T04 Gas Turbine - RESURRECTED!

[ripp]

Hi Ash,

do you know the X80 Micro Jetturbine?

Here are some pictures from the web.

The combustion chamber is already better but has also too large holes and too short evaporator tubes









regards

Ralph

translate.google.at

[ashpowers]

Hi Rip, No, I haven't ever seen the X80 before. Man that looks so much like my engine, LOL. Thanks for posting this!

OK, well I have modified the NGV and corrected the gap issue between the top of the vanes and the NGV outer ring. I also noticed that the turbine was sitting pretty high up off the ngv base, about 0.110" so I machined the shaft to drop the wheel down a little further to reduce that gap down to around 0.060" or so.

I have some SS plate here and what I have decided to do is for the time being, machine out a new exducer ring and bring the blade clearance down to an appropriate gap. The small reduction in the turbine exducer because of the previous wear should make virtually no difference in the performance of the engine - especially if I can bring that gap down to where it should be. Total diameter loss will be about 0.8mm which I dont suspect will pose much of an issue. The large blade gap losses would explain the hung start condition pretty much to the tee...

Ironically, even without using any of the maths at all and going completely off the cuff with the very first build of this engine, she at least would self sustain, LOL. I think I've been looking in the wrong place all this time!

-Ash

[ashpowers]

I've produced a new turbine exducer flange and questioning the clearance I have. I originally was shooting for the 0.25mm that Rip stated but that clearance just made me cringe. I couldn't help myself and I went to carving the flange out to a consistent blade gap of 0.45mm from the inducer blade tip through the profile and back to the trailing edge tips of the exducer. Everything looks fine and dandy but even at this blade gap it still makes me tense up. The bearing support is slop-free with the exception of that 0.0015" fluid damper clearance in the bearing sleeve, a total of 0.003" total possible displacement, which is negligible. That doesn't concern me. I'm just concerned with thermal expansion of the parts - moreso, the difference in expansion due to metal composition and heat exposure.

The turbine obviously takes a lot of heat but less expansion with inconel than with the 314SS exducer flange surrounding it... But the flange isn't getting as hot as the turbine.
The shaft is going to expand which will push the turbine further back. The axial clearance between the turbine from the top corner of the blade tips around the profile to the exducer bore is going to close as the shaft and turbine heat. The bearing housing will also expand, it is 6061 aluminum.. but this works FOR the axial clearance of the turbine-exducer flange. A pair of coupled expansion rates between these 4 different parts.

I know that the actual lengths involved with this expansion are mostly quite small but again, a 0.20-0.25mm clearance just seems a bit on the tight side. Suggestions?

[ashpowers]

OK, finally some good news to report. After the mods to address turbine clearances the engine is running notably better than any previous attempt to start her. She picks up revs significantly faster than she has been - right at the same acceleration rate I was seeing back in the day with her. Able to get her up to ~40KRPM and ~4psi with the propane whereas before I could only get up to around 20K. But bringing in the kero revealed an issue. Initially as the jetA came in she took well to it and gained speed up to around 55KRPM - much less flames out the jetpipe. But at that point she wanted to stall and wouldn't ramp up any more regardless of throttle setting but I wasn't getting the burst of flames out of the tailpipe and hanging the rampup. It was as if there just wasn't any more fuel being delivered.

So I shut her down thinking that the battery was dying but a quick voltmeter check negated that idea.

So I dropped the fuel line down into the tank inlet and ran the pump up. What I noticed was that after ~35% duty, I simply was not getting any more fuel out of the line. This line is jetted with a 0.018" orifice. I did this as an open line was just dumping far too much fuel when the pump comes up off of zero. But more notably as the pump was being throttled through its range, I could hear rather large changes in the audible noise coming from the pump through the steps. The analog output on the DAQ is an 8-bit channel so 256 steps across a 5V range which is plenty enough for a nice smooth dynamic fuel flow control for this purpose. Something is amiss in the programming. I did quite a bit of cleanup in the coding as well as added in all the code to handle the gauges but me thinks somewhere in the analog output code for the fuel system got buggered up. I do recall going through that code and taking out some old stuff that really wasn't necessary.... obviously got a little too carried away, LOL.

