Close

Project Log 114: "Final" Project Log.24

A project log for DIY Mech or Exoskeleton suit.

Mechs are not viable, nor cheap, so I will try to design and build one alone anyway.

fulanodetailFulanoDetail 18 hours ago0 Comments

A Closer Look: Main bearing Pt. 1

Mod-01 Lec-21 Centrifugal Compressor Part I 

Compressors - Turbine Engines: A Closer Look 

Introduction to Axial Compressor Design 

Is it Possible to 3D Print WORKING AXIAL COMPRESSOR? - (Testing different blade designs) 

Two POWERFUL 3D Printed Axial Compressors - feat. Uniformation GKTwo 

It Took Me 3 Months To Get This Working - Two-Stage Centrifugal Compressor 100% 3D Printed 

I 3D Printed a Compressor that ACTUALLY WORKS (but you've never heard of this design) 

3D Printed Two-Stage Compound Turbo Compressor 

The old GTM160 micro jet engine in service. Complete disassembly of the engine, repair and assembly. 

Turbo Compressor designs - Discussed 

DIY 3D Printed Vacuum Impellers 

Comparing Turbine Rotors 

Full Moon Afterburner Run 

Final Report Thermodynamic Demonstration Unit

Original Article Design of a two spool contra-rotating turbine for a turbo-fan engine

Analysis of Flow through Vaneless Contra-Rotating Turbine of Jet Propulsion Engine

(PDF) Counter Rotating Turbine Engine Compressor Blade System | Alex Gardner - Academia.edu    

I was wondering:

If I’m going through the trouble of doing all of this, then why not go back to that idea of making a turbine engine that works like a Firework Spinning Wheel.

Essentially, instead of compressing the air to go through a combustion chamber and the expanding gas going through a turbine, I would use the expanding gas from the combustion chamber to rotate the entire thing.

I mean, if the expanding gas can rotate a turbine, then why wouldn't it be able to rotate itself?

Why the combustion chambers of a turbine engine aren't used to spin the compressor like a firework wheel? - Aviation Stack Exchange 

The problems would arise from injecting fuel and/or using the heat exchanger to increase efficiency.

In the case of the fuel, you could either use a rotary feeder or a vacuum ejector connected to the combustion chamber, sucking fuel and mixing it with both the incoming air and the compressed air.

The heat exchanger would require some tubing and clever engineering, maybe making the air channels to go around the rocket nozzles both for cooling and as a heat exchanger.

https://sci-hub.ru/https://doi.org/10.2514/6.1991-3124 

According to the research it would be slightly less efficient than a conventional helicopter turboshaft, but I mean… The idea would be to run this thing with a compressor, heat exchanger and a temperature of 2000ºC.

Just now I remembered that this engine would be essentially a hero reaction turbine.

Air Powered... Propeller?

Source: Investigation of the performance and flow characteristics of two-phase reaction turbines in total flow geothermal systems - ScienceDirect 

Igniting Our Own Rocket Engines - DIY 

I built the BEST COMPRESSED AIR ENGINE (New Rotary Design) 

Source: https://www.sciencedirect.com/science/article/abs/pii/B9780123838421000081 

Maybe I could feed both air and fuel through a hollow shaft.

Now that I think about it…

I think it would be a wise decision to add a shaft seal to the fuel feel, just for precaution.

Also, the incoming air will first go through the heat exchanger surrounding the exhaust of the reaction turbine, and the same heat will be used to reform the charcoal/wood fuel into syngas to avoid clogging.

Wouldn’t spinning the entire fuel chamber with the damned engine be more practical and safer?

Besides, the fuel would be reformed and naturally maintain the solid parts at the tips and the lighter gas at the center.

You could make the fuel tank limited and only after it depletes sufficiently enough, you stop it and re-inject the fuel while the battery keeps things running.

Isn’t this just a rotary ramjet system?

Also, for some reason a lot of designs for turbineless jet engines use screw compressors without a shell. Most of them rely solely on the mechanical sealing of the screw to the static walls.

Source: Chinese scientists propose ram-rotor detonation engine for hypersonic flight | South China Morning Post 

Helical Blade Turbineless Gas Jet Engine 

I mean, bruh, just add a rotating shell to the screw itself so it doesn’t have any blades and walls to turn into powder.

