"Fat Boy" 12/118 thrust engine

[azwood]

Or fire the fuel into the tubes with them

May 10, 2018 19:09:57 GMT -5 azwood said:

What about some sort of danfoss jet on the end of your evap tubes to make sure the fuel that gets out atuly cums out in a mist just an idear I'm not that educated on them yet

[racket]

Hi

I had mist sprays until just recently , if you go back a 4 pages on the Thread, it'll explain what happened during a tube testing , the airflow through the tube was high enough that it didn't allow the spray to "fan out" and contact the wall :-( the airflow is at ~40 m/s , so only a couple of milliseconds for things to happen

I've made up a new drum of fuel "brew" , so hopefully I notice some changes, one way or the other .

Cheers
John

[enginewhisperer]

Is it worth trying to get some LPG into it for a test?

[azwood]

May 10, 2018 20:29:34 GMT -5 racket said:

Hi

I had mist sprays until just recently , if you go back a 4 pages on the Thread, it'll explain what happened during a tube testing , the airflow through the tube was high enough that it didn't allow the spray to "fan out" and contact the wall :-( the airflow is at ~40 m/s , so only a couple of milliseconds for things to happen

I've made up a new drum of fuel "brew" , so hopefully I notice some changes, one way or the other .

Cheers
John

Ok that makes sence sorry disregard what I said lol

[racket]

Hi Andrew

It'd have to be liquid propane , and I'd like to keep away from using that other than for a last resort test to see if the engine would run up to full power , just too many hassles to setup, and expensive to run as a BBQ bottle wouldn't last very long , and not all that convenient if the engine is installed in a vehicle .......................but the option is still open :-)

Cheers
John

[stoffe64]

May 10, 2018 19:01:50 GMT -5 racket said:

I haven't the faintest idea if the 50/50 mix will improve things , all I can do is try something and see what happens .

I was very hopeful that the evap mods would have improved things more than they did , but theres a chance that the mods simply meant more fuel was in contact with the tube wall and kept the wall too cool to evaporate the higher boiling point bits of the kero , perhaps all the lighter fractions evaporated compared to before where they may have been a broader range but less off it .

Vapouriser tubes never evaporate all the fuel, even in full sized aero engines, a large percentage gets blown onto the front wall of the flametube to be evaporated off , but it appears I have a greater percentage dribbling down the front wall the higher the P2 , this is whats giving me an ever increasing difference in temps between the two thermocouples as P2 increases.

Stephan , ...........on a turbo based engine theres no excuse for having an undersized combustor , but on our "micro" types it gets more complicated as axial distance is severely limited , ideally we need ~1.5 times the flametube diameter/annular width as the length of just the Primary Zone , my whole flametube length is only 1.2 times, its extremely short.

If the 50/50 mix doesn't cure the problems I can always try a bit higher concentration of petrol, then if that fails , I probably need a different flametube , with more evap tubes to change the effective flametube L/D ratios ....................the funny part is , I originally was going to have 27 X 3/8" tubes , a 50% increase on our 18 X 3/8" TV94 number to match the increased mass flow , but decided the extra complication was something I didn't need ...............duh, it might have been a smarter move in hindsight .

Cheers
John

Hello john...i know that,but if we say that it got constructed the wrong way,this could be tried maybe to get it running (but ofcourse it should be constructed the correct way to start With)
Cheers /stephan

[racket]

Hi Stephan

LOL................I might be able to answer that after the next test run , hopefully as a positive , but the "undersized" flametube , if "skinny" , probably still wouldn't work due to excessive air speeds as the flame speeds won't vary much.

Cheers
John

[CH3NO2]

Hi John,

I know it would be a major re-work and may not even be technically feasible or worthwhile, but would it be possible to extend the length of the turboshaft, CHRA and combustion liner?

Tony

[turboron]

John, remember our discussion a few weeks or months ago about jets protruding into the combustion liner to improve mixing. They sound like a possible solution for the short length to me.

