I understand that from a efficiency standpoint, a plain ol' diesel generator will be much better than a turboshaft for electricity generation.
However, just for the novelty of having one, I'd like to build one on the long term.
What kind of tools would I need to make the blades and such? I'm assuming they need to be cast or milled for strength, and there's a limited variety of materials you can use, at least for the combustion section.
You can make one from an old car turbo. The fans are incredibly hard to make even for people with the proper tools. The balance needs to be perfect so it doesn't vibrate itself apart at 15000 rpm.
>>1002120
>I understand that from a efficiency standpoint, a plain ol' diesel generator will be much better than a turboshaft for electricity generation.
A /diy/ turboshaft, yeah probably. But more generally-speaking, nah. Aircraft turboshafts/turboprops are fairly inefficient since they prioritize power-to-weight and reliability over raw efficiency, but marine turboshafts and those gas turbines used in power stations can beat out pretty much all "plain ol' diesels," with only those gargantuan 2-stroke marine turbodiesels surpassing them.
But anyways.
>What kind of tools would I need to make the blades and such?
How "right" do you want to do it? If you don't really give a fuck about performance or longevity, you can actually get away with pretty lean methods using very basic tools and materials. For example, Kurt Shreckling designed and built an outrageously-/diy/able, working centrifugal turbine engine (pic related), featuring a turbine blisk that's cut, bent and dremel-ground from a scrap of sheet-steel, and a compressor made from fucking plywood:
https://www.scribd.com/doc/3746230/Gasturbine-Engines-for-Model-Aircraft
If you want to follow the "homebuilt turbine engine" cliche, you could repurpose a turbocharger fairly easily as well, and probably achieve better performance than the Schreckling design (just as long as weight isn't a factor, that is) by running higher pressures and temps
Both of these options would require you to somehow integrate your own PTO to the shaft, or better yet you could build a separate multi-stage power-recovery turbine driven by the core exhaust as most turboshafts are.
If you're looking for more conventional manufacturing methods, see here:
https://www.youtube.com/watch?v=hOv2pGEruGk
Of course, modern engines often use blisks instead. Perhaps you could 3D-print a blisk pattern and cast that. Also you may not have the resources to cast nickel superalloys at 3000*, so watch your ITT/fuel flow if you use steel or whatever.
>>1002213
thanks for the links
I could probably get into casting them, since 3d printers are pretty cheap nowadays, I'd just have to make sure to balance it properly which would be hard.
To start with, I think a automotive turbo would be the best to start with just so I don't have to fuck around with casting or welding shit too much.
>>1002236
Fair enough.
Perhaps you could build the turbocharger-based "core" first, and then experiment with different approaches, designs and methods to make your own power recovery turbine to attach to the core and turn your generator. If it melts or kabooms, at least you still have your core. And if it doesn't, maybe you can apply what you've learned to building your own core with a higher pressure ratio and greater efficiency.
>>1002120
They're using all kinds of high-tech stuff nowadays, but the traditional material for turbine stages/rotors is high-nickel steels like inconel and hastelloy. They are difficult to machine due to their toughness/hardness.
(in the USA) McMaster-Carr sells 6-inch-square pieces of "alloy-X" (a generic nickel-steel they say is similar to inconel) 1/4-inch thick, for about $150.
Turbochargers don't work well as jet engines since the compression stage is smaller than the exhaust. They end up being more of a "turbo-rocket".
If you could join the back end of a small turbo to the front end of a big turbo, that would work better.
Turbocharger jets also have horrible RPM ranges. It's not unusual for them to idle at 75,000 rpms and top out at 100,000 rpms. It can take a BIG leaf blower pumping air onto the front to get them started.
This is due to only having one compression stage. If you could make multiple compression stages with proper sequential bleed valves, you could bring the idle RPM speed down a lot and the thing would be a LOT easier to get started.
I have never tried building a jet engine myself. Kinda wanna sorta, but may never.
A common problem I've seen other people have is that the tilted rotor blades tend to flatten out at higher-RPMs and temperatures.
For that reason I have thought it would make sense to try using rotors that are "flat" (not inclined) and have the stators tilted (since the stators don't suffer the centrifugal stress) but I've never seen anyone do it that way. And you'd have a performance hit as well. But having it work and be fairly durable would still be nifty.
>>1002120
You don't understand anything about how generators or Turbine engines work? do you?
>>1003055
Not OP, but what makes you say that?
>>1002120
Look up plans for an RC plane jet engine.(too lazy to post em right now, but it takes like 10 seconds of googling to find them, paid plans are absolute trash though)
Most go like this:
1. machine a combustor
2. machine a case
3. order a turbine/compressor from a some company in korea, because holy fucking balls is it difficult to machine superalloy parts
4. JET ENGINE!!!!!
If you choose to machine the turbine yourself, it's a fucking nightmare. We're talking months of machining here. If you choose to do this, you better go balls deep, and buy the right equipment. It will become your life.
