hydrocarbon gas, which provides a very intense light.  The carbon arc lamp is generally a pretty rugged and reliable device, which does not require a vacuum tube, but it is more complicated, and the carbon electrodes are gradually consumed as the lamp operates.

When a carbon arc lamp is operated on DC, the positive carbon is consumed about twice as fast as the negative one, and a concave pit is developed in the end of the positive carbon, known as the crater.  The crater is very hot, and most of the light is generated by this crater.  For general illumination, the positive carbon would be placed directly above the gap, as shown below in Fig. 238.  But for a searchlight or headlight, the carbons would usually be placed at an angle, as shown in Fig. 239, to direct the light in the desired direction.  Of course, in a headlight the parabolic reflector is placed to focus the light also.  (Fig. 240 shows an AC arc lamp, which doesn't concern us here.)

When the headlight is switched to dim, the direction of the current is reversed, and this makes the upper carbon negative in Fig. 239.  The result is to make the rate at which the carbons are consumed more equal, and also to diffuse the light being generated somewhat.

The mechanism for a carbon arc lamp was rather complicated, but I will present a simplified explanation.  The upper carbon is held in a movable holder, which is adjusted by two solenoids acting in opposite directions.  There is a solenoid for lifting the carbon, wired in series with the arc, and a solenoid for pulling it down, wired in parallel with the arc.  These are known as the series solenoid and shunt solenoid, respectively.  There must also be a resistor in series with the lamp when it is operated from a constant-voltage source, such as on an electric car.  And the lamp has a dash pot, to slow the motion of the holder.

When the lamp is off, the upper carbon drops by gravity until it touches the lower carbon.  In this case, there is no resistance between the two elements, and when the lamp is turned on, maximum current flows through the series solenoid and through the carbons.  The series solenoid then starts to raise the upper carbon, starting the arc.  As the arc widens, the voltage across it increases, and this voltage is applied to the shunt solenoid.  The downward force exerted by the shunt solenoid therefore increases until the net force is zero, and the upper carbon remains in place, with a fixed voltage and current.  As the carbons are slowly eaten away, the arc widens, the voltage across it increases, and the solenoids automatically adjust the gap.  If the power is turned off, as may often happen on an electric car (especially if you have third-rail gaps), gravity once more causes the upper carbon to drop down, and the process repeats.

I can only imagine everybody was glad when incandescent headlight bulbs came along.

Reference: Practical Applied Electricity by Moreton, 1911

Traction Motor Rewinding

 As long as we're going to be begging for money to pay for traction motor rewinding, we might as well make it educational:

I recently bought a book called "Electric Car Maintenance" at our own IRM Used Book Store.  Say, did we ever mention that the Used Book Store is a great place to shop?  I even saw an article to that effect in Rail and Wire, so you know it must be true.   Anyway...

The book is a compilation of articles from old trade journals.  Here's a very interesting table of useful data for GE traction motors, if you're into technical details.  The GE-66 is the last row in the table on this page.  Some of the columns are pretty obvious, but others require some comment.

First, we see that the maximum axle diameter is given as 5 3/4", but the CA&E cars have 6 1/2" axles.  Maybe that's what puts the "B" in GE-66B motors.  In the first few years, the CA&E was providing much faster service than anybody else, and they had problems with bent axles and broken gears.  The actual gear ratios on the CA&E motors are either 20-32 or 25-40, since they needed bigger, stronger teeth than the table provides for.

The "No. coils in armature" is actually what I would call the number of poles.  There are 39 poles and 39 slots, so five wires per pole gives 195 windings and therefore 195 commutator bars.  The "throw of coils" is important if you're rewinding the armature.  Basically, "1-10" means that each wire leaves a commutator bar through the first slot in one direction or the other; at the other end it then wraps around one fourth of a circle to the tenth slot and back to the commutator.  Notice that the "throw" (in this case 10-1=9) is always equal to the "number of coils" divided by 4, rounded down.  (This will be on the test.)

For some of the motors they give a "safe wearing diameter" of the commutator.  For the GE-67, for instance, they in effect say that you can safely wear down the commutator bars by more than 3/4".  I find that hard to believe, but in any case we're never going to run the cars enough to find out!

Finally, the GE-66 is wound not with wire, but with copper bars 1" x 1/8" that are specially formed and bent.  All of these bars will need to be rewrapped with new insulation.  I would expect that the winding bars should be OK; we know that in one place several of the commutator bars were melted away right down to the axle when the short-out occurred, so they need to be replaced.  That, among other things, is what makes this such an expensive proposition.

