Why the auxiliary generators on the Deltic Engines

[duncan webster 1Participant @duncanwebster1]

So how do cars get away with it, most are 4 or fewer cylinders? Is turbocharging a 4 stroke easier?

[Mark RandParticipant @markrand96270]

I believe that Honda were congratulated for their technical achievement when they turbo charged their CX500 V twin to make the CX500TC, both a twin and a “small” engine.

[Howard LewisParticipant @howardlewis46836]

The problem tends to be that an engine with a large volume cylinder takes a gulp when the inlet valve opens, suddenly reducing the pressure in the induction system.

The very large engines don’t suffer since, they are often constant speed units,(So turbo lag doesn’t matter)

The problem has decreased as turbochargers get smaller and lower inertia, so that they can speed up quickly to maintain boost pressure.

A large volume inlet tract acts as a pressure reservoir, but can result in “turbo lag” since the turbocharger has to repressurise a larger volume.

Truck, and construction equipment engines, now are often two stage turbocharged to meet emissions.

The smaller turbocharger has a low inertia, allowing it to speed up rapidly, but is wastegated to prevent overspeed at higher speeds and loads.

Private cars have less of a problem, since individual cylinder volumes are small, and the turbochargers are small and low inertia. We now have turbocharged 3 cylinder 1 litre engines, where volumes and inertias are small. Again, many turbochargers are wastegated, or have variable guide vanes, to optimise outputs without the risk of overspeed and self destruction.

(If a turbocharger overspeeds, it very soon enters the region where blade passing frequency nears the resonant frequency of individual blades. Life expectancy approaches a few seconds if that happens!)

Howard

 

[duncan webster 1Participant @duncanwebster1]

I’ve just found out that Sulzer engines used in many Brush built BR locos were 2 strokes, whereas the EE engines were 4 strokes.

To answer my own question above, having thought about it, when a 2 stroke cylinder opens, the pistons are near bottom dead centre, so a large void is suddenly.connected to the inlet plenum and quickly takes a large gulp of air, but the piston soon covers up the port. With a 4 stroke, the inlet valve opens with the piston approaching top dead centre, so a much smaller void, and the valve stays open until bottom centre, so a much more smooth suck, and with a 4 cylinder engine as one inlet valve shuts another opens, so much steadier than a 4 cyl 2 stroke

[Anthony KendallParticipant @anthonykendall53479]

“I’ve just found out that Sulzer engines used in many Brush built BR locos were 2 strokes, whereas the EE engines were 4 strokes.”
Sorry Duncan, the BR Sulzer powered Brush locos were 4 stroke – mostly the 12LDA-28 or derivatives of it.
I can see why you thought that, some of the stuff available about some of the early Sulzer engines talks about their 2 stroke diesels.
That said, it should not detract from your argument.
https://youtu.be/Y_029UD09Zc?si=YRUmincyJFUMwsZA

[HopperParticipant @hopper]

But on a two stroke the incoming air faces resistance from the high pressure gasses of combustion still in the cylinder. And the incoming air has to do the work of pushing said gasses out through the exhaust port or valve, and on down the exhaust pipe, facing more resistance. So not exactly firing into a void.

But a two stroke does this every revolution whereas a four stroke does it every two revolutions, requiring less air at the same rpm. Theoretically half as much but in practice maybe a bit more due to better cylinder filling?

[Howard LewisParticipant @howardlewis46836]

As Hopper says, the ingoing air is resisted by any pressure within the cylinder, hence the need fpr scavenging pressure.

In the simple loop scavenge two strokes used in the old two strokes, such as the Villers, or NSU Quickly engines, and simple model aircraft “diesel” engines, this is achievd by admitting the air into the crankcase and the downcoming piston acts to compress the air until the inlet port opens. In this way, the ingoing charge is under pressure, in the same way that it is in blower scavenged engines, such as the Commer TS3, or the GM uniflow engines (V71, V92 and the EMD loco engines).

In some instances the scavenge effect can be increased by careful tuning of the exhaust sytem, so that a low pressure exists at the exhaust mport when the inlet port opens. (Kadenacy)

Speedway and racing bikes tend to use this effect, (Enhanced by a megaphone at the end of the exhaust pipe), but have to be kept “On the meg”.

