Steam Engine Design

Walkthrough of various factors which affect steam engine performance, including compounding and crank layout.

Introduction

The purpose of this document is to examine the factors of steam engine design and how they affect performance. If you’re just looking for some example engines, subscribe to this blueprint on Steam and download it in the Main Menu > Content > Vehicles > Download.

Glossary of Terms

• Compounding: the practice of connecting two pistons in series: pistons are compounded when the output of one is connected to the input of another
  
  
• Compounding ratio: the ratio of pistons in one stage to the number of pistons in the next stage
  
  
• PPM: Power Per Material, an ingame figure calculated by dividing the power output of an engine by the material consumption
  
  
• PPV: Power Per Volume, an ingame figure calculated by dividing the power output of an engine by the number of blocks occupied. Each block is one square metre
  
  
• PPBB: Power Per Bounding Block, an alternative out-of-game method for calculating volumetric efficiency by drawing a rectangular box around the engine and dividing the power output by the volume of the box
  
  
• HUD: Heads-Up Display, the information displayed on the screen during normal play, as opposed to information contained in menus or popups

PPM and PPV

Introduction

PPM and PPV are factors common to all types of power, and the most commonly-cited statistics when comparing engines. If there were builder’s plates in FtD, they’d cite PPM and PPV (and would take up a full square metre). Fuel engines offer obvious choices for trading one for the other in the form of injectors, turbochargers, and superchargers. Steam engines can also be tuned for PPV or PPM through various design factors.

Compounding

“Compounding” is a technique for improving the PPM of a steam engine at the expense of PPV. Pistons are “compounded” when the output of one piston is connected to the input of another. This results in backpressure on the first piston, lowering output, but allows some of the exhaust steam to be used to generate power in the second piston instead of being dissipated. Compounding was used in real-life full-sized practice, and the many stages of modern turbines can be seen as a form of compounding.

Number of Stages

Compounding can be done more than once: the output of the first stage feeds the input of the second, the output of which feeds the input of the third, and so on. Increasing the number of stages increases the PPM at the expense of PPV; however, it is worth noting that, while increasing the number of stages increases the amount of plumbing - thereby increasing the indicated volume - the actual footprint occupied is essentially the same, as the increased piping almost always fits within the empty spaces in the design. That said, every block occupied by a steam pipe represents a vulnerable block which could otherwise have been occupied by a weapon system or another layer of armor. See section "PPBB" for more.

Compounding Ratio

Full disclosure: this section is mostly a personal fad of the author, and most players may safely skip to the next section, content in the knowledge that their equally-sized stages are within a few percent of optimal. This section is, as the APS tooltip says, “For the real experts and those with time to kill!”

The Compounding Ratio is the ratio of pistons in one stage to the number of pistons in the next. In From the Depths, the efficiency of a steam piston is determined by the ratio of input to output pressure, called Diff. pressure in the ingame tooltip. When pistons are compounded, the Diff. pressure of each stage is determined by the compounding ratio:




For many applications, the optimal compounding ratio is not the commonly-recommended 1:1, but the slightly higher 5:4. However, while 5:4 offers slightly improved PPM across the board, the benefits are small. In some cases, due to the way PPV is calculated, the additional plumbing required more than outweighs the increase in power, and for some layouts the additional clutter may interfere with the placement of boilers, auxiliaries, or other systems. See Appendix C for graphs.

Pistons per Crank




Each crank can connect to multiple pistons: four for small steam using non-cased cranks, three otherwise. Engines with more pistons per crank will have slightly higher efficiency, but may be difficult to fit depending on available space. See Appendix C for graphs.

Wheels

Adding wheels to an engine can increase PPM at the cost of PPV and PPBB, with diminishing returns. Wheels may be necessary in order to use the full output of an engine: more on this in the Troubleshooting section.

PPBB

Although the ingame UI gives PPV, some players use another method of calculating volumetric efficiency known as Power Per Bounding Box. This method works by drawing a rectangular box around the engine which extends to the widest point in each dimension, and using the volume of the box instead of the sum of the volumes of each individual engine component.

This can give a more accurate measurement in some cases because some engines have many “hollow” blocks which are not occupied by engine components - and therefore do not count against PPV - but which are nonetheless unusable for other purposes.


An engine with no empty spaces will have a PPBB identical to its PPV; an engine which is mostly empty space will have a high PPV but a low PPBB. It is possible to minimize the difference between the two by filling empty spaces with useful parts such as boilers, though, since smaller components have lower volumetric efficiency, this may come at the cost of much lower PPV and thus more blocks occupied. As always, the designer must strike a balance while considering the impact on other parts of the craft. See Appendix C for graphs.

