GT3788VA Variable nozzle turbojet build

[mitch]

racket,

Ahh, okay, that makes sense with the blocked off compressor outlet. I will probably construct the "jet nozzle" like you explained, then.

I was also considering removing the compressor housing, bearing core and turbine wheel as one unit, and leaving just the turbine housing (without turbine wheel, of course) attached to the gas producer.

I'll do some tests on the oiling system tonight, and hopefully have the engine all together for a test run soon!

[racket]

Hi Mitch

Using a complete turbo as the "mule" for testing gas producers works OK as long as the freepower turbo's turbine stage is at least minimal size , I've helped with the design of freepower blowers using complete turbos being driven by a smaller DIY type gas producer, the "freepower" comp stage produced ~6 psi of pressure for an industrial application requiring lotsa CFM at a highish velocity from a mobile unit .

Cheers
John

[mitch]

Racket,
I will keep that in mind when I test the engine. The free power turbine stage is minimal size at the least, it is 118mm.

Unfortunately, testing of the oiling system did not go well last night. The new larger pulley is not concentric at all, and creates a lot of vibration and movement when I run the pump. From what I did see though, the new pulley does create a lower draw from the motor, and oil pressures were more stable, even with cold oil.

I will try to fabricate another pulley, this time using more precisely cut components, and more carefully welded to prevent warping. I have also decided to order the OEM oil drain tube for my turbo, as it is fairly cheap and will be much faster than trying to fabricate one, as my gas producer turbo uses a drain tube that presses into the housing and seals with an o-ring, no threaded port unfortunately.

I will upload a video of the oil system test I took last night, however the misalignment and vibration of the pump is pretty embarrassing, haha.

[mitch]

Quick update,
I've done a lot of work on the oil system lately, and am satisfied enough with it for testing for the time being. I'll upload pictures later, I ended up changing pulley ratios.

I did a quick "test fire" of the engine tonight using a leaf blower to spool the turbo, and just burned some propane through it (not enough to get the engine to self sustain). Just wanted to heat things up a bit and double check everything before I try to run it tomorrow.

When it ignited, there was a large green flame out of the turbine housing, which disappeared when I put the leaf blower back to the compressor inlet. Everything sounded fine and I let it burn propane for probably 30 seconds or so, then shut it down. There was no smoke or anything, but I noticed a bit of very dirty oil sitting at the turbine outlet, and some liquid sprayed across the face of the turbine wheel after it was cool. Could this just be a bad oil seal in the turbo? It never did this before, and didn't smoke while it was burning propane, even with 35 psi of oil pressure in the bearing core.

I'm not too worried about it, just wondering what it could be. I will video the test run tomorrow, hopefully it goes well with the new oil system and kerosene injector!

[racket]

Hi Mitch

If you've been doing a lot of lube system checking then there'll probably have been some leakage past the turb seal and into the bottom of the scroll, the leaked lube will have then covered the wheel when you did the spin test , its happened to me , so don't be too concerned.

All the best for tomorrows test :-)

Cheers
John

[mitch]

Racket,

I have been doing a lot of oil system testing lately, so I think you are right!

And thanks, hopefully I don't have the runaway problem that I had last time!

[mitch]

Tried to run the engine today, couldn't get it to self sustain due to a malfunctioning fuel pump. It would start to spool up, then the pump couldn't build much pressure above 2-5 psi, so it wasn't injecting much kerosene, at least not enough to self sustain. Other than that, the oil pump worked fine, maintained good pressure throughout the testing, although it leaks from the factory machined surfaces. I didn't have time to swap out fuel pumps tonight for another test, so I'll have to wait until next weekend for another test run.

[racket]

Bummer :-(

But getting closer

[mitch]

Anyone know how to calculate/estimate fuel flow requirements for a turbojet engine? I'm familiar with how it's done on a piston engine, given air consumption in cfm, and desired air fuel ratio. For a gas turbine, I assume it's similar, use the compressor map for your the turbo to find airflow in cfm, then use about ~1/4 of that ( I read somewhere that about 25% of the air going through a turbojet is used for cooling, not combustion) then use a standard air fuel ratio? I'm also not quite sure what a good a/f ratio is for a turbojet, I've read that it ranges from stoich to very lean. Any advice on this would be helpful, as I'm trying to roughly estimate the fuel flow and pressure for my engine based on my injector nozzle size.

[racket]

Hi Mitch

A rough estimate is the airflow in pounds per second , is the same as fuel flow per minute , a 60:1 A/F ratio , so if you flow 1 lb/sec of air you'll need 1 lb/min of fuel

Cheers
John

[mitch]

Thanks Racket,

I did some calculations using the information you provided, and using 70 lbs/min airflow (which is at the far right side of the compressor map) at 60:1 A/F Ratio, the required fuel flow is about 10 gallons per hour. Right now I have a nozzle in the engine that flows 5 gallons per hour at 100 PSI, and I will be doing most of my testing with a fuel pump capable of only 60 psi. This is not taking into account the gaseous fuel flow, which is propane. So, with a 5 gallon per hour liquid fuel nozzle @ 100 psi, being run at under 60 psi, I will not be any where near able to reach maximum power on the engine it seems.

After doing some testing at low "boost" pressures, assuming that EGTs stay safe, I will order a higher flow nozzle. I am also not sure if the flow rate vs pressure curve is linear or not for these oil burner nozzles, but I can figure that out with some basic tests.

[racket]

Hi Mitch

Flow changes as the square root of pressure , but you'll need to take into account the P2 pressure which will reduce the actual pressure drop across the fuel nozzle.

Assuming your 5GPH at 100 psi , if 30 psi P2 , then only 70 psi pressure drop , square root of 70 is 8.36 vs square root of rated 100 psi == 10 , therefore flow will only be at 0.836 of rated flow , so 4.18 GPH

Using your 60 psi pump and 30 psi P2 , gives 30 psi pressure drop , sq root 30 is 5.47 , divide by our 10 , gives us a flow of 0.547 of rated 5 GPH , so only 2.73 GPH .

Fuel at ~7 lbs/US gallon so only ~19 lbs/hr ,which will only be good for 19 lbs/min of air .

You might get the engine to a fast idle :-)

Cheers
John

[mitch]

Racket,
Seems like I will need to order a bigger injector for higher power runs! I have the fittings around to get the bigger pump working, I will try to start the engine again this weekend. I also have my EGT and "Boost" monitor working again, so I can record those numbers too.

[mitch]

Good news,
I test ran the engine this previous weekend, and it ran great! It started up and spooled quickly with the larger fuel pump. I ran it for about 1 minute, and throttled it up and down just a bit. It seemed to idle around 5 psi. Oil temps climbed a lot during the test, from about 30 degrees F to about 105 degrees F in that short testing time. However, the oil system worked very well, held good pressure (about 30-35 psi). The engine was very responsive to throttling the fuel, which surprised me. I have a video of it, and I will post it later. I was not able to collect EGT readings, but the turbine wheel looked safe while the engine was running. I looked at the wheel a couple times during the run, and it wasn't glowing or anything, which is good. I will try another run this weekend sometime, up to higher power levels.

[mitch]

www.youtube.com/watch?v=Sg6YQzZS8a4&feature=youtu.be