But this only works at certain speeds, at half or double the speed, the pulses in the exhaust system might produce a high pressure at the exhaust port!

Loop scavenge two strokes often need a little back pressure towards the end of the inlet period to prevent loss of charge . (Can also be a problem with exotic valve timing on four strokes.  But the blow down effect can be used on high output super or turbocharged diesel engines as a means of internal cooling)

Howard

 

[Anthony Kendall]

On Anthony Kendall Said:

“I’ve just found out that Sulzer engines used in many Brush built BR locos were 2 strokes, whereas the EE engines were 4 strokes.”
Sorry Duncan, the BR Sulzer powered Brush locos were 4 stroke – mostly the 12LDA-28 or derivatives of it.
I can see why you thought that, some of the stuff available about some of the early Sulzer engines talks about their 2 stroke diesels.
That said, it should not detract from your argument.
https://youtu.be/Y_029UD09Zc?si=YRUmincyJFUMwsZA

Well that just shows how the interweb can lead you astray.

Can I broaden this into diesel electric loco control? Chap in our club has built a petrol electric loco. The generator field current control is tied to the throttle control. I think this is wrong, as it takes no account of the back emf from the motors as the loco speeds up. Can anyone tell me how full size are governed/controlled? Something I got off the interweb suggests that the driver handle sets the control point for engine speed, and the control gear automatically adjusts field current to keep the throttle (or fuel rack in the case of a diesel) wide open, but another source says engine speed has only 2 settings, idle and run,.

[duncan webster 1Participant @duncanwebster1]

Should point out I made the electronic gubbins for him.

[duncan webster 1]

On duncan webster 1 Said:

On Anthony Kendall Said:

…

… Can anyone tell me how full size are governed/controlled? Something I got off the interweb suggests that the driver handle sets the control point for engine speed, and the control gear automatically adjusts field current to keep the throttle (or fuel rack in the case of a diesel) wide open, but another source says engine speed has only 2 settings, idle and run,.

That’s my understanding.  Electric generators like to be run at a high constant RPM – no slowing down or speeding up.  Therefore the prime mover is governed to run continually at a set high speed.  When the electric generator is unloaded, it doesn’t take much fuel to maintain high RPM, so the prime mover runs efficiently most of the time.

Does become wasteful if the generator is only lightly loaded for significant amounts of time, as when a train waits at a big station for passengers to sort themselves out.  Then, switching the prime mover to idle would save fuel and make less smoke!  The generator doesn’t care that the idle speed is too low because it’s not loaded at all when the train is stopped.  But the prime-mover has to be revved up before the train can start.

Train speed is controlled by altering the current flowing through the DC motor field coils.  Accelerating increases the load on the generator, which in turn loads the prime mover causing its rpm to drop.  However, reduced rpm is detected by the governor, which automatically opens the throttle to compensate by increasing power output.

In basic form quite simple, though no doubt full-size systems improve performance and efficiency with other tricks.

Duncan’s example is not so basic!  It appears that the drivers speed control is linked to the motor AND the throttle of the petrol engine and there’s no governor!  As this is a semi-automatic system, I’d expect it to be a bit of a bother to set-up and operate.   The builder has to find and maintain a sensible relationship between the electric system and the IC engine such that the two cooperate well enough to pull trains.   And maybe the driver has to avoid stalling and wild revving by learning to drive the thing!

Possibly a job for an Arduino.  In a system where the driver speed control only works the petrol throttle:

• measure the RPM of the petrol engine.
• IF petrol RPM > idle_speed THEN apply current to electric motor
• current applied to the electric motor is proportional to RPM above idle_speed

Dave

[duncan webster 1Participant @duncanwebster1]

It’s more complicated than that. The electrical side is controlled by the current supplied to the field of the main alternator. This is supplied by a 12v battery, which is itself charged by a second little alternator. At least this voltage is independant of engine speed.

Let’s suppose the engine is governed at 3000 rpm and the field current is set high enough to need the throttle wide open. To get more power we need more engine revs, but the throttle is wide open, so what we have to do is reduce the field current, which is counter-intuitive, and increase the governed speed so the engine will speed up. Increasing the generator speed will increase the power required from the engine, depending on the interdeoendencies it might be possible to put the field current back up.