Kinetic Efficiency

Not all power generated by a steam piston is usable engine power. Some proportion of the kinetic energy (a unit unique to steam power which is used to represent the energy used by the pistons) is lost. The higher the proportion lost, the worse the engine will respond to changing loads; while steam engines are generally not a good choice for loads which are variable by design, in practice any load can be variable if the relevant parts are destroyed in the middle of a battle.

Compounding invariably results in lower kinetic efficiency, though it is balanced out by higher PPM. Engines with fewer crank components have lower kinetic efficiency, so designs which minimize the number of crank components will perform better. In practice, this means using as few gearboxes and as many pistons per crank as possible.

The total kinetic energy in a steam system can be seen in the Q menu for the system, under the “Current system output” tab. Piston steam to kinetic energy/s is the gross output of the pistons, before loss. Crankshaft loss/s is the total loss for the system. Gearbox kinetic energy to power is the difference of those two values, and is equal to the engine power output.

The Kinetic Efficiency is not shown ingame, but is calculated by dividing the Gearbox kinetic energy to power by the Piston steam to kinetic energy/s. This value decreased slightly as the number of stages increased.

Reference Values

By far the most common question related to steam engines is “Do these numbers look okay?”. To that end, below are some tables which should give a general idea of expected performance. Note that these are not optimal figures by any means, and that different designs may trade performance in one area for another; however, they should give a good baseline for a basic, functional engine. If your own performance figures are much lower than those given, proceed to the next section: Troubleshooting.

Troubleshooting

Steam engines do not give accurate readings unless they are tested at full power. Two conditions must be met:

• The boilers must be at full pressure
  
• The engine must be fully loaded

Pressure



Steam systems have a maximum pressure of 10, and perform very poorly below this pressure. When a boiler is connected to a system, the pressure will slowly increase until it reaches 10 or the boilers can no longer keep up with steam demand. If the pressure does not reach 10, more boilers must be added in order for the system to perform adequately. Note that the pressure is best measured at the boilers themselves; in compounding engines, the pressure at any other point in the system will by design be lower than the boiler pressure.

When the pressure reaches 10, the Material/sec and Steam/sec readout will quickly drop, usually followed by a slow decrease over several seconds depending on the size of the system. Once those readouts reach a fixed value, the engine is said to be stable, and is ready to give accurate performance figures.

Load



A steam engine operating under partial load will spin at a high RPM, will consume far less steam than expected for its size, and will have an indicated power much higher than what it is actually capable of producing. In addition, the PPM will be very low.

An easy way to provide a load for testing and design purposes is to place a Signal Jammer ECM (found under the Defence tab of the build menu) connected to an AI mainframe, either directly or through AI Connectors. By placing the cursor on the ECM and pressing Q, it is possible to adjust the power demand between 0 and 20,000 engine power. Systems generating more than 20,000 power will require the placement of additional ECM components (be sure to set the power demand on each one!).

To test a system’s electrical output, the above steps can be combined with the addition of an electric motor and battery bank. The motor will convert electric charge to engine power, which will then be drained by the ECM. The electric motor can be set to high priority in the Power menu, accessed by pressing “v” and navigating to the Power tab. This will ensure that the power will be routed through the generators first before being used, rather than being consumed directly from the power generated by the pistons.

Kinetic Energy Capacity



Each component in a crankshaft assembly is capable of holding a certain maximum amount of kinetic energy. In power-dense configurations, such as a single-stage engine with three or four pistons per crank, the amount of energy being generated per crank component is higher than the cranks alone can absorb. Such an engine will have lower than expected power output.

The kinetic energy capacity of a crankshaft can be increased through the addition of wheels (found under the gearbox menu).

Appendices

Appendix A - Data

The graphs in Appendix C below were made using data gathered by the author. The full tables can be found here[docs.google.com].

Appendix B - Links

• A steam engine demonstrator fortress, which gives examples of a few different types of engine, can be got here
  
• The steam engine designer used for these tests is available here[docs.google.com]. Note that the link is read-only, but a copy can be made with File > Make a copy

Appendix C - Graphs

[Elephant Shrew]

Thanks for the guide.

