Turbine 1.0
Start date 11-25-2002

This project has been duplicated all over the net as it is but it's just too cool for me to not do my own example.  Of course I would rather innovate than imitate so I hope to bring some new ideas to the table, even if they aren't any better than what's' already out there.

First, a little introduction.  The basis for this jet turbine is the automobile turbo charger that is used as a compressor for the system.  In the most simple since a jet turbine consists of the following:
1. Combustor
2. Compressor
As the name implies a combustor is where combustion takes place and pressure is created.  In our case we will be burning propane from a pressurized container and using the turbo to create the compressed incoming air charge.  By feeding propane and air to the combustor we can create combustion, which results in pressure, which is the force that is exerted on the turbo inlet and causes the secondary side of the turbo to compress atmospheric air into the combustion chamber.  In this way we burn fuel, which creates pressure, which spins the turbo, which creates compressed air, which allows us to add more fuel..  All things theoretical we could continue this cycle until the turbo reached a critical mass, that is to say, it has reached it's maximum flow volume.

Most combustors are made via the following model.  An outer housing and an inner 'flame tube'.  The flame tube is drilled with various holes to allow for proper air mixing and to introduce some cool air into the turbo's''' inlet to prevent overheating.  The air enters via the holes in the flame tube, which if designed properly, creates a proper air/fuel mixture somewhere toward the middle of the combustor and allows the bottom holes in the flame tube to deliver the excess air to the turbo inlet.  Usually an automotive spark plug initially ignites the mix but once combustion occurs it self sustains after a few seconds.  In my design I was hoping to create what is commonly known as a 'motor jet' to act as the combustor.  The combustors purpose is to provide an area for the fuel/air to combust and create pressure which in turn, exerts the force on the turbo inlet.  The reason for a combustor with a flame tube is to help keep the outer wall of the combustor cool, and to allow some unused air to enter the turbo and cool it.  My hope is to use the motor jet to create the force from combustion.  The design is not very different from the above model.  Simply remove the flame tube in the center and that is a motor jet.  A serious consideration is the heat that will be created and directed into the turbo.  I am hoping that the cooling system I have envisioned will be adequate to cool the turbo, oiling system, and exhaust cone.  We will see....

Here are some pictures of preliminary testing on an old piece of automotive exhaust pipe, propane bottle, and compressed air from a home air compressor.

Here is the exhaust of the motor jet.  Mike is running the engine while I snap pictures.  In his right hand is the propane bottle and flow control valve.  In his left hand is the air compressor nozzle which is controlling the air inlet into the motor.  By tilting the air inlet to an adjacent angle we can create a vortex and get the air and fuel to mix evenly.  In the lower picture the vortex is slightly more visible, although I couldn't get a straight on shot due to the rather substantial heat generated.  I bent the end of the outlet into a small cone to created some restriction and in turn allow for some combustion pressure.  I found that this actually caused more turbulence in the exhaust outlet than it benefited by creating some backpressure.  The next step will be to make a smoother transitioning cone/nozzle and monitor the exhaust to see if turbulence is lessened, and combustion pressure is increased.

In case it isn't obvious, this is sustained burn.  Not a one time boom or series of booms like a pulse jet engine.

11-26-2002 Got some more exhaust tube and various parts donated by the local Bellevue Midas ( that's Bellevue Washington).  So new combustor is in progress.  3 inch diameter and approx 18 inches long the new combustor will also use a better control valve for the air source,  but I am going to reuse the fuel injection portion since it worked so well.  In fact, when the fuel was valve full open I found turning the bottle upside down and feeding liquid propane caused a richer air/fuel mixture and was more fuel than the incoming air supply at 90 psi could handle.  The nozzle used to control the air inlet has a very small outlet hole and was probably a large reason that not enough air could enter the combustor even at this higher pressure.

