Fleet Boat Speed Control 1: Fairbanks-Morse 38D 8-1/8 9-cyl

[Rosencrantz]

Last edited: 21th August, 2015 Updated 2/2.1. and added 2.2.

1. Prologue

During the last few days I have been researching speed control in the boats equipped with Fairbanks-Morse 38D 8-1/8 9-cylinder engines. Purpose of this and following topics is threefolded:

1. "Which are the correct speeds for different engine order telegram settings" is one of the most frequent question among submarine simulator users, modders, and other non-professional submarine enthusiasts. Even if there is heap of true information availible today, there is still many questions left, and even more simplified, inaccurate, insufficient or even wrong information availible in the internet, books and other publications. Aim of this and following topics is to provide in the future accurate, reliable and as completed as possible information about the Fleet Boat speed control, based on original US Navy documents.

2. Few years ago I started a project called "Fleet Boat Engine Hour Mod", discussion here:
http://www.subsim.com/radioroom//showthread.php?t=180989
Project was then hindered and later abandoned due to computer loss and incoming real life tasks. However, I'm now slowly getting back in the saddle. My intention is restart previous project, because some progress was achieved previously.

3. If possible, bring an end to "yes, no, maybe or maybe not" discussions regarding speed control in Fleet Boats. I have found topic somewhat more complicated and diversified than often seems to be understood.


"How to?"

a) All information provided by me in this and following topics will be based on real life US Navy documents, especially Submarine War Patrol Reports 1941 - 1945, not forgetting fleet boat and equipment operation manuals (NavPers), other technical and operative documents, and other possible sources considered both reliable and accurate enough by me.
b) Excerpts short enough will be presented here, otherwise link or reference information will be shared.
c) Original post will be edited by me as more data is found. Date of the last update will be presented on top. As I do this in my freetime, updates won't be frequent. Active and passive periods will vary.
d) Screening boats and war patrol reports is based on type of engine. At the moment I'm researching boats with Fairbanks-Morse 38D 8-1/8 9-cyl. Boats with other engines will follow later.


"What to expect"

Along new knowledge also new question must be expected to be raised during this research. However, I have found speed control question quite interesting and I have already learned something new myself.

Mistakes in grammer and spelling will be present in texts written by me, as I'm not a native speaker. However, I hope data, excerpts and links provided and research done will outweight my personal shortcomings.


2. Narrative

2.1. Even if we are talking about submarines, I want to start from totally different kind of a warship. Different pretty much in all aspects like size, speed, type of missions she was build for and tasks she completed. We'll start from USS Enterprise, CV-6. We'll start from her, because I think she might be able to teach us something new about speed control in US Navy ships during WWII.

From a dedicated site we can find parts of the Ship's Operations Manual (http://www.cv6.org/ship/logs/default.htm). Chapter 9 Ship Control is very interesting regarding our topic as Section 904 describes carrier's speed control. There we can see how for example an order "Standard speed" actually did not mean some particular speed in knots, but quite opposite. This order was situation-dependable to a great extent. While in one situation it could mean as slow propeller rpm as 54 (revolutions for 6 knots) in another situation it could mean rpm 309 (revolutions for 33 knots). Full speed was always 5 knots and flank speed 10 knots faster than current Standard speed. (Probably those orders were not used if stadard speed was sometimes set at 33 knots..?) For additional information about speed control in Enterprise see also Sections 905 - 908.

Previous example raises a question if speed control in fleet boats was organized and handled in a similar manner as in flat tops. Quite often it has been understood that for example "Standard speed" meant always 13 - 14 knots speed in fleet boats. This is actually not true. Standard Submarine Phraseology originally issued by ComSubLant is a booklet describing the voice procedures and phraseology recommended "as standard for all submarine interior communications". I have a copy published by Periscopefilm. On the page 19 "Engine and battery combinations" there is one example of setting stadard speed: "Standard speed will be one eight knots." USS Triton's (SS-201) second War Patrol Report offers another example:

Jan 25, 1942: "Standard speed was set at 13 knots but could not be maintained at all times due to rough head seas..."

2.2. Speed range for one engine on the line

Quote from "The Fleet Type Submarine", NavPers16160: "For surfaced operation... the motors develop power ranging from 20 hp to 2,700 hp per propeller shaft at speeds ranging from approximately 67 rpm to 282 rpm."

Determining the slowest possible speed for a Gato/Balao/Tench and maybe also for a Tambor-class fleet boat is pretty easy. While there is no straight relationship between engine rpm and speed the latter is close tied together with propeller rpm. Knowing the top speed of a Gato-class boat was 21 knots at 282 rpm, 67 rpm equals with speed of 4,99 knots. Tambors were a bit slower, probable slowest speed on the surface being some 4,85 knots in a fair weather and the calm seas. USS Trout (SS-202), Report of the sixth War Patrol: "October 31, 1942: Patrolling on east-west courses... at 4 knots". So far4 knots on the surface has been the slowest speed I have run into while reading the reports.