Thanks, Ron

[turboron]

John, if you can solve this problem it is a potential breakthrough for DIY gas turbines based on turbocharger components.

Thanks, Ron

[racket]

Hi Tony

A lengthening of the shaft is possible , but it brings with it a whole new set of potential problems and would effectively require a completely new engine being built , not worth the trouble, one of the positives with our current setup is standard "off the shelf" bits that work .

Running on liquid propane/LPG would be the simplest fix if I only wanted a "test stand engine" , but if it was mounted in a kart/bike its not practical .

I'm doing more research on vapourising systems to see if I can find a solution/s

Cheers
John

[racket]

Hi Ron

I'm using a number of propane injectors fitted into some of the Primary Holes to produce swirl within the flametube , swirl has its place , but it doesn't increase residence/dwell time within the flametube, and I feel that is one of my big problems .

I've also got to do some numbers to check what heat transfer there actually is across the vapouriser tubes , and what percentages of the fuel can be actually evaporated , by adding petrol to the mix, even the hot T2 air will evaporate petrol without the wall heat .

LOL............lots more to learn ,........ not such a disaster as this build is about the journey as theres no defined destination :-)

Cheers
John

[turboron]

John, the reason I like gas turbines is that really require you to get the gray matter working whether you want to or not. If you solve this problem you may be breaking new ground which is always worthwhile.

Thanks, Ron

[finiteparts]

John,

I have been giving some thought to your previous runs and I feel like the combustor is not the problem. Typically, the combustor experiences the most problems at light off and low power conditions. This is primarily due to the fact that chemical reaction rates are proportional to the 1.8 power of the pressure and exponential with temperature. So typically, you struggle with combustor issues at low power (low pressure and temperature) and if you can the engine up to speed, the problems resolve themselves. Now I do agree that the combustor sizing that you have is challenging, but that space rate number must be off due to some sort of error in measurement, because if that was real, you would be seeing signs of severe overheating when you took the engine apart. I think your combustor might be somewhat inefficient, but I do not believe it is the real limiting issue.

So if it is not the combustor as the primary issue, what is it? I revert back to my previous statements about the clipping. The severe clipping has reduced the flow turning in the rotor and consequently, has reduce the blade loading. The reduced blade loading means that there is less pressure drop occurring in the rotor and the bulk of the stage pressure drop is occurring in the NGVs. Hopefully, that clicks in your mind as a key design parameter, stage reaction, and with this distribution of pressure drops, it means that you have turned the turbine stage into a more impulse design. This follows from the discharge flow angles as well. If we define the discharge angles relative to the axial direction, an impulse stage would operate with the NGV discharge angle > rotor discharge angle...which is also likely due to your modification.

The problem with a more impulse stage is that it all becomes more important about how well the stators perform. Since you take the bulk of the pressure drop in the stators and we assume them to be adiabatic, and thus not doing any work, you have less total temperature drop through the stage. This could easily be your high EGT issue.

Impulse stages also choke earlier than the 50/50 stage or the reaction stage.  SO your intent to increase the choked flow capability of the stage might have inadvertently been hampered by the change in stage reaction.

Additionally, if you look at Rohlik's plot showing the estimated breakdown of losses in the radial turbine, you will see that since you have pushed the rotor to a higher specific speed, you will experience higher exit losses. The blunt trailing edges will also shave away precious efficiency, especially due to shock losses if you get the relative discharge speeds up to the transonic speeds, where designers worry about 0.030 inch or smaller trailing edge thicknesses.  Thus it is likely the stage efficiency was negatively impacted as well.

It does no good to state all this if I can't offer a solution, so this is what I would try.

If you have an unclipped rotor, I would replace it and see how far you can then push the engine. You should be able to push the stage to an overall higher power level and as such achieve a wider operating range.  I also want to make sure that when you calculated your rotor choking area, that you used the relative gas properties in the rotor and not the discharge gas properties, as this would have produced an erroneous effective area target.  Then of course there is the question of how you translated the effective area into a geometric area, estimated Cd, nozzle efficiency guess, etc?