Even if you have a fucking 5 axis CNC, it's still pretty fucking difficult.
>>1002236
Good luck casting superalloy. Oh and did I mention it needs to be singlecrystalline. J/K, this is only a problem if you're running it for a long time. This is not a problem, because you probably won't be able to afford to run it that long.
But hey, if you buy one of those turbines from some korean company and have a mechanical engineering degree, and studied turbomachinery, you might actually be able to make something that could power your house.
Micro gas turbines have been used for combined heating and power
>>1003086
>Look up plans for an RC plane jet engine
That would be wrong tho...
Airplane-style jet engines are dynamic-thrust engines, most of their energy output is in direct intake-exhaust thrust. OP wants to spin a generator, so he needs a turboshaft (like a helicopter engine).
>If you choose to machine the turbine yourself, it's a fucking nightmare. We're talking months of machining here. If you choose to do this, you better go balls deep, and buy the right equipment. It will become your life.
>Even if you have a fucking 5 axis CNC, it's still pretty fucking difficult.
I've seen guys on web forums cut mini-jet-engine parts before, and it doesn't take months.
All the ones I think I've seen had real metal-cutting milling machines converted to CNC tho, not 80/20 CNC routers.
And you might burn through a few micro mill-bits which can cost ~$15 or so each, but it's hot a horrible sum.
The trouble with making jet engines is that the specific fuel efficiency (the amount of power produced per unit of fuel burned) goes up as the engine's RPMs goes up.
So if you want it to be very efficient at all, it has to be engineered (the mechanical properties AND the internal flow design) to operate normally at 90-95% of its maximum RPMs.
It is this sort of engineering and metalworking that ordinary mortals have difficulty doing at home.
If you just want something that makes jet noises and spins at ~20K RPMs, that isn't real difficult to do (with machine tools).
It's not gonna be very fuel efficient tho, and if you run it very hard it's gonna fail pretty quickly.
The power comes from the force of expansion of heat that's why there is a compressor section and where the gas goes in the GSM pressure has to be higher than the air compressor section. Also the amount of nozzles on the compressor section have to be thought of for even heating of the combustion blades
>>1003086
>Good luck casting superalloy.
Watch the video: [>>1002213]
>it needs to be singlecrystalline.
>>1003107
>Airplane-style jet engines are dynamic-thrust engines, most of their energy output is in direct intake-exhaust thrust.
There are several turboshaft engines that are based on jet engines which simply have the nozzle (and afterburner, if applicable) removed and a separate power-recovery turbine attached instead.
https://en.wikipedia.org/wiki/General_Electric_LM1500
https://en.wikipedia.org/wiki/General_Electric_LM2500
https://en.wikipedia.org/wiki/General_Electric_LM6000
Several turbine RC helicopter engines work the same way, starting with a model turbojet and attaching a power recovery turbine with output shaft to the exhaust.
>So if you want it to be very efficient at all, it has to be engineered (the mechanical properties AND the internal flow design) to operate normally at 90-95% of its maximum RPMs.
Temperature is the more pressing limitation for efficiency. It's possible to overengineer the turbomachinery to withstand even higher RPMs, but generally engine designers don't make the RPM limit any higher than it has to be in order to reach operating pressure ratio and temps. Doing so allows them to design the engine with less material, reducing both weight and cost.
>where the gas goes in the GSM pressure has to be higher than the air compressor section.
No, it doesn't. In fact, achieving that condition is impossible. Pressure at the turbine inlet can only be less than or equal to pressure at the compressor exit.
>Also the amount of nozzles on the compressor section have to be thought of for even heating of the combustion blades
I don't even... what?
>>1003077
Turbine engines are very high speed, very low torque. In order to use one as a generator you would need one hell of a reduction gear assembly. You will get a lot of torque that way but also use a lot of fuel.
Building a turbine engine is not something that someone just throws together. You need special alloys, very precise tooling and the list goes on.
Even if OP uses an off the shelf Turbine or Jet engine the whole thing is.... silly.
I hope OP has a full machine shop.
>>1003086
>needs to be singlecrystalline
i can throw my money behind this point. all the best high performance turbine blades are very specific single crystalline nickel - iron "superalloys".
Basically the super means they can operate at super fucking high temperatures increasing the efficiency of the engine because you can push up the inlet temp.
>source: im fucking retarded
My cube neighbor used to work for a turbocharger company. I think he said they cast the turbine blades. But I thought aircraft compressor shit is inconel++ and has to be ultrasonic/xray screened.
>>1003397
>Turbine engines are very high speed, very low torque. In order to use one as a generator you would need one hell of a reduction gear assembly.
Or you could just use a flux-switching alternator, but I digress. Gears are no big deal.
>Building a turbine engine is not something that someone just throws together. You need special alloys, very precise tooling and the list goes on.