  Pictured left is the 'fuel manifold' that distributes the propane inside the combustor.  On the left side you can see the original hardware used, from the top threaded portion of the finished unit is a hole that feeds all the way through to the six holes in the base of the unit. The bottom bolt is just to cap it off and make the fuel travel through the lateral holes instead of straight out the bottom.

11-30-2002 Well after 5 days of playing with the motor jet design, I have come to the conclusion that going along with the conventional combustor/flame tube design will ultimately be more successful.  Not to say that what has been done up to this point was a waste, since basically this has helped determine afterburner design for the finished project.  Today I built the combustor and flame tube and tested it.  It worked surprisingly well the first time.  I noticed a drastic drop in outlet temperature which was a serious concern I had with the motor jet, since I wasn't sure what 1000 degrees F would do to the turbo inlet.  After tinkering with the holes in the flame tube I got a nice consistent burn and could even see all the way into the back of the combustor where the combustion started.  Pictured right is the new combustor and flame tube.  At the right side end of the combustor pointing down is the compressed air inlet.  This proved to be a bad way to simulate the compressed air coming from a turbo since the turbo puts out higher volume with less pressure and a shop air compressor does the opposite.  At the far right of the combustor is the propane injector and the hose that connects it to the propane bottle.  Fuel is controlled by the valve at the end of the hose.  Air inlet is controlled by a ball valve in-line with the supply inlet ( you can barely make it out in this picture).  This design also has and ignition coil and spark plug running off of a 9V battery to ignite the mixture.  The gauge at the top is to measure combustion pressure.

   1-13-2003  The ignition system uses the basic automotive coil found on older cars and can be had for less than $10 at your local auto parts store. Mine is off a 1982 Volvo 240.  To modulate the coil I have seen lots of various ideas from a 555 timer circuit down to a basic momentary switch.  For those of you confused on how a coil creates a spark, here is a brief description.  The coil is made from an iron rod in it's center and is surrounded by two separate windings of wire and is then enclosed in a metal case.  The first winding is known as the primary and is where the input voltage (in this case a 9v battery) is applied.  The secondary windings are where the spark (high voltage) is created or induced.  The output of the secondary windings is connected to the spark plug and also to ground.  As we energize the coil by applying voltage from our battery we are creating and electromagnet, this is of no use as it is so we break the ground source from our battery.  This in turn causes the magnetic field to collapse, as it collapses it falls through and into the secondary windings.  Now is a good time to mention that the secondary windings have about 1,000 to 10,000 more windings than the primary.  This difference in windings is responsible for the increase in voltage from 9v to around 9,000 volts or more.  This high voltage is enough to ionize the gap of our spark plug and create a small spark that is capable of igniting our fuel mixture. 

    So as you can see all that is needed is a way to cycle (turn on and off) the ignition coil at a rate that suites our needs.  Now if you've looked over my site at all you have come to realize that I like to make stuff quick, easy and cheap, of course that means no complex 555 timer circuits and no lame manual switches.  For my ignition system I used a flasher relay from an automobile to cycle my coil, very easy and less than $5 at a parts store.  By using a 3 pole flasher we can do this easily and still adjust the speed at which it cycles.  The 3 terminals on the flasher are power, ground, and control.  The first two of these is obvious, the third is connected to the ground side of the coils' primary.  The coil's primary positive terminal goes straight to battery voltage.  The flasher will start to click on and off at the normal speed we are all used to when we drive our cars and use the turn signals.  If this is not fast enough for you, then you can open the flasher unit and modify the resistors inside to speed things up.  Upon inspection of the flasher you will notice that it is actually a chip driven timer much like a 555 circuit.  By reducing the resistor values we can speed up the flashers cycle and get more sparks per second.

2-12-2003  Today Mark Seagul from the Mercedes Benz franchise of the dealership I worked at donated a used but good working garrett T3/T4 turbo.  Inspection of the turbo looks promising, almost no play in the turbine shaft, good sealing waste gate valve, and not much rust beyond the surface.  This turbo has no water cooling passages in it which makes me wonder how well it will take the heat we are going to be creating when running it.  Guess we'll wait and see.

