"Fat Boy" 12/118 thrust engine

[Deleted]

Hi John

Glad you found the problem :-) looking forward to more video and full power run

All Best
Andy

[smithy1]

Hmmm...Interesting that she appears to run well at the lower P2's but the issue arises when higher P2's are used. Interested to see how this turns out and what mods you use...I felt sure it was running really nicely at the kart track, goes to show how things change when the parameters change.

Also goes to show how little I know.

Cheers,
Smithy.

[racket]

Hi Smithy

Yep , a bit confusing , I just hope I've "diagnosed??" it correctly .

Perhaps the higher fuel pressure and flow along with the greater airflow rate and swirl within the can setup the right?? conditions for the aberration.............dunno .

Today I "adjusted" the injectors alignment so that they all spray through the throat of the evaporator waist and impinge on the inner wall just downstream of it , rather than upstream of the "dam wall" , once the engine is positioned with a slight forward lean to counteract the slope of the lower evaporators theres no backflow whatsoever , I think once the high speed air is going through the throats it won't matter what the alignment will be .

Its a simple "solution??" to see if it makes a difference, ideally I feel longer syringe needles and slightly larger size would be the way to go , but I'll try this quick fix first , if it doesn't work then I'll go up to a 21 gauge needle instead of the 22 gauge ones I'm using .

I've had a good look through all the flametube holes and everything still looks serviceable so I'll slap her back together and give her a quick run here in the backyard .

Cheers
John

[finiteparts]

Hi John,

The engine is looking great and I can't wait to see it going full-bore!

Did the adjustment of the injector location help?

I was wondering how the inside of the combustor looked and what pressure drop did you design for on this combustor?

It looks like the dome slip fits and is retained with several screws. Could the gap at that location be uneven and thus cause the top to be locally hotter?

Another thought is that we should be careful interpreting the oxidation coloring on the liner, since it is a time average of all the temps that happened on the entire run, not at one condition. Combustion designers often use thermal paints to evaluate the liner temperatures, but they do very specific tests so that they don't get off-condition effects baked into the thermal paint. They are at best a time average of the local heating, but by running to a certain condition and holding it, they can get a decent representation. If the test messes up and they run at some unintentional thermal condition, the thermal paint results are garbage. But it does seem suspect that you saw such a high EGT that may line up with the hot spot...suggesting that they did occur simultaneously.

I like your little necked region in the vaporizer tubes, but I wonder if it wouldn't be better placed at the outlet of the vaporizer tube? If you read the literature from Rolls, SNECMA, AVCO Lycoming, etc, you see that most of them do not believe that you can completely vaporize the fuel in the given injector length. They claim the main mechanism for the superior performance of the vaporizer is the airblast atomization of the fuel as the vaporizer discharge jet dumps into the main combustor volume. The free shear layer that forms as the jet discharges combined with the thin fuel film creates an excellent means to shear the fuel into very small droplets. With the heat that has been pumped in the fuel from the vaporizer body, the droplet vaporization time is highly reduced thus supporting a very quick and clean flame structure. So if you necked the outlets such that you got the highest flow velocity at the vaporizer discharge, you would drive a stronger free shear layer and help atomization of any fuel that didn't vaporize in the tube.

Finally, do you think the soot on the inner case and the fuel dribble could have just occurred at shut-down as the fuel that was in the manifold drained into the combustor?

Good luck!

[racket]

Hi Chris

Yep , things still look OK , just a minor problem .

I haven't had a chance to do another test run to test the changes , but having looked at the Solar T62 manual and how they spray their 6 jets of fuel through 6 convergent/divergent air nozzles into the flametube , I might just be producing something similar by squirting through the throat of my waist, the outlet ends of the evaps are a tad "reduced" in diameter by the pipe cutting process , but the D'ing of the tube produces a good bit of turbulence within the evap for fuel spread across the heating surface as well as mixing the air and vapour .

Yes , I agree with your thoughts on the "colouration" , it requires a bit of flexible thinking to accommodate all of the variables associated with failed/hot/dodgy starts , but the one thing that stood out was heat colouration of a section of the thicker NGV material not normally associated with transient heating as can happen with the sheet metal of the flametube .

As for pressure drop .............I give it 5% in calcs .

The soot in the outer can wasn't "at the top" , as one would expect from post shutdown dribble , it was to one side sorta indicating it was from the dribble out of the lower evaporators which I found during the water injector testing and the residual airflow swirl within the can during operation .

There shouldn't be much of a gap where the screws hold the endcap to the rest of the flametube as the inner wall will run hotter than the outer rolled edge like a slip joint , LOL, certainly not enough air leakage for a 300 C degree difference in temps , I'm pretty certain it was dribble back out of the lower evaps

Cheers
John

[madpatty]

Hi Racket,

Did you ran the engine again?
Did re-alignment of fuel injectors help.

I am curious to see what this re-alignment changes.

Cheers.
Patty

[racket]

Hi Patty

I haven't had a chance to retest the engine, life has been getting in the way :-(

When I get a chance I'll post results for you .

Cheers
John

[racket]

Hi Guys

Well its been a couple of months since my last test session, Anders efforts on the weekend have shamed me into action so I took the engine out for a test to see how the injector mods affected things.

