Been a while - new project, T04 Gas Turbine.

Moderator: Mike Everman

Post Reply
Ash Powers
Posts: 176
Joined: Sat Sep 04, 2004 7:17 am
Antipspambot question: 0
Location: Edgewater, FL
Contact:

Been a while - new project, T04 Gas Turbine.

Post by Ash Powers » Thu Nov 30, 2006 10:19 pm

Its been some time since I've played with the GTs. I purcahsed a T04 turbocharger and a Schwinn Stingray scooter earlier this year with the idea of building a jet-powered scooter.

The idea is to use just the rotating group from this turbocharger and build the other components around it for an annular combustion chamber design and a radial NGV for the turbine. I cast a few of the pieces, machined them and am currently working out the details of the NGV blade angles and the compressor diffuser blade angles. I ordered a 5" diameter 304SS tube for the outer casing today and Boca Bearings finally shipped out the ceramic bearings this afternoon. Stainless sheet (309) was ordered today as well to produce the flame tube and the wooden pattern was cut out last night for the front compressor diffuser plate. Still need to make the wooden pattern for the inlet nozzle for the compressor, but I'll do that after the diffuser cover has been cast and machined.

The compressor is a T04B 60-1, 3.00" major dia, 2.324" inducer.
The turbine is a T04 P-trim, 2.92" major dia, 2.54" exducer.


Here's some pics:

Cross-section drawing:
Image

Casting the diffuser:
Image

Machined and mounted diffuser:
Image

Inner NGV machined:
Image

Turbine exducer case:
Image

Hot Parts:
Image

Turbine/NGV/exducer case
Image

I am seeking some assistance in determining the blade angles for the diffuser as well as for the NGV. I'm hoping to produce around 30-40lbs of thrust from this turbine which will be ample power to drive the scooter with someone on it - probably more than enough. :)

Mike Everman
Posts: 4928
Joined: Fri Oct 31, 2003 7:25 am
Antipspambot question: 0
Location: santa barbara, CA
Contact:

re: Been a while - new project, T04 Gas Turbine.

Post by Mike Everman » Fri Dec 01, 2006 4:44 am

I can't contribute, but as a professional machining critic (design engineer), I must say, nice work!
Last edited by Mike Everman on Mon Dec 11, 2006 6:11 am, edited 1 time in total.
Mike
__________________________
Follow my technical science blog at: http://mikeeverman.com/
Get alerts for the above on twitter at: http://twitter.com/mikeeverman

Ash Powers
Posts: 176
Joined: Sat Sep 04, 2004 7:17 am
Antipspambot question: 0
Location: Edgewater, FL
Contact:

re: Been a while - new project, T04 Gas Turbine.

Post by Ash Powers » Tue Dec 05, 2006 1:18 am

Spent some more time building the casting patterns and doing some machine work.

Cast and turned the front diffuser plate, the compressor inducer nozzle and the inlet duct. The compressor inlet nozzle and the inlet duct were cast as two seperate pieces and the inlet duct has a lip-fit over the compressor inlet nozzle. I will be using some small fasteners to bolt them together and the wall thickness is enough to allow for boring/threading and mounting of an RPM pickup.

Image

Closeup of the cold section housings:
Image

Hot section components:
Image

I also machined a new NGV inner guide plate with the recess to keep the bearing tube from soaking a ton of heat. A cover plate was machined to fit (pictured on the far left with the fasteners) that will be attached to the NGV during assembly. I also purchased a 6" vertical/horizontal rotary table for the mill, a slitting saw, and arbor for the blades. This will be used to cut slits into the NGV and compressor diffuser plate for insertion of the vanes, which will be seam welded on the backsides of the components to seal things up and hold them in place.

Image

A new bearing tube will need to be machined and I am leaning towards using stainless rather than aluminum. Should be able to make it slightly smaller and it wont be as susceptable to thermal deformation/failure. The rest of the machine tools will be here later this week and I'll knock out the diffuser and NGV, and I should have the SS casing to start on by midweek. It will be a 3-piece design where the rear section will have a 45-degree step into the center mounting plate for the NGV outer housing and the turbine exducer pipe.

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

re: Been a while - new project, T04 Gas Turbine.

Post by larry cottrill » Tue Dec 05, 2006 2:59 pm

As I've said before, I don't know squat about turbos, but I recognize that this is beautiful work. Very cool indeed!

L Cottrill

mk
Posts: 1053
Joined: Mon Jan 19, 2004 8:38 pm
Antipspambot question: 0
Location: FRG

re: Been a while - new project, T04 Gas Turbine.

Post by mk » Tue Dec 05, 2006 6:59 pm

Some nice piece of workmanship! Cool.
mk

Ash Powers
Posts: 176
Joined: Sat Sep 04, 2004 7:17 am
Antipspambot question: 0
Location: Edgewater, FL
Contact:

re: Been a while - new project, T04 Gas Turbine.

Post by Ash Powers » Sat Dec 09, 2006 7:11 pm

Things are moving forward nicely. I recieved the rotary table, slitting saw, and arbor yesteday and couldn't keep myself from setting it all up and going to work at it. This process worked just as planned. I cut out the NGV slots as well as the compressor diffuser slots.

After a good bit of time putting together an excel spreadsheet to allow quick and easy calculations for the NGV properties, I nailed it down to 15 vanes @ 0.065" thickness at 22 degrees to produce the required 3.3 square inches of area leading into the turbine inducer blade tips.

Image

Image

Image

I also recieved the 5" stainless pipe for the outer casing and did the final machining of the front compressor diffuser plate for a snug fit. The diffuser plate was cut with a 2-degree taper and it will be later notched for installation of an oring to provide an air-tight seal between the two parts.

Image

Looking down the bore:
Image

I have the stainless 0.065" plate here to make the NG vanes but still need to source some 0.065" aluminum sheet for the compressor diffuser vanes. Should be able to get ahold of that easily though and will machine those parts and have them TIG welded next week.

I'm still waiting for the T04B 60-1 compressor wheel I ordered to show up so I can create the final diffuser vane pattern but it should be here soon.

Mike Everman
Posts: 4928
Joined: Fri Oct 31, 2003 7:25 am
Antipspambot question: 0
Location: santa barbara, CA
Contact:

re: Been a while - new project, T04 Gas Turbine.

Post by Mike Everman » Sat Dec 09, 2006 9:27 pm

Looking good. I'm watching your progress with great interest!
Mike
__________________________
Follow my technical science blog at: http://mikeeverman.com/
Get alerts for the above on twitter at: http://twitter.com/mikeeverman

Johansson
Posts: 1161
Joined: Sun Nov 21, 2004 9:42 pm
Antipspambot question: 0
Location: Northern Sweden

re: Been a while - new project, T04 Gas Turbine.

Post by Johansson » Sat Dec 09, 2006 10:08 pm

Very nice work!

//Anders

Ash Powers
Posts: 176
Joined: Sat Sep 04, 2004 7:17 am
Antipspambot question: 0
Location: Edgewater, FL
Contact:

re: Been a while - new project, T04 Gas Turbine.

Post by Ash Powers » Mon Dec 11, 2006 5:50 am

Spent some time making up a template NG vane to cut the final parts from SS out of. The template was made with about 0.015" additional "size" to account for small variations when the vanes are final-welded and I will setup the grinder on the lathe to get the final contour for a snug fit with the NGV's outer case.

Image

You can see the scribe lines on a couple of the vanes out at the rounded top corner - a rough finish for the moment but it is easier to take material off than try to put it back on, right? :)
Image

I also stumbled across an oring from a turbocharger that was a good suit for sealing the front diffuser cover to the main body. A little bit of time with the diamond wheel to make an oring cutting bit and a few thousandths at a time to get the depth just right and its a perfect fit. I used some parker's o-ring lube to keep it from sticking when inserting the plate into the main body and she slips right to seat with just the right amount of force to make me happy. This oring is used in a Garrett turbocharger to seal the compressor housing to the cartridge, so it is perfectly suited for this application. I mounted the main body into the lathe and sanded a short taper into the inner wall to prevent the edge of the body from nicking the oring during installation.

Image

After installing the part into the main body and removing it as seen above, there is a good 0.125" worth of sealing "margin". The entire cold section of the engine will not be solidly mounted to the main body so as to allow for thermal expansion/contraction without putting any of these loads on the internal components, and 0.125" is more than enough to compensate for this.

Image

Image

Image

I left a little bit of a seat on the plate (0.200") to have a face to mount a cover for the front of the engine to give it that "jet engine" look. I'll make up a wooden template to spin the aluminum to form this component sometime this week. Looking forward to doing that as I have never spun any aluminum pieces and from what I take of it, requires a bit of elbow grease but is a pretty fun process especially when turning out parts for jet engines. :)

Ash Powers
Posts: 176
Joined: Sat Sep 04, 2004 7:17 am
Antipspambot question: 0
Location: Edgewater, FL
Contact:

re: Been a while - new project, T04 Gas Turbine.