2-22-2003  Ok, so here's the mostly finished unit.  Thanks again to Steve at the Midas shop for donating all the pipe and doing the bending at no charge!  As you can see the turbo now has the combustor bolted to it's exhaust intake and the induction pipe has been welded to the combustor from the outlet of the induction portion of the turbo.  The outlet nozzle is bolted to the exhaust outlet of the turbo and will hopefully provide some backpressure to help the thing run and also provide a some direction for the thrust generated by the afterburner.  At this point the only thing left is to build the oiling system.  I have had to do some thinking on the oiling system since I have seen quite a bit of other peoples experimenting on homebuilt turbines and the oiling varies from gravity feed funnels to full on oilers running at 20 psi.  I'm still looking for parts for mine.  So far this project has cost $0 and I'd love to be able to say I built it for exactly that amount.

Just in case you can't figure it out on your own, here's how the turbine works according to the picture above.  The turbo inlet takes in the outside air and compresses it via the inlet fan blades.  This compressed air is routed to the combustor where it is mixed with fuel (LP) and burned.  The combustion pressure, heat, and expanding gasses pushes against the fan blades in the exhaust portion of the turbo which is connected to the intake fan blades.  This in turn causes more air drawn in to the turbo and more combustion creating more thrust and the cycle continues.  The after burner figures into this whole thing as basically a liquid fuel rocket on the back of the turbine.  Hot exhaust gasses light the incoming liquid propane which begins combusts in the outlet nozzle just ahead of the taper toward the end.  This creates pressure in the nozzle, when it escapes the restriction at the end of the nozzle it creates additional thrust.  Of course you might as well forget fuel economy at this point because you'll be burning propane at an incredible rate.  I like the way Patrick Arnold refers to his turbine as the 'Whoosh master 2000 rotary blowtorch'.  Next step will be building a test platform for the thing to run on.

3-1-2003  Had a small setback today that cost some time.  I managed to strip out the spark plug in the combustor when removing it.  So I used a hole saw and opened it up and inserted a blank nut that I threaded to match the plug threads 14mm by 1.25 - welded that sucker in and is now good to go.  One other change that I made was on the ignition system.  The flasher relay was sucking so much power that after 2 min of running the battery would be flat. So back to the drawing board and for the sake of solidity I built a 555 timer circuit to run at 5 HZ and used a TIP120 power transistor to switch the coil.  Now I get about an hours worth of good spark on a single 9v. 

Ok, so here is the schematic for the igniter circuit.  As you can see the basic 555 timer oscillator circuit is used due to its simple design and low cost.  The resistor values may not be the exact ones I used, I drew this from memory.  I used a 9v battery for power but did regulate it down to 5v for consistency.  The LED on pin 3 is driven off the high side of the signal while the coil is actually fired off the low side of the signal, the main purpose of the LED is just to tell whether the circuit is on or off. 

We have rigged a temporary oiling system so Sunday (march 2) Mike is coming over and we will try to make it run.  Cross your fingers.

We are getting close to the part of the project where people can get hurt or killed so if you're not sure what you're doing STOP NOW and take up needle point.

3-3-2003  Picture right is the semi-completed turbine on its test stand.  I say semi-complete because the oiling system is quite temporary and the fuel supply is coming from a small 16oz propane bottle which is barely adequate.  The oiling system is comprised of a reservoir on top and a drain tube and collector at the bottom.  I am running synthetic ATF for oil since it has antifoaming additives and has a low viscosity at room temp.  This has proven to work well since the oiling is gravity fed at this time.  This system also limits how long it can run since the hotter it gets the more oil the turbo needs.  Now the propane supply I am using at this time is only allowing me to run the engine for about 15 seconds before it gets too cold and loses pressure and floods the engine.  The feed line from the propane bottle is rubber and turns rock hard as it reaches about -10 degrees F which is about 10 seconds after it is at full flow.  The propane bottle is turned upside down to help get enough fuel volume into the combustor, although this could be flooding the engine since after it stops you can here boiling in the combustor which I assume is raw liquid propane.  More testing will definitely be necessary to determine how to make it run well for longer periods of time.  Big thanks to my neighbor Eric the Viking for donating his unused 20lb propane tank.  I already have the regulator, line and fittings for the larger tank the only mods will need to be on the combustor fuel inlet.  