Now let’s think what happens when we get to a hill. Engine as before running at its governed speed with the throttle wide open. The loco will slow down, which means the back emf from the motors will reduce, which means the generator current and power   requirement will go up, which will slow the engine, thus engine power will go down when we want it to go up. What we should do is reduce the field again.

With an intelligent driver and 2 controls (governed speed and field) I think it’s a lot easier, but he wants one stick control.

[ChrisLHParticipant @chrislh]

If it’s of any interest, the way it was done 60 years ago in full size (EE, Brush, etc) using DC traction motors and generators was that the driver’s handle controlled engine power. The combination of engine speed and torque for each handle position was built in and followed a roughly minimum engine fuel consumption line. The engine governor looked for a particular speed and load and adjusted generator load to achieve this. The generator, amongst other windings, had a seperately excited field winding which was modified by a series of variable resistors under governor control to achieve the required load. However the traction motors supplied by the generator required different supply voltages depending on their speed which of course was fixed by the vehicle speed. If the loco and train encountered a rising gradient and slowed as a result, the motors would take more current which would increase the load on the generator and finally slow the engine. The system would then, with no intervention by the driver, adjust itself once again to have the engine speed and load at the values dictated by the driver’s handle position. No doubt not how it’s done these days.

[duncan webster 1Participant @duncanwebster1]

Thanks, that’s similar to what I was thinking, but even more complicated. I think I could do that with an Arduino, but it’s not my loco, and I’ve got enough projects on the go for the foreseeable future

[ChrisLH]

On ChrisLH Said:

If it’s of any interest, …

Very much of interest, thanks!

 

No doubt not how it’s done these days.

Maybe, maybe not.   Often as not the same principles apply, but improved by applying modern technology.

Electric trains are likely very different today, because semi-conductors can now handle high-power AC.  AC out-performs DC, but not easy to get AC to deliver.

In the past, high-power speed control pretty much demanded brushed DC motors because they could be controlled by big resistors and hefty mechanical switches.   Unable to provide the best performance or efficiency, but DC systems were a reasonable and above all practical solution.   This came to an end when high-power AC also became practical, providing better performance, and opening the door to sophisticated computer control.  I don’t have any details!

Dave

[Howard LewisParticipant @howardlewis46836]

If the engine has a governor, the driver’s “Throttle lever”position will determine the speed set.  The governor will adjust the fuel level, (Throttle opening or rack position) to produce that selected speed.

(Even a simple lawn mower has a crude pneumatic governor which adjusts the engine speed to match the load on the governor spring tensioned by the “throttle lever”. The speed is sensed by a flap in the cooling air supplied by the cooling fan, and the fan is directly connected to the throttle in the carburettor)

As I understand it, for most traction purposes, the engine will be fixed speed unit, possibly with an idle setting, and the electrical control system will adjust the field supply to deliver the power selected by the driver.

The locomotive speed will be determined by the power available.

Analogous to a 50 or 60 Hz alternator supplying a load, which might vary. Being AC, the governing will be tight, (or even isochronous, if the unit is required to parallel with other sets)

Howard

[noel shelleyParticipant @noelshelley55608]

An interesting point is that 2 AC generator once in phase will pull together. To get them in phase a bulb could be wired between them and the bulb would pulse, as the setting was close this pulsing would get slower, when the bulb was out, due to no difference then you through the switch to link them. Never tried it ! Does it work in practice ? Noel

[ChrisLHParticipant @chrislh]

Howard, as mentioned above my information is 60 years old but back then the diesel engines were certainly not constant speed, varied from about 400 idle to 850 max. The idea was that the driver selected a power output using his handle and the engine responded with appropriate values of speed and fuel rack position (max efficiency line). The electrical system then sorted irself out as outlined above.

[duncan webster 1Participant @duncanwebster1]

Yes, you have to have the 2 going at pretty much the same speed to start with. If you just connect 2 alternator out of phase it can be pretty spectacular I’m told.

When I worked for a short time on steam turbines we used Woodward governors on some jobs. These were reputed to be able to synchronise an alternator set to its brethren without going through the light bulb routine. Things of awesome wonder. Probably superceded by some electronic whizz bang now

[Howard LewisParticipant @howardlewis46836]

The normal idea when paralleling two or more alternator sets, was that one was the master and was isochronous governed, while the other(s) were slave and droop governed.