I really think this fuel pump is overkill for this setup.. Well, I know it is. Its a stock Nissan 300ZX twinturbo fuel pump that is capable of supplying enough fuel to support around 725 horsepower in a twinturbo 3L piston engine. It takes a certain minimum amount of power to get the pumps motor turning and once it gets over that "hump" and starts pumping, there's a large rush of fuel flow. Going from zero flow to that jump requires a jet to be installed into the fuel line to help dampen that initial burst but not too small else it will kick off the pressure relief valve on the pump at a certain point and you just aren't going to get any more out of her.

It is amazing just how lucky I was when I originally built this engine. It literally was like everything just fell together. I dont remember what jet size I put into the line 6 years ago but I do remember that day back then "Well, I'll just put a jet in there to limit the flow. Here - this size will do" and BAM, it just worked, LOL.

So, after discovering this issue I decided to run propane through the main fuel rail. Fired her up and she ramped up very quickly to 60K and ~7psi, pulled the air start back and she actually ran completely on her own... If I could have gotten the propane tank to give me more I wouldn't have hesitated. She was running right around 600C in this condition - I let her run like this for about 45 seconds - using my WD40 can hooked up to the bearing lube inlet and giving her a squirt every few seconds. Bearing sleeve exit temperatures were running around 200C if I recall correctly.

It is clear that I severely underestimated the effects of the blade tip clearance.

I am going to straighten out the code for the throttle control, already opened up the fuel line jet, and then I'll crack her open to take a look to see how things are faring.

Bearings still running smooth as silk.

Will try another start later this evening.

-Ash

[ripp]

Sept 13, 2014 11:26:24 GMT -5 ashpowers said:

I've produced a new turbine exducer flange and questioning the clearance I have. I originally was shooting for the 0.25mm that Rip stated but that clearance just made me cringe. I couldn't help myself and I went to carving the flange out to a consistent blade gap of 0.45mm from the inducer blade tip through the profile and back to the trailing edge tips of the exducer. Everything looks fine and dandy but even at this blade gap it still makes me tense up. The bearing support is slop-free with the exception of that 0.0015" fluid damper clearance in the bearing sleeve, a total of 0.003" total possible displacement, which is negligible. That doesn't concern me. I'm just concerned with thermal expansion of the parts - moreso, the difference in expansion due to metal composition and heat exposure.

The turbine obviously takes a lot of heat but less expansion with inconel than with the 314SS exducer flange surrounding it... But the flange isn't getting as hot as the turbine.
The shaft is going to expand which will push the turbine further back. The axial clearance between the turbine from the top corner of the blade tips around the profile to the exducer bore is going to close as the shaft and turbine heat. The bearing housing will also expand, it is 6061 aluminum.. but this works FOR the axial clearance of the turbine-exducer flange. A pair of coupled expansion rates between these 4 different parts.

I know that the actual lengths involved with this expansion are mostly quite small but again, a 0.20-0.25mm clearance just seems a bit on the tight side. Suggestions?

Hi Ash,


Those are good news ( 55KRPM )
du you have closed the large holes near the evaporator tubes?

The position of the shaft is held by the ball bearing on the side of the turbine wheel.
A preload of ball bearing is always present at the front ball bearing,
and the shaft tunnel should be made of normal steel because this expands
much less than aluminum.



www.jetmodelltechnik.de/Shop.htm
www.jetcat.de/

If the shaft tunnel must absorb a lot of heat and is made of aluminum,
the result can also look like this picture shows.




regards

Ralph

translate.google.at

[racket]

Hi Ash

Congratulations on getting her running :-)

Turbine clearances need to be "close" but our engines will run with larger tip clearances than axial wheels, I once got a "replacement" scroll for the TV 84 engine and found it was a bit harder to start and ran a bit hotter , when I investigated I found the wrong scroll had been sent , there was 2mm of clearance on the wheel's exducer :-(

When I make the TV94 turb bits I make the inducer to shroud axial clearance ~0.015" and the exducer to shroud radial clearance at 0.025" ( 0.050" diametrically) , the extra clearance on the exducer is because I use a mild steel shroud with a lower expansion rate to the wheel material .