To find out top speed or speed for 80-90 combination are both much more difficult. Raw calculations taking account the shape of the hull (displacement hull), lenght of the waterline, displacement, and both engine and motor limitations and estimated mechanical efficiency of some 84 % suggests that Tambor-class top speed with one engine on the line was more than 12 knots but less than 13 knots. Simplified example:

1600 hp x 0,844 = 1350 hp (shaft)
3 304 000 lbs/1350 hp = 2447 lbs/hp (shaft)
2447 x 1.19 = 2912 lbs/hp (engine)
2912 = (10.665/x)3 => 0.75
Waterline length 300.85 ft (Tambor-class)
x/square root 300.85 = 0.75 => 17.35x0.75 = 13 knots.

1280 hp x 0,844 = 1080 hp (shaft, 80% load 90 % rpm)
3 304 000 lbs/1080 = 3059 lbs/hp (shaft)
3059 x 1.19 = 3640 lbs/hp (engine)
3640 = (10.665/x)3 => 0.69
x/square root 300,85 = 0.69 => 17.35x0.69 = 12 knots.

Archerfish's first War Patrol Report gives some support to previous calculations:

"Jan 2 - 4, 1944: ...turns for 11.5 knots with one main engine..."

Previous remark covers two days enroute so it is not likely that the crew took all out of the one single engine on the line but engine was running somewhat lower rpm. It is even possible 80-90 combination was used.


To be continued.


-RC-

[Sailor Steve]

Sounds like a cool project. Here's hoping it works out. :sunny:

[TorpX]

I'll certainly be tuning in to see what you come up with.

[Rockin Robbins]

Ditto. Past attempts came smack up against unrealistic game handling of fuel, speed and duration. If you can crack the puzzle, I'm a fan!

[aanker]

Thank you for your renewed interest in this project. I liked it in 2011 and I like it now. Hopefully the SH4.exe will cooperate.

[Rosencrantz]

Hello, Steve, TorpX, Rockin, aanker!

Thx for the positive comments! I'm quite excited about some findings so far. (See 2.1. in the orginal post).

What comes to Fleet Boat Engine Hour Mod, I have to say I lost all data with my previous computer. Also - even if some progress was achieved - results were not totally satisfactory. After doing some raw calculations based on data found from The Fleet Type Submarine Main Propulsion Diesels Manual, I have a feeling that for example at 80-90 combination with one engine on the line later fleet boats (from Tambors onwards) were much faster than typical 6-7 knots used in most simulations and mods.

Also at the moment research comes first, mod second.


Greetings,

-RC-

[CaptBones]

Just a couple of comments to help point you in the right direction, I hope. Which is to say I hope this isn't too much "preaching" and doesn’t discourage you from trying to find a better approximation of real engine/ship speed control simulation in the game. I guess I'd also say that using the steam plant operating guide for CV-6 as an indicator of the concept or "philosophy" of plant operations for the "Fleet" submarine is not going to really help at all. That's not even an "apples to oranges" comparison, it's more like a "bicycles to pomegranates" comparison.:hmm2:

OK then, first; remember that the US “Fleet” boats had diesel-electric propulsion plants. Using diesel engine HP and speed (engine RPM) in calculations concerning ship propulsion shaft HP/RPM and ship speed is incorrect. The diesel generators operated on the basis of load-speed settings (e.g., 80-90) that were very different from the shaft RPM/ship speed called for on the Engine Order Telegraph (EOT) and Revolution Indicator. Note that only the EOT is modeled in our submarine simulation games.

The load-speed setting mentioned refers only to the engine-generator sets; as you know, 80-90 is 80% load on the generator (and hence on the engine) at 90% of engine full speed rating (720rpm for the FM OP engines). Bear in mind the main generators were all rated 1100kW at 415V/2650amp and the OP was rated 1600BHP at 720rpm. Thus, 80-90 means running each engine-generator set operating on propulsion at 880kW/650rpm (yes, it was “rounded off” a bit). BTW, you’ll see that the arithmetic to convert kW to HP doesn’t come out exactly right. That’s mostly because the engine rating is in BHP (Brake Horse Power), which is different from the output HP at the engine coupling to the generator. A certain amount of developed HP is absorbed by the engine-driven components and accessories (scavenging blower, water pumps, lube oil pump, fuel oil supply pump, fuel injection pumps, etc.).

Next, add to that the complexities of a DC propulsion plant with several main engine-generators and multiple main motors on separate propulsion shafts. There were two motors on each of the two shafts, driving the main shaft through a reduction gear. That results in “multiple armatures” in a series-loop configuration, a situation that is far beyond the ability of the game engine to even roughly simulate. The utility and flexibility of four main generator armatures that can be connected in different combinations to four main motor armatures is, well, "beyond the scope" of any commercial PC game engine.