Ideally, they were all in sync and no circulating currents happened (Lamp out).

Parallelling with an infinite bus, (mains) was an interesting activity.

Some multiple sets incorporated relays which would isolate a slave set, if the circulating currents were too large.

Many years ago, at Southdown Central Works, there were three 1000 rpm alternators, powered by Gardner 5LWs which had been removed from buses and the governor arms extended to tighten the governing so that they would hold a constant speed at 1,000 rpm. The idea was to save money, during winter,  by powering the lights from the alternators rather than the mains.

The governing was too tight to be stable and damping was minimal, so the attendant spent all winter afternoons frantically trying to keep the lamps out; and failing. They just flickered on and off in sequence while he juggled the controls.

In another completely different context, I was unpopular with Woodward Governor Company for slugging a droop governor so that it did not try to govern out torsional vibrations!

They had spent thousands of dollars to develop a speedy response, and I was damping it heavily, to slow the response; hence the unhappiness!

An engine capable of a high instantaneous rate of acceleration could very difficult to control with a droop governor, to get a steady speed; especially if the selected speed was near to peak torque speed.

But an isochronous governor was more expensive, so droop was the first choice, where possible.

Howard

[duncan webster 1]

On duncan webster 1 Said:

On Anthony Kendall Said:

“I’ve just found out that Sulzer engines used in many Brush built BR locos were 2 strokes, whereas the EE engines were 4 strokes.”
Sorry Duncan, the BR Sulzer powered Brush locos were 4 stroke – mostly the 12LDA-28 or derivatives of it.
I can see why you thought that, some of the stuff available about some of the early Sulzer engines talks about their 2 stroke diesels.
That said, it should not detract from your argument.
https://youtu.be/Y_029UD09Zc?si=YRUmincyJFUMwsZA

Well that just shows how the interweb can lead you astray.

Can I broaden this into diesel electric loco control? Chap in our club has built a petrol electric loco. The generator field current control is tied to the throttle control. I think this is wrong, as it takes no account of the back emf from the motors as the loco speeds up. Can anyone tell me how full size are governed/controlled? Something I got off the interweb suggests that the driver handle sets the control point for engine speed, and the control gear automatically adjusts field current to keep the throttle (or fuel rack in the case of a diesel) wide open, but another source says engine speed has only 2 settings, idle and run,.

This is perfectly OK. It is the basis for the Ward-Leonard speed control system that was used on a lot of large motors before electronic controls. https://en.wikipedia.org/wiki/Ward_Leonard_control

I tried this on a small gas turbine engine with a alternator as a possible hybrid electric landrover about 20 years ago. The gastubine was fitted with a constant speed fuel control system to keep the output frequency at 400Hz. The voltage matched the motor in a electric tug I had an option to buy. As it was a brushed motor it didn’t mind AC. I got as far as sitting the the turbine on the tug and connecting to the motor. It seemed to work. The turbine throttled up as the increased load caused the RPM to throttle up. Not a racer but would make an interesting off-roader Two smaller motors, one for each axle would be better. The turbine was a Rover so it should have been eligible for Landrover competions. But my employment changed and the project was axed s I no longer had access to suitable premises to do the work.

[duncan webster 1Participant @duncanwebster1]

Small world. I worked at Lucas Aerospace for a while in the bit that made gas turbines. They made the APU for the Harrier, and had rights to the Rover turbine. I designed a gearbox to use it in an industrial application, burning cane alcohol. The aircraft chaps simply couldn’t get their heads round the fact that weight didn’t matter and had conniptions when my design turned up in the flesh. It worked tho’.

[Anthony KendallParticipant @anthonykendall53479]

“Parallelling with an infinite bus, (mains) was an interesting activity.”

Certainly is Howard.
Often used to parallel up with the mains at high power transmitting stations where the diesel capacity was a fraction of the full load and used as emergency. Not normally possible to change over totally because of the break and resulting reduced power.
Only way to test alternator was to parallel up with the mains.
All fine after syncing, paralleling up and increasing load to the full rated output.
All fine, that is, until the mains disappears momentarily and for a short period of time before the overloads work, the alternator attempts to power the grid and mains returns out of sync! Only saw it once – could best be described as an amazing moaning noise, a cough, much rattling of OCBs, followed by fans winding down and a deathly silence.

[Author]

Posts