Things are actually worse than before

In the July test at 0.5 bar P2 - 1.5 PR the TOTs were 639 C on one thermo and 570 C on the other , roughly a 70C difference ,not ideal , but manageable , todays test had 682 and 495 C , 187 C degrees difference , bugger , definitely not acceptable .

RPM was roughly the same at ~33,000 , temp rise in compressor again roughly the same, as well as fuel pressure of ~17psi .

At 1.0 bar P2 - 2:1 PR , todays temps were 658 and 473 C , but still roughly the same difference at 185 C degrees between thermos, RPM ~44,000 as before and fuel pressure 33 psi .

Once at 1.5 bar P2 - 2.5 PR the temps spiked on one thermo as before to 946 C whilst the other was in the low 600s so ~300 C degrees difference which is similar to last time where temps were 927 and low 600s ..........fuel pressure this time was a lot higher at ~85 psi vs 72 psi last test .

Now the only change I made to the engine was to adjust the syringe needle injectors so that they discharged through the evaporator waist preventing "splashback" , the changes meant that the fuel jet impacted the evap inner wall at a shallower angle rather than the previous ~45 degrees ,and further into the evap , the shallower angle may have caused the fuel to shoot through the evaporator quicker than previously allowing more raw fuel to discharge onto the flametube front wall where it may be flowing down the wall and collecting in the bottom where its finally evaporating and creating a hot streak , this is the only scenario I can think of which could have caused the larger variation in temps between the two thermos at the lower PRs.

I think I may need to construct a new fuel manifold with larger bore injectors , to reduce fuel pressure and fuel jet velocity , and longer injectors to pass through the "waist"with a 60 degree bend on the end so that the fuel is sprayed "harder" onto the evap inner wall so that it spreads around the bore before being dragged through the evap by the airflow .

A disappointing test run , but at least it produced changes which can only be related to the injectors , so they're the most likely problem .

LOL.....There were some interested spectators youtu.be/MgJsF1ykoXY
Cheers
John

[Johansson]

A pair of very hardened spectators since they stayed during the run.

A very interesting theory and something I will consider as well, the whole bent syringe needle arrangement has always felt a bit wonky to me since it is prone to clogging, difficult not to say impossible to clean and is easy to damage during overhaul since the silver soldered syringes are very brittle.

[racket]

Hi Guys

I've been doing some experimenting ..........again

Theres been a need to get away from a solid fuel jet entering the evaporator at a shallow angle at high velocity , fuel pressures need to come down to a max of ~20 psi pressure drop across the injectors , so a max delivery pressure of ~60 psi or so at a P2 of 40 psi .

I setup a system to test syringe flows vs pressure



Started with a 18 G X 1.5" needle , but this flowed too much at even 5 psi, ~250 CC./min , but it was a good strong needle being 1.2 mm OD so I tried bending over 1/3rd of the sharp tip to see what might happen , success a nice fan shaped spray was the result which projected out to the side of the needle tip at an ~70 degree angle to axial , even at only 5 psi the fan maintained itself .

A 21 G needle with its end bent over was inserted into the test rig and some tests conducted , at a flow of 30 CC/min the flow was in a screwing solid jet at minimal pressure, at 5 psi it produced a fan and 60 CC/min , 10 psi 100 CC, 20 psi 150 CC and at 40 psi 200 CC/min , this one would be a good candidate for the 12/118 where I need ~150 CC/min for each of the 18 injectors .

An advantage with the smaller gauge needle was the spray angle at ~80 degrees from axial .

Next a 23 G needle produced a flow of just 30 CC/min whilst maintaining a fan at a very low pressure , at 40 psi flow was 100 CC/min with extremely fine atomisation , just a mist , and at virtually a right angle to the needle axis .

With the injector to one side of the evaporator and the fan spray covering the evap bore , the incoming air should transport a fairly well atomised spray down the evaporator hopefully facilitating a greater amount of vapourisation with less raw fuel exiting the evap and potentially causing problems .

Some pics

[racket]

OOPS

Continued .....








At low pressure of <5 psi


At 40 psi pressure


Cheers
John

[gidge348]

Hello John, This may be a silly question, but has anyone experimented with small automotive style fuel injectors?

Found these www.ecotrons.com/files/ECOTRONS%20Fuel%20Injector%20technical%20spec.pdf that are designed for a 110cc scooter, seem to run at between 30 & 240 ml/min on 100% duty cycle but can be triggered to pulse about 0.5ms.

Cheers Ian...

[racket]

Hi Ian

There have been guys who tried automotive injectors on their turbo based DIY type engines , but their temperature rating can be an issue for us where they're subjected to combustion unlike when installed in an inlet tract.

In out "micro" type engine there simply isn't the space for them unfortunately , we're limited to something small , simple and cheap , the syringe needles are generally OK , but for some reason I've run into a problem using them on this engine , maybe its the fuel flow rate vs the length of evaporator , don't really know , but I think/hope this "spray" outcome from the syringe needle rather than a solid jet might be a solution.............LOL, time will tell , better get on and pull her apart :-)

Cheers
John

[racket]

Hi Guys

Engine removed from test stand and outer can removed .

3 evaporators have had their silver solder melted , one evaporator hanging loose and rattling around in its hole .

Time to sit and think

Cheers
John

[Deleted]

Hi John

OH..... :|could it just be a result of the injector problem

All the best
Andy