Cheers
John

[ashpowers]

Well, the news gets better but still not quite all the way there. I've found my mistake in the code and corrected that. Opening the fuel jet up to ~0.025" has stabilized the fuel flow and much more controllable now.

I've run her up to around the 70KRPM mark, ~10psi and she is well behaved - only running about 675C TOT on straight jet-A. It is getting late here and didn't want to completely disrespect my neighbors so I shut her down and will have another round with her tomorrow to attempt a WOT run.

I have also obtained some 1/8" 304SS plate to make the scroll housing for the ST50 freepower turbine.

This will be my first freepower build and admittedly I'm not as versed in the thermo involved with calculating its design points.

Once the gas producer is running reliably I plan to temporarily attach a Nissan mass airflow sensor to the inlet of the engine and plot out a massflow graph for this engine so I can see exactly what she's consuming. Should be helpful in the design of the FPT section.

Does anyone have links/references to FPT design or how the calculations for the turbine stage in the gas producer are adapted to a FPT? I would imagine that there may be some modification to the ST50 turbine to adjust its flow properties to suit the gas producer which is no big deal. But obviously I will need to know what those modifications need to be before I can build the housing for it.

Looking through my Cohen & Rogers text I'm not exactly seeing a direct example of a freepower design. Can anyone point me in the right direction with this analysis?

Ironically my first Gas Turbine calculator spreadsheet has a freepower analysis section in it, LOL. Problem is, its been 6 years since I put that together and it was laid out for use on an axial freepower.

[rythmnbls]

Ash,

I have the 4th edition of Cohen and Rogers, there's a set of example calculations in chapter 2 section 2.2 for a gas producer and freepower stage.

Regards.

Steve.

P.S.

Very pleased to hear you got it running !

[racket]

Hi Ash

Generally its a good idea to have the gas producer sorted , then take some jetpipe total pressure and temperature readings to determine the maximum energy level thats available to the freepower, the size of the jet nozzle will give an indication of the mass flow once a rough calculation is done using those total pressure and temperature to give us a velocity from the jet nozzle , the thrust reading will also give us the mass flow after another calc., a quick cross check of velocity, density and jet nozzle area will give us another mass flow , the actual flow will be somewhere close to those theoretical calcs , ......good enough for ballpark freepower designing .

Are you going to use a gearbox on the freepower to allow maximum rpm or just direct drive with some kart sprockets and chain at reduced rpm ??

Is your ST50 turb wheel similar to the one I used in my 2 shaft kart jetandturbineowners.proboards.com/thread/40/2-shaft-turbine-kart-build ??

Inducer 119.5mm ,tip hight 14.5mm, exducer 89mm with a tip angle of ~35 degrees , ...........if so , then you shouldn't have any problems getting your 0.86 lbs/sec through it , but you'll need those jetpipe "numbers" to determine what sized "A/R" of NGV throat area and angles to use to optimise outcomes .

Cheers
John

[ashpowers]

Hi John,

I do have a Nissan mass airflow sensor that I had a company produce a very accurate transfer function table for - I will be able to quite accurately determine how much air mass this engine is moving and the TOT probe would fill in the rest of that equation.

My ST50 turbine looks just like the one in your pictures and I was intending to gear it down to try and maximize what I can get out of her. At least one step of gear reduction combined with a chain reduction to the rear wheel.

Do you have a solid mathematical approach to the A/R value of the housing? If so, I would definitely like to see that.

-Ash

[racket]

Hi Ash

LOL..............."solid mathematical approach ".........mmmmmmm , yes and no , lotsa guess work for the scroll , with final "tuning" being made via the dump jet nozzle sizing .