Then, in order to accurately model those real-world propulsion plants, you really need to consider Propeller Law and start with a Propeller Curve and multiple engine power output curves on the same graph. As I'm sure you realize, the correlation between propeller RPM and propeller HP (and therefore ship speed) is not a straight-line relation; 282rpm/21kts is not equal to 67rpm/4.99kts. Without an actual propeller curve for the “Fleet” boats, we could just use an “ideal” propeller curve. If you assume that 100% power represents two engine-generator sets on one propeller shaft, the omission of one engine-generator set would mean 50% available power on that shaft, if the remaining engine-generator set could operate at full rated speed. With only one generator operating on propulsion on that shaft at full speed, the total system voltage in a series-loop circuit would be 50%, which would reduce the motor speed to about 50%, since motor speed is approximately proportional to the voltage generated. Using the propeller curve, you would see that 50% shaft speed is equal to 20% of the propeller horsepower demand, which is not equal to the power output available from the engine-generator set on propulsion on that shaft. But, by weakening the motor field, the main motor speed can be readjusted upward along the propeller curve to about 80% shaft speed, where the full capacity of that engine-generator can be absorbed.

Now, since we have a more flexible, and more complicated, plant you could also assume that the Propeller Curve is plotted to represent four engine-generator units. In order to simplify that case to something manageable and easily understandable, we have to assume that both propeller shafts are operated at the same shaft speed at all times. Even so, the following simplified explanation is not entirely correct.:03:

Nevertheless, to illustrate the flexibility of the “Fleet” boat propulsion plant, let’s consider the omission of one generator set on battery charge, which would mean 75% available propulsion power, again with the remaining three units capable of operating at full rated speed. However, with only three generators on propulsion (at full rated speed), only 75% of total system voltage would be generated, assuming a series-loop circuit with three main generators connected to all four main motors. This is also only possible if the design of the generators and motors is correlated so that the sum of the generator voltage rises is equal to the sum of the motor voltage drops. Thus, motor speed would be limited to 75%, since motor speed is approximately proportional to the voltage generated. Using the propeller curve, you would see that 75% shaft speed is equal to 43% of the propeller horsepower demand, which is not equal to the power output available from the three engines running on propulsion. But, by weakening the motor field, the main motor speed can be readjusted upward along the propeller curve to about 91% shaft speed, where the full capacity of the three engine-generators can be absorbed.

Finally, the whole matter is really made more difficult to readily model by the application of the 90% normal speed limitation on the generator sets. This artificiality can be observed in the generator rating; they were designed for a 450 volt output at the terminals (with nominal insulation resistance for 460V) and rated at 415 volts. The FM OPs could, and still can and do today, run constantly at 720rpm for thousands of hours without excessive wear. When “in extremis”, the MoMacs and Enginemen in “Fleet” boats could move, or even remove, the rack-stops and push the OPs to nearly 800rpm. Commercial variants of those very same engines run at 900rpm.

OK…so does that help or hinder your project?:06:

[fithah4]

:o:huh::o:huh:........WOW!!

Very detailed information on this. A graph would diffently help in viewing info with each example listed as a color to compare to each other, through the speed and hp / bhp ranges.

Fith

[Sailor Steve]

As I'm sure you realize, the correlation between propeller RPM and propeller HP (and therefore ship speed) is not a straight-line relation; 282rpm/21kts is not equal to 67rpm/4.99kts.
Not just for propellers. I was just reading a similar set of equations concerning race cars. One of the big problems is drag. Hydrodynamic drag increases as the cube of the velocity, meaning that it takes 9,261 times as much power to maintain 21 knots as it does to maintain 1 knot. it only takes 1,331 times as much power to maintain 11 knots. That gives almost 7 times as much power needed to maintain full speed as it takes to maintain half speed.

This is true no matter what type of propulsion is being used.

[CaptBones]

fithah4...Thanks.:salute:

Yes, graphs would certainly be helpful, several actually. They do exist and could be extracted/copied from various sources. I'm sure a good search engine would find quite a few useful examples of Propeller Curves with corresponding engine speed/hp curves for a variety of propulsion plant types. In the end, though, I think that is probably far beyond the potential, or even the possibility, of producing a technically complete and correct model to incorporate as a MOD in our beloved subsims. "Tweaking" certain files, as was done in MODs such as "Improved Ship Physics", is probably as close as the game engine itself will allow.

Steve...hello again and yes, you're correct.:up:

Hydrodynamic drag is the biggest factor, but it's not the only one and it is also not "consistent". The drag factor typically used in hull/propulsion plant design is actually the speed-to-length ratio, or speed in knots divided by the square root of waterline length. As that ratio increases, the propeller horsepower required may increase as the 3.5 to 4th power of the speed in knots. Other factors that enter into the problem on a day-to-day basis include bottom fouling, actual displacement vs design displacement, trim, list, sea conditions, head winds, etc.

Also, I referred to the "ideal" propeller curve, which is a cube-law curve. Although we've been coming closer in recent years, there still is no "ideal" propeller; funny things like "slip" keep us from getting there. So, the relationship between a real propeller's rpm and its horsepower to drive the ship, is not an ideal curve. The "ideal" curve (Propeller Law) is used in design work, but the performance characteristics of the ship's propulsion plant aren't even known for certain until they are finally determined at Sea Trials.

AND...if you want to add one more complication directly related to diesel engines, there's BMEP (Brake Mean Effective Pressure). But, I've probably added far too much "confusion" to the basic discussion as it is, so we'll forego that (for now at least).:arrgh!: