Been a while - new project, T04 Gas Turbine.

[Ash Powers]

Got to working on the fuel ring and injector tubes and am pretty happy with how the assembly turned out. Not too much to look at but accounts for about two hours of my time.

The primary ring is 0.125" o.d. 304 stainless with 6 injector tubes of 0.032" o.d. and 0.025" i.d. I used a high-fill silver braze and flux with the small mapp gas/O2 torch. The part was final tested by connecting to a propane source - no blockages and no leaks! :)



Fitment:


I also got the turbine/shaft back from my turbo builder and have discovered a flaw in my turbine design. The stock shaft of the turbine has a quill diameter of 0.250" and the compressor wheel has the same i.d. to match. The original drawing didn't take the length of the quill into account and the shaft extension would have been an impossibility: with an i.d. of 0.250" and an o.d. of 0.250, the shaft would be non-existent! So the assembly was lengthened to account for this (as I will not be modifying the quill length) and the shaft diameter was bumped up to 0.750" (from the 0.625" it was previously).

The bearings are a perfect fit for the reground shaft:



And the new section drawing of the engine:

I've lengthened the bearing tube, combustion chamber, main body, and shown dimensions of the stock shaft and how it will fit into the extension. I also added a shroud at the back of the motor around the jet pipe which will be made from thin aluminum sheet for a little extra protection from touching a really hot part and it makes the back end of the motor look a little more appealing. :)



The tool steel for the shaft extension will be here next week along with the longer section of 5" pipe for the main body and the 6061 bar to make a new bearing tube. I'll have to remake some of the combusion chamber parts but I'm willing to give spinning another shot. I still can't wait for the PC board to show up though.. :)

[Ash Powers]

Made some more progress over the weekend. I picked up some steel rod from the local Home Depot to play around with making the shaft extension - just to get the "feel" and process down pat. After two attempts/approaches, I nailed it down but that was after some modification to the turbine/shaft itself.

Original:


Modified:


The new threads were cut with the shaft mounted between centers and a thread size/pitch of M6X1.0, left-hand. The shaft was then cut to length.

I ended up going this route as trying to bore a 6.35mm bore into the extension for the length it had to be was proving to be impossible while holding a truly straight bore. This modification also allows me to keep the engine within the original design length. Here is the rotating group with the shaft extension, bearings, compressor nut, and bearing load spring.





I also put the parts out to take a picture of them all: 27 parts so far (not including the fasteners).



I also spent some time putting together the rear jetpipe/flange and the jetpipe shroud. I have an appointment later this afternoon with teh TIG welder to finish these parts. :)





Here's "about" what it will look like.


The jet nozzle shroud will have a ring welded to the back edge of the main body to mount. The shroud is two pieces: the cone and the flange. The flange is a straight cylinder of about 3/8" length which will slip over the mounting ring. The ring is 0.065" thickness and will be drilled and tapped for some small stainless button-cap screws to hold the shroud in place.

Tool steel shipped yesterday. :)

[Johansson]

Beautiful work as always, too bad it makes my bike look like a pile of crap... *LOL*

//Anders

[Zippiot]

What is this jet going in?
Any specific plan or just for fun, cause it is way to beautiful to sit on a desk!

[Ash Powers]

I bought a Schwinn Stingray scooter about this time last year along with the T04 turbocharger to build the GT from.



I weigh about 150lbs, the scooter is about 15lbs and the turbine, accessories and fuel will be around 25lbs. 190lbs with ~30lbs of thrust should make for a fun ride. :)

[Ash Powers]

Beautiful work as always, too bad it makes my bike look like a pile of crap... *LOL*

//Anders

Bah - your jetbike is a cool project, I dont care what anyone else says about it. ;-) Looking forward to seeing it completed!

[Ash Powers]

The TIG welding of the nozzle shroud didn't go so well as I expected given the thin 0.020" 304SS sheet I made it from. I had the shroud mounting ring welded to the main casing and that went well (it is 0.065" thickness) and rolled a new shroud and spot-welded the seam. I drilled and tapped for three small screws to pass through the mounting ring and into small holes drilled into the shroud. The shroud is a very tight fit into the ring but fasteners will be required to hold it securely. Here's the pics:

Mounting ring:


The welded jetpipe and flange:


Shroud mounted and secured:


Side view: Can't wait to see it this way on teh back of the scooter. :)


The GT Control circuit board with the pressure sensors, thermocouple signal conditioners, PWM control, etc is scheduled to arrive tomorrow. :) Good times!

[Ash Powers]

Progress is a good thing, and quite a lot of fun too. :)

I recieved the PCBs late wednesday and had time over the past two days to start installing all of the components onto the board. Took a good bit of time to assemble - probably have about 2-3 hours cutting legs on caps and resistors and soldering everything down while checking continuity along the traces. Unfortunately when I put together the frequency to voltage schematics, I built them directly from the datasheet provided for the LM2917N of which a few examples dont incorporate a diode along the ground terminal and signal input terminal to draw the input signal completely to ground. Neither of the F-V converters worked at first until I discovered this and had to do a little patchwork to integrate a diode into the circuits. No big deal, but I wasn't too happy with having to hack up one of these pretty boards. :)

This GTControl board has both a 12V and 5V regulated power supply of which the 12V powers the DAQ device and the 5V powers the bluetooth device. The thermocouple circuits also pull from the 5V source and the rest of the circuits are driven at 12V. Two heatsinks were attached to the voltage regulators and are having no problems handling the load - the system complete is only pulling 0.38A total.



Right now each of the electronic components are just connected together straight with their respective connectors but I will put together some short interface cables for the bluetooth to DAQ and then from the DAQ to the GT Control board. All three items will stack on top of each other and take up a much smaller amount of space. I'm thinking a control unit casing around 5.75" X 3.1" X 2.0" - quite small. :)



A closeup of the GTControl board (missing one pressure sensor but I will install another one if I really feel thet need for an additional pressure measurement)


I've spent the last few hours working on the program for the PDA. I have calibrated the thermocouples for celcius measurement, calibrated the pressure sensor for psi measurement, calibrated the F-V input to read in Hz, and will have to wait to build the tacho circuit before I can final calibrate the RPM input. Both of the PWM outputs were tested and are swinging from 0% to 100% duty nice and smoothly and the potentiometer adjustment for base frequency works as planned. Still need to finalize the threshold safety routines to pull fuel if max EGT or RPM are exceeded as well as complete the startup and shutdown sequences.

I also assembled the engine just to check fitment of everything and clearances - BTW, I also machined the new shaft extension with the tool steel I recieved earlier this week :) Everything is fitting properly, clearances holding well. I will be sending the rotating group to my turbo builder for final balancing next week and working on putting together the high-power mosfet driver board (for fuel pump and starter motor control) as well as assembling the small optical tachometer circuit which will be mounted beneath the compressor shroud.

[Ash Powers]

Getting closer - I can almost smell the jet fumes. :)

Mounting flanges, EGT, and pressure sensor quick release port bosses welded. Stand for engine has been built and painted.





Optical tach IRLED/Phototransistor interface board schematic completed, board has been built. I employed a Linear Technology LTC1050 chopper amplifier setup as a comparator to produce a conditioned signal output for the RPM circuit on the GTControl board. There is a POT for both the IRLED's output and the LTC1050's input to allow adjustment for sensitivity. Currently there is about 1.4V of hysteresis built into this circuit's comparator so fingers are crossed that this will provide ample headroom for producing a clean output signal.









Compressor inlet duct was drilled, tapped, and threaded holders for the IRLED and phototransistor were machined. The IRLED and PT were mounted with epoxy.



The power MOSFET board for the fuel pump and starter motor has also been built using standard breadboard. Four NTE2985 MOSFETs, carrying a maximum of 30A each were used.





I also constructed a small potentiometer throttle input for the GT Control board using a 2K POT with a pair of 680 ohm resistors; powered by the 5V on-board power supply of the GTControl board and signal output fed into the AD channel 4. You can see this in the second image above. A thumb throttle will be constructed for the scooter which will tie into this POT for throttle control. I finalized the software for this input and it is also linked to the PWM output for the fuel pump through the software to allow "fly-by-wire" control over fuel flow. A threshold configuration page has been assembled in the software which allows for user-adjustable maximum RPM and TIT values as well as user-configurable fuel-cut parameters in the event those values are exceeded.

At this point I need to pickup some kerosene and turbine oil to fire her up for the first time.. Still waiting on some connectors to arrive for the DAQ to GTControl board connection so I can construct the final casing for all of the electronics to reside within but those should be here Monday or Tuesday. Looks like this week will tell all. :)

[mag]

Man, thats the greatest work I´ve ever seen! Can´t wait to see it run!

You´re realy a Professional GT builder, you should get this bluetooth stuff patented!

[Ash Powers]

Well, I fired her up for the first time this evening after getting everything setup on the bench. I dont have a glow-plug installed into her at the moment so I just used a torch in the tailpipe method. Turned up the propane feed a little, spooled her up a little with the compressed air until I got ignition and steadily increased the spool with more compressed air and more propane. I brought her up to about 25,000RPM where she was producing about 2-3psi of combustion pressure and removed the compressed air feed. I only ran her for about 20 seconds and slightly ran up the revs to about 30K or so and TIT was only about 650C which is promising given the low combustion pressure. I didn't want to run her for longer given that there is no lubrication feed to the bearings when running on propane, but it sure was exhilirating to see her run on her own. :) :) :)

I have to work out the fueling system though. My attempt to try and keep the service ports to a minimum is proving to be problematic. I cannot switch over from propane to kero using the same feed port so I need to find a solution to this.

Does anyone know how this is accomplished on the commercial engines? Is there a seperate feed tube for the propane and kero? If so, does the propane get fed into the evaporator sticks like the kero or some other location? I'm concerned with each of the evaporator sticks not getting warmed up enough if there is only a single propane port in the CC.