One interesting thing about this type of engine:

Since they don’t rely on turbines, but rocket propulsion to rotate and/or generate thrust, they are not limited by the efficiency of turbines, nor the speed limit of blades, which is the speed of sound.

Blades must always avoid reaching the speed of sound due to drag, mechanical limitations, efficiency etc. But what happens when you don’t have to worry about these factors?

While looking at these I found a few interesting things:

Thermodynamic performance evaluation of a turbine-less jet engine integrated with solid oxide fuel cells for unmanned aerial vehicles - ScienceDirect

US20060230746A1 - Turbineless jet engine - Google Patents

Gas turbine operating parameters? - Thunder Said Energy

You could also make it double intake and double exhaust, kinda like a steam turbine.

I was wondering if it would be better to add the fuel through the exhaust (since the exhaust is doubled in the middle), so the fuel gets hot enough to be reformed.

In either way, I do think I should increase the pressure instead of working with a helium turbine engine at 1000ºC, this way you could also reduce the temperature and make nitrogen viable again.

Source: GAS-TURBINE COMPRESSORS: Understanding stall, surge – Combined Cycle Journal

Source: Influence of Operation Conditions and Ambient Temperature on Performance of Gas Turbine Power Plant Advanced Nuclear OpenAir-Brayton Cycles for Highly Efficient Power Conversion 

There is also the possibility of maintaining high temperature (1000ºC) and high pressure (200 bar/20 Mpa), increasing the power density of the system.

How to light an IR plate burner and Why ? (I was wondering about the possibility of using a similar burner for the heat-exchange system)

One weird thing, graphite’s resistivity is the lowest at 1000 kelvin (726 ºC):

Source: (PDF) Numerical Analysis of a Radiant Heat Flux Calibration System 

My Wind Generator Was Useless....Until NOW! (MPPT) 

Two plane dynamic balancing machine for model jet engine 

Any machine works best at its rated power input/output, but I don’t know if it would be practical to make an “efficiency-regulator” that purposefully cuts off the output during low output operations. Like I did with the electric motors, which can close off some of its wires to “virtually” reduce its output.

On an unrelated note, what should I name it?

I guess that “Guaraci engine” would be a good name, since it is supposedly meant to work so unbearably hot you could even use molten steel as a coolant…

Now that I stop to think about it…

Wouldn’t the best solution be to use both of the power source options (molten carbonate fuel cell and helium turbine) in a Combined cycle power plant?

It is said that the molten carbonate fuel cell can only reach efficiencies up to 80% by recycling its heat source such as in the case of a steam turbine engine. The helium turbine is used just like a steam turbine engine in order to convert heat into work…

The only problem is weight and bulk.

🔶🔶🔶🔶

(I will write these orange squares to keep track of parts where I’m procrastinating, so I finish them already)

I need to make the model of all the heat exchangers into spiral heat exchangers (because these are the most efficient ones), these heat exchangers will also function as the fuel reformers.

The heat exchanger will serve both for the helium “turbine” engine generator and for the cryocooler (in case people actually use it for something).

Spiral Heat Exchangers Explained

3D Printed Heat Exchangers For High Temperature And Pressure | The Cool Parts Show Bonus 

Plate heat exchanger - Wikipedia

Which Heat Exchanger Is Best? The Three Main Types Explained... 

Heat Pipes and Other Heat Transfer Techniques 

Powder and other chemicals can be trapped inside of a heat exchanger. So it could be interesting to make it in a conical way so it works like a cyclone filter.

Dust Collection: Small Changes, Massive Effect! 