Reverse Engineering

While looking through some old papers, I came across a sheet with calculations I made several years ago, trying to figure out what actual resistances were used in the control system for the wood cars.  It's rather cryptic, so I reformatted it and added some explanations.  This should apply to all CA&E cars with GE-66 motors.

We start with the data for the grid elements used on these cars, taken from the GE data sheet:

Grid Type	Resistance (ohms)	Current Rating (amps)
4	0.023	75
6	0.038	60
7	0.047	55
8	0.059	50

I might note that because our grids are old and rusty, the actual resistance may be higher than the nominal value.

Then here is the resistance part of the motor circuit schematic.   R1 through R7 are labels both for the contactors, and for the resistances between them.  It should generally be evident from context which is which.  All of these contactors are connected to the 600V supply when the trolley contactors are closed.  RX is just my name for a junction point in the circuit.

The next table lists the arrangement of grid elements in each box, the total resistance of each group, and the effective resistance for the resistors shown in the schematic.

Box	Grid Arrangement	Connections	Resistance (ohms)	Resistor #	Total Resistance
1	8A10	R1-R3	0.590	R1	0.590
	6A8	R3-R4	0.304	R3	0.456
2	6A4		0.152
	4A14	R4-R4A	0.322	R4
3	6A18	R2-RX	0.684	R2	0.682
4	4A10	R4A-RX	0.230	R4	0.552
	7B8	RX-R7A	0.094	R7
5	8A7	R5-R6	0.413	R5	0.413
	6A5	R6-R7	0.190	R6	0.190
	7B6	R7A-R7	0.071	R7	0.165

Note that some of the resistors are split between two boxes: R3, R4, and R7.

Now, starting with the control sequence, we can calculate the total resistance on each point as the controller is advanced.   I also calculated the percentage by which the resistance is decreased on each step.  Points 1 through 5 are series, 6 through 10 are parallel. 

Point	Contactors Closed	Resistors in Circuit	Total Resistance (ohms)	% of Previous Point
1	R1	R1+R3+R4+R7	1.76	--------
2	R1,R3	R3+R4+R7	1.17	66%
3	R1,R3,R4	R4+R7	0.72	61%
4	R1, R3 to R5	(R5+R6)║(R4+R7)	0.33	46%
5	R1,R3 to R7	None	0	0
6	R2	R2+R7	0.85	--------
7	R2 to R4	(R2║R4) + R7	0.47	55%
8	R2 to R5	((R2║R4) + R7) ║ (R5+R6)	0.27	57%
9	R2 to R6	((R2║R4) + R7) ║ R6	0.135	50%
10	R2 to R7	None	0	0

+ means “in series with” ║ means “in parallel with”

So starting with the question "Why in the world did this have to be so complicated?" we have a least a partial answer.  This was really an interesting problem to work out.

Bearing Project

I was out at the Museum briefly yesterday to work on the 309 bearing project.  I've been revising the patterns at home, and the big sanding disk machine was needed to smooth off the edges.  The patterns now look like this.  I showed them to Bob Sundelin, who will be doing the machining on the castings, and we talked over what needs to be done.  So we're ready to take the patterns to the foundry in Chicago.  Thanks again to everyone who contributed the funds needed to make this happen!

In the shop we happen to have a spare GE traction motor, which might help show what we're trying to do, for those of you unfamiliar with the gritty details.  I believe this is a 207 from the L4, but it's not important.  The basic design is pretty standard.  

This is the motor case with the axle caps and axle bearings removed.  This is the commutator end; the commutator is under the large hatch cover (yellow arrow) on top of the case.  The inner (solid) halves of the bearings fit into the circular parts of the case at each side (white arrows).  The large bolts above and below the bearings hold the axle cap castings to the case, and support the motor on the axle.

At the top (white arrows) are the axle cap castings, with the oily waste visible.  Below that (red arrows) are two bearings, with the other two out of sight behind the axle caps.  The outer half of the bearing has a large opening for the waste to contact the axle surface and keep it well oiled.  These particular bearings are broken, as pointed out by the arrows -- ouch!   In fact, that's why this motor was removed from the locomotive.  But when the axle bearings are assembled, they wrap tightly around the axle and keep the motor pinion properly aligned with the gear on the axle.

And finally, here's the pinion end of the motor.  The black, greasy pinion is obvious, and below it is one end of the gear pan.  The white arrow points to the lid covering the opening for the armature bearing.  The bearing itself is inside the case directly behind the pinion.  You have to open the lid to check and refill the oil on the armature bearings.  It's an unpleasant task, but somebody's got to do it.

Meanwhile, the usual projects continue.   Here Ed Oslowski is working on the floor in the Liner.  I wasn't around long enough to get many other pictures.

However, Barn 13 was being switched out for various reasons, and the two THI&E cars wound up outside next to each other, so it was a good opportunity to get several good shots.

That's it for today, but we'll keep you informed on the bearing status.

Sunday operations

Frank writes...

I barely took any photos on Sunday but of course Chuck was out and was taking good photos so I'm off the hook! I was conductor on the 308 and 319 with Jeff Obarek as the motorman. The crowd was light, even lighter than on Saturday, with the similar heat (highs in the mid-90s with the heat index up around 105 I think) and predictions of heavy thunderstorms in the afternoon keeping people away. Oh well. There were still visitors and they certainly enjoyed the parade of equipment that was out. There were six trains running: the CA&E cars, CRT cars, steam coach train, and caboose train pulled by ComEd 4 on the main line; CSL 144 on the streetcar line; and IT 101 doing shuttle trips.

Here are the CA&E cars on Station Track 1. Beyond the train is the 101 and if you look closely in the distance you can see the 144 passing through Car Line Junction. Usually we run Jefferson Street Shuttle trips out of 50th Avenue but of course the 101 isn't very conducive to doing that. So Joel, who was running the car with Thomas as his motorman, just ran out of Station 1; he'd follow one of the revenue main line trips out and short-turn at the east switch, heading back to the west end and straight back into the station. That seemed to work pretty well.

More of the same. The CRT wood cars, with Bob Opal and Andy Sunderland as the crew, were brought out in the morning as a three-car train with the 24, 1268, and 1797. But almost immediately the 24 developed a fault and had to be set out on track 41 east. It turns out that insulation between two of the control wires in one of the conduits failed, leading to a reverser failure that meant the car would only go in one direction. Once it's fixed the car will be back in service.

When we returned around 2pm from our third main line trip, it was reported that a line of thunderstorms was approaching from the southwest and due in around 3:30. Rather than risk another main line trip at 2:15 we swapped places with the 101; they did a main line trip and we did a short-turn shuttle trip to Jefferson Street at 2:25. Then the train went straight to Barn 8 thanks to some deft dispatching by David Streeter. And it's a good thing too - an advance line of showers had popped up and passed through about three minutes after the cars made it indoors. An hour later, as predicted, heavy rains came through and drenched everything that was still out on the railroad.

By that time Joel was hard at work on a repeat of Saturday night's cookout in the car shop. While we were hanging around after dinner, Scott and I started looking over the C-21 controller that was donated on Saturday, soon joined by Richard. It's definitely more complete than our other examples and, most notably, still includes the large torsion spring on the handle shaft. See here:

Richard figured out that there's a spring-loaded button that engages and disengages the direct gearing between the handle shaft and the contact drum. At the start of the video the direct gear is disengaged, so the handle shaft (above the torsion spring) is advanced with the handle while the contact drum (geared to the lower plate that he is holding from advancing) stays motionless until released. At :08 the button is depressed, so the entire controller is "solidified" and functions like a normal C-6 or something; Richard tries holding the lower plate steady but can't. Then the button is released again and you can see how the handle doesn't advance the contacts until the pressure - in this case, provided by hand - is released.

But wait, there's more. That video showed the contact drum advancing almost immediately once the pressure was released. But there are small dogs underneath a gear on the contact drum (Richard is fiddling with them at the very start of the second video, above) that seem to affect how the contact drum advances. With these moved, the contact drum advances more slowly, with the speed controlled by a flywheel inside the black-painted cast housing at the top of the contact drum.

I'm pretty curious exactly what the thinking behind all of this was, particularly the function of those dogs shown in the second video. My best guess is that perhaps there were two different ways to automate the car's acceleration: either by limit relay (this would be shown in the first video) or by simple timing, with the flywheel controlling the rate at which the contact drum advanced (as in the second video). The CA&E controlled acceleration using a limit relay under the car but Richard had mentioned that simple timing was another way of doing it. Any thoughts?

Modern technology

Frank writes...