As always, any help is greatly appreciated!

We took video of the engine running, however, my wife didn't start the video until after I had her spooled up and running. I'll post video of the next run from start to finish though. :)

[multispool]

Ash,

You will need separate feeds!

Make up a T off the gas pipe with two 0.8mm needles feeding the vaporizer sticks, use roughly opposite sticks!

The kero feed can be 0.8-0.9mm needles feeding each stick.

Start feeding the kero once the EGT is showing over 100c, slowly at first, increasing the rpm at the same time. You should be able to shut the gas off at around 23krpm or so.

Don't run for long on gas only or the bearings will dry up and the pre-load stick.

If you are using a fuel solenoid, try to program it to shut off the fuel if the EGT is below 100c or rpm is below about 4krpm or you can end up with a nice fire! Else just keep vigilant...

[Mike Everman]

That's phenominal, Ash!

[Ash Powers]

Multispool: Thanks for the advice. I constructed an additional inlet port and propane feed tubes for two vaporizers, 180-degrees apart: two 0.032" o.d., 0.025" i.d. connected to a 0.125" service port tube coming out the front of the compressor inlet duct.

I got her spooled up on the propane (and it took everything the tank had trying to push through those two little tubes) and then turned up the fuel pump slowly. When the kero hits the CC, you can definately tell by a quick rise in spool and the EGTs settle down a little as CC pressure increases. I only ran her up to about 40KRPM this time with a TIT of 700C or so and about 4-5psi of pressure. Now that I can get her spooled more with proper bearing lubrication I felt good to take her up, however, once I try increasing the fuel flow after this point, I begin to hear a distinctive "growl" coming from the engine. I'm not sure if this is compressor surge or not, but I am mostly inclined to think this is the case.

Unfortunately when I designed the diffuser plate, I was not so much aware of the critical aspects of its design and more/less took a guess on the number of vanes. I'm using 15 vanes with a 24 degree seperation and with the knowledge I now have, I realize this is too much blade seperation. I figured I would give it a shot to see if she would run, and she certainly does, but compressor surge is a nasty beast and I wouldn't want to grenade this project. I'm going to further test to see if there is compressor surge by simple means of hanging a short 1" string from a stick and placing it near the compressor inlet to see if there are any back-pulses of air which will indicate surge. I'm also going to remove the rear turbine jetpipe shroud as this may be causing a sort of "whistle" (read: groan) effect as the fast-moving jet of air passes by it.

It could also be erratic combustion causing it so I'll need to go about this methodically to eliminate the possibilities and identify the culprit.

In the mean time, I had my wife take two videos of the 2nd and 3rd run of the engine for your entertainment. Sorry for the quality and lack of good lighting, but its there. :)

(These are about 3MB each)

http://ashspecz.com/Image%20Gallery/Gas ... t2-384.mpg

http://ashspecz.com/Image%20Gallery/Gas ... t3-384.mpg

[Ash Powers]

Hope you guys enjoyed the videos. :)

After considerable testing, I have come to find that the compressor is NOT in surge. The noise is due to erratic combustion. This only occurs when the propane fuel is fed at the same time the kerosene is fed. As soon as I bring the engine up to enough speed with the kerosene to turn off the propane feed, the surging noise goes away. Increasing the liquid fuel rate from self-sustain RPMs doesn't cause it to make this noise so long as the propane feed is turned off. Still have to see how the compressor performs at higher pressures and airflow levels though - I'm prepared to construct a new diffuser if necessary.

The test with the small ribbon on a string at the compressor inlet never showed any signs of surge and given the results from above, it is apparent that the noise was due to turbulent combustion.

I'm happy to find that the "J" tube evaporators are performing well enough to provide for quick combustion and prevent overheating of the turbine. The engine has been run several times solely on liquid fuel (after pre-starting with propane) within acceptable operating parameters. I had some concerns about the "depth" of the evaporator tubes into the combustion chamber (or lack thereof) but with the results of the tests, they appear to be sufficiently vaporizing the kerosene and not showing signs of excessive heat (they are still a medium-brown color). Obviously I'm not through the woods as of yet since I have only tested the setup to about 40-50% of is maximum, so fingers are crossed that they will continue to perform this well as I turn up the juice.

I successfully ran the engine up to 60KRPM this evening and CC pressure increased to ~7psi with a TIT of ~685C. This was achieved with a PWM output to the fuel pump at 20%. So far the IRLED and Phototransistor linked into the 4017 decade counter appears to be working well. We will have to see if the tach's performance stays consistent when the revs are turned up to peak operating speed.

At this point I have run the engine about a dozen times without flaw and slowly inching up in fuel flow and revs... running her between 1 to 3 minute intervals or so at a time. The bearings appear to be pretty happy thus far, although they aren't rated for RPMs much higher than what I have currently run her up to.

Tomorrow I will get the scooter from my storage facility and get to work on developing the plan for the engine mounts, fuel tank, electrical layout, etc etc.

GOOD TIMES!!! :)