By the way, sheet lamination 3D printing can allow for very interesting heat exchanger designs:

Source: 3D Printed Heat Exchangers For High Temperature And Pressure | The Cool Parts Show Bonus 

Sheet Lamination 3D Printing: An Additive Manufacturing Process Explained #3dprinting #engineering 

It is also possible to use Plasmatron fuel reformers. You can also use cold-plasma/non-thermal plasma/non-equilibrium plasma to reform the fuel. A Developed Plasmatron Design to Enhance Production of Hydrogen in Synthesis Gas Produced by a Fuel Reformer System

Non-thermal plasma-catalytic processes for CO2 conversion toward circular economy: fundamentals, current status, and future challenges | Environmental Science and Pollution Research

Electrification of gasification-based biomass-to-X processes – a critical review and in-depth assessment - Energy & Environmental Science (RSC Publishing) DOI:10.1039/D3EE02876C 

Experimental Investigation on Compact Current Non Thermal Plasma Assisted Hydrocarbon Reforming Hydrogen Rich Gas

Low Current Non-Thermal Plasma Assisted Hydrocarbon Reforming Hydrogen Rich Gas

Partial oxidation and autothermal reforming of heavy hydrocarbon fuels with non-equilibrium gliding arc plasma for fuel cell applications - Drexel University 

Non-thermal plasma-assisted steam methane reforming for electrically-driven hydrogen production (although this one talks about steam reforming, since you will take wood to turn into fuel for the carbon fuel cell, it will also contain some amount of water, which in turn, will turn into steam) As far as I could see, most of the articles focus on producing hydrogen or syngas in more processes both for hydrogen fuel cells and synthesis of other fuels and materials

I don’t know if this will be relevant, but every now and then I check out radiators for spacecraft.

They are a fascinating subject, although I don’t know if they would be that useful in an atmosphere.

Ribbon radiators seem to be the most practical ones, since they expand with the conversion of liquid coolant into vapor and contract once it releases heat.

Source: https://www.projectrho.com/public_html/rocket/heatrad.php 

🔶🔶🔶🔶

(I will write these orange squares to keep track of parts where I’m procrastinating, so I finish them already)

Since I want to actually finish the mech project, I will abandon this idea and focus on the helium engine.

BEFORE HANDLING LITHIUM CARBONATE (and other chemicals), PLEASE REMEMBER THAT I AM NOT AN ENGINEER AND YOU DO ANYTHING AT YOUR OWN RISK.

PLEASE READ:

Hazardous Substance Fact Sheet: Lithium Carbonate https://www.nj.gov/health/eoh/rtkweb/documents/fs/1124.pdf

Hazardous Substance Fact Sheet: Sodium Carbonate https://www.borderjanitorial.co.uk/files/COSHH/SodiumCarbonate.pdf

Hazardous Substance Fact Sheet: Potassium Carbonate https://www.chemos.de/import/data/msds/GB_en/584-08-7-A0216451-GB-en.pdf

Observation:

I was revisiting metal hydrides and I found out that aluminum hydride, lithium hydride and lithium aluminum hydride can hold hydrogen with 10% of its weight and releases hydrogen passively even at ambient temperature. You would need to cool it down to around -70ºC to -100ºC in order to avoid that.

Well, you would still need 100 kilograms of aluminum for every 10 kilograms of hydrogen, but in this state, it is 2 times denser than liquid hydrogen, which already has 142 megajoules of energy per kilogram.

But if you use an alkaline fuel cell and if you expel the resulting water, you could maintain 100 horsepower for around 10 hours with 100 kilograms of aluminum and 10 kilograms of hydrogen.

Even if you had an efficiency of 50% instead of 70% to 80%, you would still “just” need around 200 kilograms of aluminum.

However, as you can imagine, it is extremely dangerous and difficult to synthesize aluminum hydride and lithium hydride.

ChatGPT keeps saying that you need to convert lithium hydride into lithium aluminum hydride in order to get aluminum hydride. And in order to get lithium hydride:

“Reduction of Lithium Hydroxide:

Another method involves the reduction of lithium hydroxide (LiOH) using a reducing metal. In this process, anhydrous lithium hydroxide is mixed with a reducing agent (such as aluminum or magnesium) and heated in the absence of air. The reaction occurs at temperatures between 300 °C and 550 °C, producing lithium hydride and the corresponding metal oxide “

In the end you would need a smaller molten carbonate fuel cell to slowly recharge the hydrogen fuel cell. But if you want to save weight as much as possible, then it could be a good compromise.

Observation 2:

Cold plasma, or non-thermal plasma, or non-equilibrium plasma is a type of plasma that can be made without melting everything that it touches.

However, since it is full of free electrons and free ions, it can be used as a catalyst for chemical reactions.