[Zlorfik [CH/BY]]

7. Can I accurately predict/measure the maximum amount of stable/constant power production by a steam engine system?
8. Why not make a turbine & electrical engine instead? Is there a guide from you on that? :]

[Zlorfik [CH/BY]]

5. What are good ways to use exhaust steam without influencing the efficiency of the steam engine? A good chunk of steam is always wasted and that is generally also just a waste of materials. One can not use turbines, since they run on pressure and pressure should be 0.1 for the last stage of pistons output. I was thinking steam jets?
6. The gearbox allows me to set a maximum power produced value as well as maximum rotation speeds. From own experiments, any rotation speed lower than 100% seems to waste energy. What about maximum power production? According to the gearbox tooltip, a system can transform 1 kinetic energy to 3 engine power. So if my crank assembly has a total kinetic energy of 2100, a maximum loss of 600, what maximum power would I set? My guess would be max power = 6300 and make sure I can actually supply the necessary 2700 kinetic energy/s via pistons.

[Zlorfik [CH/BY]]

1. How many pistons should act on a single crank? Is the answer simply "depends on the number of wheels"? Or should it be "as many as possible"?
2. You said that gearboxes and cranks increase the inefficiency of systems. Is it therefore never a good idea to have 2nd and subsequent stages work on entirely separate crank shafts?
3. From web searches and own tests I found that from the max 5000 steam that acts as pressure on a piston, almost 50% of it can be used as steam consumption, 20% being used up and 30% output for the later stages. How can I control the amount of steam used by pistons?
4. If I have 3 pistons in stage 1, they can use up to almost 7500 steam under optimal conditions, which is exactly what a single medium boiler produces. However, I have never reached ~2500 steam consumption of pistons in such a system (more in the region of 2000). Why is that?

[Zlorfik [CH/BY]]

Great Guide! I love steam engines (and electrical engines) and I have been struggling to figure out all the details. I generally want to go for the most efficient engine possible, since in the campaign I have found that all my ship designs suffered from such high material consumption that I basically lost the games due to running out of materials from upkeep alone.

Let's say Volume is not a concern. So from your guide, I assume that I would build 4 stage engines, since this has the best PPM value. (More than 4 stages should increase PPM even more I assume). I would also add wheels to the crank shaft in order to increase PPM even more. This raises several questions for me:

[Ekental]

So the guide talks about 5:4:4 ratios but the linked spreadsheet calculator is for a 5:4 ratio per stage.

Doing a little bit of testing on small 3x expansion piston systems I got:

Piston Systems - 3x expansion
2:5:5 - 545 ppm, 3880p
3:4:5 - 566.4 ppm, 5007p
4:4:4 - 577.4 ppm, 5869p
5:4:3 - 580.3 ppm, 5960p
5:3:4 - 577.4 ppm, 5742p

5:4:4 - 594.8 ppm, 6569p

5:5:4 - 596.5 ppm, 7055p
6:4:4 - 593.3 ppm, 7073p
6:5:4 - 595.9 ppm, 7651p

We see 5:4:4 turns out not quite the same as the data here 602.7ppm v 594.8 ppm, but it's close.
The spreadsheet advocates 6:5:4 as that's the closest to 5:4 per stage and that is very slightly better than 5:4:4 but :shrug:

Anyways, is it 5:4 per stage or 5:4n stages?

[Taverius]

Fun detail I just found, for triple expansion steam 5:4:3 is ever so slightly more efficient than 5:4:4, but with a decimal difference only, while the 5:4:4 has considerably more power.

Comparing Medium 2-Bank Triple-Expansion engines (1 wheel each):
- 5:4:3 14200 Power 691.4 PPM
- 5:4:4 15261 Power 691 PPM

[Polaras]

@Rnev Depends on the design and the load. Adding fly-wheels and generators increases kinetic loss on the shaft (decrease kinetic efficiency) But because the shaft is heavier (max total kinetic energy is higher), it should spin slower and the total kinetic loss will be lower resulting in a net increase in power at maximum load. This is what I would expect from a big optimized engine.

I was able to get power to drop from adding a 2nd generator on a 2 stage, 2 piston, large engine because in that setup, there are only 6 sources of kinetic energy loss, and the engine was only running at 10% max RPM (lots of overhead for kinetic energy storage). Even on a 6cylinder 2 crank 1 stage Large engine, having a second generator resulted in a loss of total power, PPM, and PPV.

Generally speaking, you never need a second generator unless you are running a small class engine with a very large expected energy load compared to power load. ALWAYS test under expected load. It depends.

[Rnev]

Great guide, interesting read too.
any chance someone knows why when i add belt-generator and fly-wheels - energy generation from the engine actually goes *down*?

[arognar]

how do i deal with the noise?