Ultra micro turbines

[remembereverything]

New to turbine and jet engines but more recently decided I want to build something unique. So I came across this awhile ago and recently rediscovered it.

www.ecojetengineering.com.au/technology/

I'm curious how worthwhile a turbine like this in the 500w to 1kw range could be. From how small it is could this be a decent diy project. Yeah I'd have to order some expensive custom made parts. I'm just wondering if something this small be reliable. Even if it was a bit bigger could a micro turbine like this have a decent 1000 plus hour rebuild time? Or could things that this design lacks but could be modified like a specific bearing type for example. I'd love to have a small turbine powered bike or longboard. At the size at least of this example that would be totally doable.

Guess I'm just looking for more info on a potentially below 2kw turbines that I could use as either a genorator in a electric hybrid build or gear down to run direct.

[remembereverything]

Related to this idea I'm also curious if a small carburetor with the proper jetting could be used after the first stage for fueling.

[racket]

Hi

For such small applications the RPM are extremely high , gearing , bearings, everything, is just that much more difficult , .....read expensive , generally for military use where cost isn't such a consideration .

Carbs aren't used on GTs as we only burn part of the air so no point having fuel in all of it , it'll only end up melting the turbine wheel

There are Papers on the Net about extremely small turbines if you'd like to do a bit of searching .

LOL...............we're generally heading in the other directing , trying to get bigger and bigger units to play with :-)

Cheers
John

[turbotom]

I've got some experience with very small gas turbine engines, but still of "makroscopic" size, i.e. based on the smallest commercially available turbocharger components. The smallest one was in the 30mm rotor diameter range, called the "Kolibri" by Lambert Turbines. I contributed with compressor and turbine redesign to increase the thrust of the tiny turbojet from approx. 15N to >30N at basically the same external dimensions. These little technological marvels spin at 250krpm and are basically used for aeromodelling (toys... ).

The second smallest engine I've worked on was a design of my own and performed as a turbo generator. It's been a contract job from about a decade ago and it utilized modified 45mm turbocharger rotor components, spinning at 160krpm. The unit was designed to produce 5kW of electrical power. Unfortunately, it still can be considered an unfinished prototype since commercial interest wasn't there to an extent that's been anticipated and technical difficulties made series production prohibitively expensive -- at least in the low quantities that would have been requested.

FYI, I attached a few pictures of the prototype.

My take at even smaller units is that they are probably feasible, but at steeply dropping efficiency. Below 2kw, I'ld say a recuperator is mandatory, and units of less than 500W will be limited by the self-sustain conditions of the engine itself, i.e. regardless of the power, the turbomachinery cannot be reduced in size anymore. This will probably incorporate impellers of 15...20mm diameter and speeds of 350...500krpm. It would be an interesting but pretty challenging project, both from a science/engineering as well as financial point of view...

[drumwilldrum]

Hi Tom, 

Fascinating engine you have there. 

Can you comment further on the technical difficulties that made series production prohibitively expensive?

It looks like you used ball bearings, how did they hold up to the high rpm and how were you lubricating them? 

Did you bore out the turbine or is the shaft extension mated to the original turbo shaft in some way?

Thanks
Will 

[turbotom]

Jan 18, 2024 0:53:59 GMT -5 drumwilldrum said:

Hi Tom, 

Fascinating engine you have there. 

Can you comment further on the technical difficulties that made series production prohibitively expensive?

It looks like you used ball bearings, how did they hold up to the high rpm and how were you lubricating them? 

Did you bore out the turbine or is the shaft extension mated to the original turbo shaft in some way?

Thanks
Will 

Will -

you may want to have a look at the attached drawing of a very early version of this turbo generator (comments in German, sorry...). But it shows basically all the design features.

The rotor concept is over-critical, i.e. the engine is running above the first shaft bending eigenmode. This means, the shaft can be considered "soft" and the shaft and the turbomachinery components need to be balanced individually to very high accuracy. The shaft is designed as an "exoskeleton" for the generator magnet. The brittle nature of the solid SmCo magnet requires a reinforcement at the high rotational speed of the generator (160krpm). Thus, the magnet's outer diameter is precision ground and the hollow shaft section which is machined a few hundredth of a millimeter smaller than the magnet, shrunk onto it. After this, the "far end" of the shaft is pressed into the hollow end and laser welded shut. Only after this, all the bearing and highly accurate surfaces are ground to size. The shaft is made of an ultra-high-strength maraging steel (Rp ~2000N/mm²). Since the generator is of a (magnetic) two-pole topology, balancing requires some care since the interaction of the powerful magnet with the earth's magnetic filed tends to manifest itself as a dynamic imbalance.

Suspension isn't too much of a difficulty, modified hybrid ball bearings (with machined bronze or phenolic cages) and axial/radial squeeze film damping work quite well here. Since all the bearings are located in the "cold" area of the engine, lubrication with a dedicated pump and a circulating type lube system is adequate and sufficient. The cooling shround of the generator will also sufficiently cool the oil.

The turbomachinery is set up of modified turbocharger spares. The compressor impeller has the bore slightly enlarged vs. the original configuration so it can be shrunk onto the projecting generator shaft. The turbine wheel has its shaft shortened and ground to a smaller diameter and its end threaded to directly screw into a matching internal thread in the generator shaft. So the turbo's original friction weld is kept and the wheel isn't drilled through.

Axial grooves in the projecting part of the generator shaft guide compressed air from between the compressor and the turbine to the compressor spacer area where it serves as labyrinth pressurization air and seals the lube system. This works so well that there's no oil loss observable at all.

Since the generator is configured as a three-phase, air-gap coil annular type with a stack of three individual toroids, winding it is an arduous job. Each of the split two phase coils are internally connected in series, but since all individual toroid connections are still "brought outside", there's a total of 18 electrical terminals. This appoach turned out very beneficial for starting since less voltage is required to reach self-sustain speed if only part of the stators are used for motoring. But winding, assembling, potting and soldering this setup is not far from a nightmare.

The disc-type vaporizing combustor works better than anticipated -- there's a strong swirl motion present inside the primary zone (some of the compressor diffuser exit swirl is utilized for that) which effectively lengthens the primary flame zone. As with many single-shaft turbine engines, no-load lean burning characteristics are somewhat problematic and the engine tends to flame out off-load. The approach is to apply a certain minimum load already during acceleration above 100krpm to keep the required fuel mixture rich enough to avoid that problem.

Another shortcoming of this engine is the ridiculously bad fuel efficiency. We're talking of a specific consumption of round about 1kg/kWh of Jet A1 or diesel fuel... Better economy would require a recuperator with its weight penalty and added complexity.

In order to cut production cost, the quantities to be produced would have to be quite high. A few hundred such units would basically still be a "manufacture product" which requires very high skill levels and attention to detail, thus too expensive for most applications.

I guess this gives you an idea of the difficulties involved...

Cheers,
Thomas

[drumwilldrum]

Hi Thomas,

That is both fascinating and seriously impressive engineering. Was the design and manufacture self funded!?

It isn't clear to me from the drawing how the oil is returned to the pump, it is circulating not total loss correct? In via one bearing and then out via the other possibly, but how does it flow between the bearings?

Do you have any pictures of the disk-type vaporizer and combustor?

What was the driver for the selection of that architecture for the generator, power density?

As a casual observer I can see a lot of areas for improving manufacturability but I guess the efficiency put a stop to iterative effort.

Cheers
Will