Nonthermal plasma - Wikipedia 

Cold Fire You Can Touch - DIY Cold Plasma Torch

The Cold Plasma Wand That Heals (Microjet)⚡  

You could (or should) add this to the fuel cell.

Well, I’m writing this now because I missed a really important issue: Vegetable coal has sulfur on it.

Sulfur can “poison” the Molten Carbonate Fuel Cell Electrolyte, degrading the fuel cell performance/efficiency.

Something as low as 0.1 grams can affect the cells.

Vegetable coal can have around 100-200 grams of sulfur in 200kg. And since the idea was to use vegetable coal as fuel…

And there is no simple method to remove it in biomatter, you need acid baths using sulfuric acid, water bath etc.

There are sulfur and chloride scrubbers, but they are applied when the biochar/biofuel is converted into Syngas.

But I can’t find a compact and easy solution for it, they mostly use giant chemical filters that need to be constantly re-treated using a myriad of different methods to avoid saturation.

I need to check if all of this is true, but essentially, ChatGPT said the following:

All of these methods release sulfur in a gas form.

Now for chloride:

So the idea is to:

Also, Pyrolysis (the process of heating charcoal/biomatter in a low oxygen environment) only works at around 800ºC to 1000ºC with 90% efficiency in conversion of matter into syngas.

Some papers describe how they only use sodium carbonate and potassium carbonate, not even worrying about lithium carbonate.

They also describe how the cells work at 800ºC because all of the carbonates are in liquid form at such temperature.

So…

There will be multiple zinc scrubbers and sodium carbonate scrubbers that will be selectively closed to go through the reversing process.

You can “just” copy the parts and replace them with the proper materials, sodium silicate, potassium silicate, lithium silicate and calcium silicate (not obligatory) mixed with sodium carbonate, potassium carbonate and lithium carbonate.

I also talked about how these fuel cells output is based on the surface area contact of the electrodes with the electrolyte, so in order to keep it with the biggest surface area as possible, you would need to use super thin stainless steel wires stacked parallelly for that.

More specifically, 120 milliwatts per square centimeter, divided by 300,000 watts, the area in square meters would be around 4774m² assuming 120 miliwatts per square centimeter.

But assuming around 96 mW/cm² I would need 1,041,666 cm² for 100,000 watts.

This would be the surface area of a stainless steel 0.8mm thick wire with 45 meters of length weighing around 0.2 kilograms. For 1mm of thickness it would be around 450 meter long stainless steel wire weighing around 3 kilograms.

Observation:

I always assumed that the energy generated each hour by the Molten Carbonate Fuel Cell powered by wood would be 30 megajoules per kilogram of wood (either because I’m insane or because I didn’t question ChatGPT), but that is incorrect.

The energy density of wood is 16 megajoules per kilogram, and assuming the whole process has a 50% efficiency, then you would have a total of 8 megajoules per kilogram.

Since using 100 horsepower per hour would require 264 megajoules, then it would require 2647 megajoules in 10 hours.

2647 Mj / 8 Mj per kilogram = 330 kilograms of wood in total for 10 hours at 100 horsepower. 33 kg of wood per hour, 5.5kg of wood per minute and 0.1 kg of wood per second.

But since circles/cylinders does not have the best surface area contact, I will have to make some considerations:

As you can see in these wires, only a small part of it is in contact.

On the right, the cylinders have 18.8mm² of area, on the left, they have 5.8mm² of area. So you would need to increase the number of wires in contact by 3.2x times (a little more, 3.3 or 3.4). That is true for any thickness of wire (I tested it).

So, you would need 0.66 kilograms for every 100 kilowatts for the 0.8mm wires and 10 kilograms for every 100 kilowatt for the 1 mm wires.

In order to enhance the performance of the molten carbonate fuel cell you can add a few additives.

I will make a list, but remember: I’m talking about adding around 1% of these additives per weight of electrolyte (not the total weight of the cell), all of them or each one of them and even then, they can cause unexpected issues in either quantity (or simply not affect the cell significantly, for better or for worse).

They can also be dangerous in their own right.