So today I realized that in my older age (28) I am getting either too cautious or just chicken.  I was actually a little scared to try and light this thing up.  I had tried to run up the turbo with a high speed drill on the input shaft and turn on the igniter and open the fuel to get it started but this proved to be fruitless since the air volume at that low speed (8000 RPM) is not enough to start the engine, but it did make some loud bangs and make some neighbors poke their heads out the window.  Finally I broke down and did what everyone else has done to start theirs and got out the leaf blower to spin up the turbo to the needed 20,000 RPM to start.  After that starting was easy and smooth.  With the igniter turned on, I spun up the turbo and turned on the gas and very smoothly and sounding just like a real jet engine she came to life, of course only for about 15 seconds but man, what a show.   I think I am running too rich since a yellowish flame exits the exhaust nozzle about 12 inches back.  

3-4-2003  Today I boldly took the turbine to the Volvo training center with me to attend training for 2 days.  Surprisingly when demonstrating it, I was able to get it started with ease and it stayed running for about 2 minutes.  I did end up holding the bottle right side up which is most likely why.  I was able to  increase throttle to about  3 to 4 psi of turbo boost when running.  Oil flowed nicely with no pressure through the turbo.  One of the fellows I work with (James Boggs) graciously donated a 30lb propane tank that was almost 100% full.  So I will bring in the regulators and lines tomorrow so we can use it.

3-5-2003  Well......Today had its good parts and bad parts.  The turbine, with new fuel source, lit off very nicely and accelerated to 5 psi of boost with no problem but as it ran we noticed no oil was flowing down from the reservoir on the top.  We noticed this a bit too late and after 3 or so minutes of running the turbine came to an immediate stop with a loud screech.  Once it cooled off I could see that the shaft had noticeable play which meant the bearings are probably shot and may not run again.  Later once it had completely cooled I checked again and it didn't have quite as much movement as originally anticipated.  Guess I'll take it home and see what I can do.  Fortunately we were able to really get some great sound and heat from the turbine and got a feel for running it. 

3-6-2003  Today I bodged together a temporary but viable oiling system with an old fuel pump and coffee can.  Can control up to around 5 lbs of oil pressure so this should solve our oiling problem.  I turned on the pump and spun up the shaft with the leaf blower and it seams to spin smoothly and without vibration so we will hope for the best on the next run.  I tried to start the turbine today but with no success.  I wasn't sure what the problem was at first but upon further inspection I found that the igniter was not working reliably, some smoke coming from the resistor inline with the TIP 120.  So off to the local Radio Shack for some more parts and on the way there came up with a more simple design for igniter using less parts and less $.  Below is the schematic.

As you can see the blinking LED is the control and the TIP 120 handles the current, we still use the auto coil for voltage amplification.  The resistor inline is 10k ohm and limits current through the LED.  This doesn't spark as quickly as the previous design (about 3 times every 2 seconds) and I'm not sure if this will present a problem or not.  Testing so far has created nice spark that jumps a greater gap then before.  

3-7-2003 Today more efforts on firing up the turbine.  Had some limited results today.  The turbine never fully fired and ran but it did run for about 2 seconds before I flamed it out on accident.  The new igniter looks like it will not work very well.  I had great difficulty getting the turbine to fire just the one time and this took quite a few attempts.  I have a feeling that the slowing spark rate is the problem.  Upon examining the old circuit using the 555 I found it was really sparking closer to 20 times per second not 5.  I think what is happening is that once the correct fuel mixture is obtained in the combustor and it lights from the spark plug it is blown out before the next spark occurs and keeps it burning.  The reason it's being blown out is because the combustor air inlet creates a rather turbulent condition in the combustor so fuel mixture varies slightly throughout the flame tube and may be preventing smooth ignition.