Nope, doesn't have to be. See: [>>1002213 https://www.scribd.com/doc/3746230/Gasturbine-Engines-for-Model-Aircraft]
>>1003449
you can use xrays to check the crystal structure.
https://en.wikipedia.org/wiki/X-ray_crystallography
>>1003318
>Temperature is the more pressing limitation for efficiency. It's possible to overengineer the turbomachinery to withstand even higher RPMs, but generally engine designers don't make the RPM limit any higher than it has to be in order to reach operating pressure ratio and temps. Doing so allows them to design the engine with less material, reducing both weight and cost.
Yea, but jet engine compressors are open-flow, I.E. not restricted-flow like a piston engine or piston air compressor is.
Jet engine compressors (axial or radial) tend to have a working efficiency that is directly proportional to their rotation speeds--they produce higher pressures the faster you spin them.
A piston-engine or piston-type compressor can reach it's maximum pressure with just 1 revolution, even if it is only turned once at a speed that is equal to only a few revolutions per minute. The factors that determine what compression it reaches are much less dependent upon the RPMs it spins at. Most piston engines maximum power and torque ratings do -not- occur at their highest-rated RPMs.
>>1003449
>My cube neighbor used to work for a turbocharger company. I think he said they cast the turbine blades. But I thought aircraft compressor shit is inconel++ and has to be ultrasonic/xray screened.
The wheels are cast but then machined afterward.
What kind of quality-control turbo companies use I don't know--but I'd bet it's not as strict as what airplane jet engines use.
Also I recall reading that a car turbo's exhaust rotor gets way hotter in use than the rotors inside an airplane jet engine.
Jet mech here. The poster who suggests getting a working core is correct. That's the toughest part by far.
If you genuinely want a power plant, look at Capstone turbine generators for inspiration and copy them. They are not only good for backup power but powering hybrid vehicles.
http://www.capstoneturbine.com/
I suggest finding a surplus helicopter or "jet fuel starter"/APU turbine (you aren't going to build one from scratch unless you become a machinist AND have buckets of money to spare) then adapting a generator head of your choice via a reduction gearbox. Making the gearbox would be challenge enough as would the rest of the system.
Older turbines can tolerate a variety of fuels because they "trim to temperature".
Turbine gensets have been used in aviation for many decades. The ubiquitous "-60") military generator also produces high volume compressed air which can drive air starters, air conditioning carts (turbine powered by air) or anything else you wish.
Fun to study even if you decide not to build one.
>>1004683
Why do the small ones always have radial turbines? Is it possible to make high powered multistage radial ones?
The engine used in tomahawk missiles has a really nice compact axial core but they don't sell them. They aren't restricted or anything apparently the manufacturer just likes selling them to the military at inflated pricing so they won't risk undercutting their own prices by selling them.
>>1004972
>Why do the small ones always have radial turbines?
They don't. The turbines are almost always axial, but centrifugal compressors are common on small engines because they are far simpler to manufacture and operate, especially for a given pressure ratio. A single-stage centrifugal compressor can achieve a pressure ratio of 2.5:1 with decent efficiency, whereas an axial compressor would need several stages to attain such performance. Furthermore, axial compressors are much more temperamental, and often use complex variable stator vanes to overcome compressor surge and choke issues and broaden performance.
Downsides to centrifugal compressors include large frontal area (and drag), convoluted flow (complicating the arrangement of such compressors into stages), greater mass and high moment of inertia (and thus jet lag).
>Is it possible to make high powered multistage radial ones?
Sure. It would be somewhat more complicated and unwieldy, but sure.
>The engine used in tomahawk missiles has a really nice compact axial core but they don't sell them.
The F107 has a 3-stage compressor including a centrifugal HP stage. It's not entirely axial.
It ALSO has a 2-stage bypass fan on a separate LP shaft. There's a reason the engine costs what it does; it's not JUST mil-spec markup.
If you want a cheaper turbine of a similar class, might want to look at the PBS TJ100. It's a much cheaper and simpler single-spool turbojet. There's also a turboshaft variant (which is just the TJ100 core with a bolt-on power recovery turbine) that puts out around 240 hp. OP might also want to look at this for inspiration.
>>1003695
>compressor
>inconnel
no
Turbine - yes, totally. But not the compressor. Those are made from cheaper stuff, they don't run nearly as hot. Aluminum or titanium alloy, some are from stainless...
>>1006572
Titanium alloy and steel are both only good to a working temperature of about 500* C (aluminum is even less). That may be high enough for a subsonic engine with fairly low pressure ratio, but engines with higher pressure ratios (>25:1) or those designed to operate at high mach must employ high-temperature nickel superalloys in the compressor (at the very least, in high-pressure stages). The J58's inlet temperature was so high at mach 3.2 that 8 of its 9 compressor stages had to be made out of nickel superalloys to withstand the temperatures of the highly-compressed air.