Okay, so this post is the result of my visit to IRM some two weeks ago, but I haven't been out since and what it makes up for in information it lacks in time-sensitivity. So I figured it would be okay posting this tardily. You're getting what I got when I stopped out on a Sunday afternoon: a primer in some of the most modern technology ever applied to an interurban car. In this case, it's a Houdaille rotary shock absorber off of CA&E 460, which was on the pit for its annual inspection.

Here's a photo of one of the shock absorbers in place. The car has four (as does sister car 451), one on each side of each truck, and they're designed to lessen the roll of the car by dampening the action between the truck frame and truck bolster. Houdaille made tons of these things, mainly for the automotive market, though those tended to be much smaller I was told.

Anyway, here's what one looks like disassembled on the bench in the shop. Richard, Greg, Joel, and Good Nick were working on this thing. It had been removed from the car and disassembled the previous evening in what was, by all accounts, a herculean team effort owing to incredible buildup of gunk and crud. Anyway, it's not a tremendously complex system. In the left foreground is the body, which bolts to the truck frame. There are four chambers inside the body; two of the dividers are cast into the body and the other two attached to the shaft, which rotates as the arm (attached to the truck bolster) moves. The arm to the right, fits right over the shaft after the top portions of the housing are installed. When assembled these chambers have hydraulic fluid in them, which originally would have been glycerin, and the rotary motion compressing and expanding the internal chambers provides the dampening action.

Here's the cover after quite a bit of cleaning. If I recall correctly, the top of the main chamber (visible in front of that McDonald's cup in the previous image) gets installed over the body and then this cover screws on over that, with the space between the top and cover used as a well for hydraulic fluid, filled using the port visible at the top of the cover. Hopefully Richard can comment with corrections of whatever I'm getting wrong here.

And here's what a cleaned and serviced Houdaille interurban rotary shock absorber looks like when it's been reassembled. All four of the ones on the 460 appeared to be complete and in generally good condition, but none seems to have any hydraulic fluid left in it and this one contained quite a bit of gunk and goop. The result is that there's no dampening force; they're pretty much just along for the ride. Once these are all rebuilt - and refilled, albeit not with glycerin - the car should ride more smoothly and roll less. Time to put this technology to work!

Pattern Making Made Cheap!

As mentioned before, we have a problem with the #1 truck on the 309.  The bearing surfaces on both axles were turned down at some point, which would require thicker bearings.  But none of the bearings we have available are thick enough, so there's too much play between the axles and the motor mounts.  This needs to be corrected.

Here's what an axle bearing looks like.  It's in two halves: the one on the left I'll call the inner half.  It fits up against the motor casting and has no holes, but is completely solid.  The other half is the outer half, and is held in place by the axle cap, which is bolted to the motor.  There's a large oval hole where the waste rubs oil onto the bearing surfaces.  There are also holes for pins to keep the bearings from rotating relative to the motor.  There are four holes, but only two pins, so that the bearings are the same for either left or right.  The thicker end of the bearing, which is sitting on the table in this picture, butts up against the wheel or the gear and takes the stress of lateral movement by the motor.

For this particular motor, the GE 66B, the outer surface of the bearing has a diameter of 7 1/4".  The nominal axle diameter, and therefore the inner diameter of the bearing, is 6 1/2", but on this truck we're talking about, that's been turned down to as little as 6 1/8".   Therefore, we need to make two patterns for new brass castings that will be thicker.  After casting, they can then be machined to the correct diameter for each axle.  And the spare brass castings are an excellent basis for patterns.

Tim Peters pointed out that we had some leftover pieces of 6" PVC pipe.   The inner diameter is 6", of course, and it's 1/4" thick, so the outer diameter is 6 1/2".   That means it fits almost perfectly into a spare bearing.  So I took it home and started to cut it to size in an obvious fashion.

I prefer to use an ordinary hand-held hacksaw.  It won't get away from you like power tools can.

The next step will be to attach the PVC with contact cement, and carefully smooth the edges flush with the brass, using a rasp.   First, though, I want to check with the experts in the car shop before going any farther.

In any case, we should have usable patterns at basically no expense.  Pattern making is often a large part of the expense of casting, and if we didn't have these spare bearings available, it would be an extremely challenging task to make them out of wood, or whatever.

Once the patterns are done, we will get quotes on casting new bearings, then start passing the hat.  I promise to do my best to make you feel guilty.   It's the least I can do.