Its probably better to make a scroll plenum with a NGV stator to do the nozzling into the turb wheel , we have better control of area and flows that way ,

Once you have the jetpipe data from the gas producer its just a matter of designing the freepower NGV the same as when making the gasifier NGV ,.... the pressure ratio , density and velocities will be lower but we can still use the same mathematics.

Because you'll be gearing the rpm you can use a slightly different approach to my karts freepower where I was designing for a restricted rpm, so was trying to maximise mass flow through the freepower by using a higher/steeper gas inlet angle ( large A/R) , you can go with a lower angle to compliment the higher tip speeds , also your mass flow is smaller which better matches the lower angle and relative gas velocity into the turb inducer at higher tip speeds . ..............I ended up with a relative angle of ~50 degrees , you'll be looking at closer to 80-90 degrees , virtually radially inwards................my gas approach angle was ~37 degrees you'll be down below 30 degrees possibly 25 degrees above the tangent.

Cheers
John

[ashpowers]

OK, have been working at getting the engine running up to her max and she's just not having it. It seems that after ~80KRPM the temps just want to shoot sky high and I'm getting flames out of the tailpipe once again. =(

What I am going to do is I have ordered all of the materials to rebuild the combustion chamber just like what I had before that worked really well but I will be adding two additional J-tube vaporizers for a total of 8 and probably make them slightly longer than previously. In the old configuration from years ago she would run really well up to around 18psi but any more than that and I think I was at capacity of the vaporizers and the temps wanted to shoot up very quickly above that. I've gone down to 304 Stainless Steel Tubing, 5/16" OD, .257" ID, .028" Wall from the 3/8" O.D. that I was previously using. This new evapo tube will give me a flow area of 0.052 sqin per vaporizer X 8 = 0.416 sqin. Probably close enough to the 0.42 sqin recommended. I know you want about 6 times the inducer area for surface area in the evaps which will require these to be about 4.12" in length. My question on this is it would seem that by increasing the number of vaporizer tubes there will be less fuel moving through each and thus require less surface area for each evaporator - or in this case, less length.

I've also ordered 0.017" ID injector needles to build a new fuel injector ring with. I am currently running 0.027" with 6 injectors and I honestly think that is too large as it is. In the previous setup years ago I was only able to run the fuel pump at about ~50% duty before I was at her max output (TOT) so I think dropping down to 0.017" with a total of 8 should fare well.

[racket]

Hi Ash

Normally the combustion should improve as the rpm and air pressures increase , at 80K your compressor wheel would be working at its best efficiency , this is a strange one .

Cheers
John

[ashpowers]

Hi John,

That was my thought exactly. I still think it comes around to where the vaporizers are located and the fuel injectors being so shallow in the bores. There is just more and more air moving through the dilution ports adjacent to them as she gets spooled up more.

Fortunately I have exact notes on how the old combustor was laid out. I'm becoming a master at building new parts these days it seems, LOL. New materials should be here by the weekend so I'm looking forward to finally having her running.

Also, when I got her up to this speed I started to get a good bit of vibration out of the engine. She's nice and smooth up to this RPM. Got me to thinking that perhaps the oring supports between the bearing sleeve and the bearing housing may be at resonance at this higher RPM. I'm thinking I may want to add in an additional set of orings in the support system to help raise the resonant frequency if this is indeed what is happening.

What do you think of the revised vaporizer layout I'm entertaining? I'll be using the "J" style tubes like the old combustor had, just more of them and I'll put more length into them as well. Is my thinking correct on the surface area? What would you calculate to be the correct length for these J-tubes using that i.d.?

-Ash

[racket]

Hi Ash

Could the "vibration" have been caused by combustion/surge ??

Do you have a jet nozzle on the engine ??

One other thought about the airflow into your evaporators ............your turb wheel has always been a tad marginal in size , therefore the airflow rates could be down below "normal" , this would then require less hole area in the flametube , those large dilution holes could then be a "greater" problem .

A bit of extra evap length shouldn't be a problem, the 6 X inducer for total surface area is more a rough minimum area rather than a maximum , as long as the ends of the evaporator are at least 1.5 times the evap diameter away from an end wall they'll be OK

Cheers
John