3-11-2003  Today I rebuilt the igniter back to the 555 timer circuit.  This time I used a 10k ohm resistor inline with the TIP 120 input so that it doesn't fry like last time and also upped the voltage from regulated 5v to 9v direct from the battery.  So if your using the first schematic be sure to change the 500 ohm resistor to a 10k ohm.  Retesting the igniter showed that this was the way to go, the faster spark rate was definite necessity for good ignition.  I didn't really get a chance to fire up the turbine since it was late and the kid's were already asleep but I did spin it up with compressed air and I did get a few consistent 'booms' out the back end before turning in for the night.  I also need to get a new spark plug since the first one somehow cracked almost in half and the only replacement around was a shallow reach instead of a deep reach and I'm sure this isn't making ignition any easier.  I normally lived a charmed life when it comes to my pet projects but I swear this one is cursed, every time I touch it I break something.  Tomorrow I'll put in the new spark plug and try to get the turbine up and running.

3-12-2003  Today I replaced the shallow reach plug with a deep reach and tried firing up the turbine and surprisingly it fired right up with no problems, very smooth. Ran the turbine upto 7 psi of boost and it was SCREAMING, highest I have ever run it.  What can I say, the only way to describe the noise was cool!  So now that I know the turbine isn't completely toast I will build a permanent oiling system and then run the thing until I either max out the LP tank flow or completely burn out the bearings.  Should be fun to see how much boost I can really get out of it. 

3-23-2003  Since last time, I have run the turbine about 4-5 times after taking it to work for the guys to see.  I have run the turbine upto a max boost of 17psi and that is probably as high as I will go until I get new bearings for it.  I have to keep it intact for the guys at my new job in Oregon to see, then maybe I'll run it till it blows and then get new bearings or perhaps build a two stage twin turbine.  As it is the bearing are BADLY worn and very loose.  At 17 psi of boost the outlet turbine will sometimes touch the outer wall and emit a large cloud of rust dust.  The turbine also passes quite a bit of oil past the bearings and is burnt upon exit.  Stay tuned,  I'll be back in action sometime at the middle of April after the move!!

4-28-2003  The move to Oregon is now finished and I decided to run the turbine today for the guys at work.  It fired right up but would not run on its' own without some form of pressurized air on the inlet.  I thought the bearing drag might be the cause for this but upon further inspection I found that the exhaust impeller has large chunks taken out of the fin blades.  The impeller will have to be changed with a new one and the bearings and seals need to be redone.  Luckily the turbo I've been using is a Garrett and there must be millions of the things made because I was able to find a rebuild kit almost immediately after searching the net.  Pictured left is the damaged exhaust impeller, as you can see the lower blades at the 4 & 5 o'clock positions have big chunks taken out of them.  Looks like a turbo rebuild is in order....

8-29-2003
Finally!!!  The new bearings and new impellar are here and installed.  Fired the turbine up today and it ran very smooth and with much greater power than before.  Big difference in how much drag the old bearings had versus the new ones.  Ran the total boost upto 15psi before shutting down.  Once shut down the impellar spins for about 2 minutes longer than before.  So all things good and now that it runs the guys at work don't think I'm full of hot air.

10-29-2003  Good runs the last few times the turbine has run and this time got some decent video shot while we were at it.  I finally have finished the afterburner prototype with the new combustor.   The old combustor had enough oxygen left over for the afterburner to run but the new design is too effecient.  So I took the pressure output port that normally runs the wastegate and ran it to just before the afterburner fuel inlet.  This allows self governing and a nice ratio of air flow based on turbine speed.  Hope to test it soon.

For now.....check out the video Turbine.WMV