My second though is to trim the NGVs back a bit to open up the NGV throat areas. This might allow you to get to a higher flow at the same pressure ratio and perhaps make more power in the turbine to better match the comp/turbine.

Lastly, if you don't want to do either of the previous, you could get the engine stabilized and try to very slowly increase shaft speed. The thought here is that the overfuelling that you need to produce enough unbalanced torque to accelerate the rotor is producing too large of a temperature rise, so by very slowly making smaller temperature rises, it will give you a margin to the surge or whatever is limiting the rotor speed increase.

Good luck,

Chris

[racket]

Hi Chris

Thanks for your thoughts , much appreciated .

Yep , if the engine will fire up and have reasonable operation at lowish power settings, the combustion should improve as the rpm and P2 increase ............LOL, thats what the theory's been telling me .

What I have is a steadily worsening difference between the two thermocouples in the jetpipe , at 1 Bar temps are 627 and 673 , just 46 deg C difference which is acceptable for a fuel rate of ~1.5 lpm , but at 1.5 Bar temps are 693 and 772 so 79 C degrees difference ( 72% increase ) with 2.2 lpm of fuel , but at 1.75 Bar its 816 and 967 a 151 C deg difference ( 91% increase from the 1.5 Bar figure) and a massive >4 lpm .

I'm feeding in ever increasing amounts of fuel , way more than what is being burnt given the TOTs , and theres greater temp differences between the thermocouples which sorta indicates fuel is dribbling down the front face of the flametube and burning where it shouldn't , or not burning at all .

The high TOT at 1.75 Bar - 2.75 PR is due to the "turbine stage" which currently has "restricted??" exducer flow , which will be requiring a greater share of the available pressure drop , and is the reason why I made the new jetpipe with its less restrictive flow which should take care of the "inbalance" between NGV and wheel pressure drop requirements , I didn't use it as I wanted to test the new square spring vapouriser tubes against the previous setup without any other changes.............LOL, other than removal of the comps extended tips.

On my Garrett TV84 engine I clipped its turb exducer the same as this ones , yet I was able to power that engine up to 45 psi P2 and have 12 psit of total pressure in the jetpipe , 12 psit represents a massive amount of "unused" energy , I'm currently running an "open pipe" , effectively allowing that 12 psit too be used to drive the comp , a comp running at 80% effic unlike the 71% of the TV84 .

A standard unclipped rotor would probably put the comp into surge unless it was run with considerable "exducer clearance" to provide an "escape route" for some of the exhaust gases, a standard wheel has ~6.6 sq ins of exducer throat area , I'm running ~8.6 sq ins , area into the wheel is ~11 sq ins

The NGV throats are currently 13.4 mm X 20.5 mm by 18 of for an area of 0.053 sq ft - 7.66 sq ins and were sized for 3.7 lbs/sec at a choked speed of 2100 ft/sec plus a 10% area allowance for "boundary", this is the same sort of setup I've used for all my engines except that the gas angle is ~33 degrees vs ~20 degrees for the other engines , this has reduced gas deflection a bit , but the good comp efficiency as well as nil backpressure on the turb should compensate .

I'll take your advice about a slower increase in fuel flow to prevent any "thermal choking??" type problems when I next run the engine with the 50/50 fuel mix , if theres minimal differences experienced from the fuel mods then I'll fit the larger jetpipe which will increase mass flow and get the NGVs working harder , just at the moment I'd like to try and find a reason for the large temp differences from one side of the engine to the other , I'm confident there aren't any blocked injectors as the insitu flow tests are producing the same flow vs pressure as when the fuel manifold was tested after manufacture.

Thanks again for your contribution ...............its too easy for me to get single minded about things , a fresh minds looks at things differently :-)

Cheers
John