The list:

Magnesium carbonate, calcium carbonate (this one is said to “clog” the electrolyte in some case), barium carbonate, inconel, lithium oxide, samarium oxide, magnesium oxide, nickel oxide, copper oxide, titanium dioxide, cobalt oxide, lanthanum oxide, manganese oxide, iron oxide, aluminum oxide, vanadium oxide, molybdenum disulfide, platinum, ruthenium, tungsten oxide, silicon carbide, yttria-stabilized zirconia and rhenium.

Observation:

When I try to buy zirconia powder by itself it is insanely expensive, but ceramic knives are made out of it and they are super cheap, even in packs.

The weight of the electrolyte may heavily depend on its thickness.

If you assume the electrolyte has a density around 2.3 g/cm³, 1mm of thickness and the area I calculated before for the wires (3,437,497cm² per 100,000 watts), you would need around 790 kilograms.

To keep it around 1 kilogram, you would need to make its thickness around 0.001 mm, which is 1 micrometer.

Which is not very realistic.

You would need to increase its weight by 10 to 100 times (10kg to 100kg) in order to keep the thickness around 0.01mm to 0.1mm.

That is the weight before mixing it with the silicates (sodium, potassium etc), by the way.

What should I call this fuel cell?

Tupã-5?

Meaning from Wikipedia:

“Tupã or Tupan (also Tupave or Tenondete) is the word for God in the Tupi and Guarani languages, including the Guarani creation myth.

Tupã is considered to be the creator of the universe, of humanity and of the spirits of good and evil in Guarani mythology referred to as Angatupyry and Tau respectively. Tupã is more specifically considered the creator of light and his residence is the Sun.

A contest was run by the Centro Paraguayo de Informaciones Astronómicas in Paraguay in 2019. This bestowed the star HD 108147 with the name Tupã in December 2019.”

By the way, I will increase the total output to 750 volts and 500 amps, 375,000 watts in total (500 horsepower), because of the possible inefficiencies in the system and in the actuators.

Also, I’m wondering what kind of material I should use for the electrodes.

Yes, normally nickel, inconel and stainless steel are used as electrodes, but they almost never use these in the micrometer thickness. Even less in the powder form as I’m intending on doing.

A few electrically conductive ceramics I found:

Silicon carbide, tin oxide, lead oxide, bismuth ruthenate, bismuth iridate, iridium oxide, ferrite/iron oxide, Indium Tin Oxide (ITO), Lanthanum-Doped Strontium Titanate (SLT), Yttrium-Doped Strontium Titanate (SYT), Yttria-Stabilized Zirconia (YSZ), Gadolinium-Doped Ceria (GDC), Lanthanum Strontium Gallate Magnesite (LSGM), Beta Alumina, Silicon Nitride (Si₃N₄), Zinc Oxide (ZnO), Barium Titanate (BT), Lead Zirconate Titanate (PZT), Molybdenum Disilicide (MoSi₂), Lanthanum Trifluoride (LaF₃), Lead(II) Chloride (PbCl₂), NASICON (Na₃Zr₂Si₂PO₁₂), Silver Iodide (AgI), Rubidium Silver Iodide (RbAg₄I₅), Barium Titanate (BaTiO₃).

Well, I think the most practical alternative for electrode material would be using silicon carbide and/or tin oxide.

Well, I slowly asked ChatGPT to check one by one just to be sure, because it doesn’t delves into details when talking about a lot of subjects at once:

Now all I have to do is choose.

… I'm not in the mood at the moment to do so…

Making Ionic Liquids: The Future of Solvents and Catalysis? 

I was wondering about the possibility of having an alkaline biomimetic fluid system similar to blood.

Alkaline fuel cells can work with any kind of fuel, but any content of carbon can turn the hydroxides in the cell into carbonate crystals.

So the idea would be to continuously pump hydroxide based electrolyte fluid/gel through the membrane and continuously feed the carbonate through a machine that would reverse said process.

I asked ChatGPT and DeepSeek:

Causticization with Calcium Hydroxide: Process: Sodium carbonate reacts with calcium hydroxide (slaked lime) to regenerate NaOH, while calcium carbonate precipitates out. Requires a steady supply of calcium hydroxide.

The byproduct, calcium carbonate, must be thermally decomposed (∼900°C) to regenerate CaO (quicklime), which is then hydrated back to Ca(OH)₂. This adds complexity and energy costs.

Electro

Discussions