Post by Ash Powers » Tue Dec 12, 2006 1:45 am

I've been kicking around some ideas for building an engine control
unit for my gas turbine and have looked into the PIC route using LCD
character screens, etc etc, but the amount of development time
constructing the proto board, having a final board produced,
soldering everything together, and further development to refine the
control just doesn't float my boat all that well.

My professional specialty mainly deals with modification of OEM
engine computers for the Nissan Z-cars and I've got thousands of
hours over the past seven years disassembling code, modifying it,
creating my own EPROM emulation system for real-time tuning, as well
as R&Ding and manufacturing many other bolt-on performance parts, but
of course, all that hard work pays. :)

So I got to searching for any sort of solution for using a PDA with a
data acquision device and finally came across a "budget" solution
that looks promising. Many of the PDAs available have a serial port,
and newer ones have a USB port, which can be converted to a serial
port using an adapter. I came across a company, www.bb-elec.com
which has a ton of DAQ solutions. And I came up with this device:

http://www.bb-elec.com/bb-elec/literatu ... 3804ds.pdf

It has:
7 channels of 12-bit A/D
4 channels of 8-bit D/A
2 Digital Inputs
1 Digital Output
37 samples per second when sampling all channels (~120S/s for single
channel)
9600baud, 8,1,N communication protocol
Power supply voltage of 12V-18V.


Programming for this device will be really simple as there are only
four commands.

The fact that is has more than enough channels for inputs and outputs
and a reasonable sample rate for controlling/monitoring a GT is nice,
but to top it off, this device is only $89.95 US.

A/D channels can accept inputs from a TIT thermocouple, pressure
transducer, and RPM pickup (using an LM2917 freq to voltage
converter). With expansion capability for four additional inputs
(afterburner temp & pressure sensor for "tuning" the A/B?). Or also
for monitoring oil pressure/temperature? In my case, I am thinking
to use a potentiometer on an input for controlling one of the analog
outputs to handle fuel delivery (pot on a thumb throttle for the jet
scooter). Other fail-safe functions can be implemented as well such
as using a digital output connected to a fuel solenoid to completely
cut off fuel flow, user-adjustable max TIT and RPM with closed-loop
fuel control to prevent overspool or overheating the turbine.

A PDA unit is quite compact and this DAQ device is also rather
small. The DAQ device can be located closer to the engine to avoid
having to lengthen thermocouple wires and a serial cable can connect
the DAQ to the PDA.

I haven't written any code for PDAs but given that many of them are
running WindowsCE or PocketPC, finding a compiler for these OSes
shouldn't be difficult at all, and a program to handle something like
this can be forged over a lunch break.

I'm not looking to reinvent the wheel here so if someone else has
already done this and is using such a system for operating their GT,
I would love to hear from you (and hearing from anyone is always nice
too.:)


Thoughts?

Ash Powers
Posts: 176
Joined: Sat Sep 04, 2004 7:17 am
Antipspambot question: 0
Location: Edgewater, FL
Contact:

re: Been a while - new project, T04 Gas Turbine.

Post by Ash Powers » Tue Dec 12, 2006 11:03 pm

So I have spent some time looking into various solutions and have come up with something better than I imagined: PDA-based engine data acquisition and control.

Bought one of these: ($135 on ebay, new)
Image

Bought one of these (RS-232 DAQ Device): $89.99 through bb-elec.com
Image
pdf here:
http://www.bb-elec.com/bb-elec/literatu ... 3804ds.pdf

And Bought one of these (Bluetooth wireless to RS-232 adapter): ($71 shipped on ebay, new)
Image

The idea here will be to use the PDA to wirelessly connect to the DAQ device via the bluetooth communication protocol/device. I will probably purchase NSBasic/Palm development software to write the program for the PDA to communicate with the DAQ device, or some other programming application (still deciding on which to use). A small board containing the pressure sensors, thermocouple signal conditioners/amplifiers, fuel pump speed controller, and the optical RPM sensor's signal conditioner will be produced to connect the engine to the DAQ device.

The PDA program will be written to have a display to show all engine vitals as well as have user-definable limits for things such as TIT, RPM, and have safety functions to reduce fuel flow if maximums are exceeded or shut down the engine completely. A small potentiometer can be used on an analog input for throttle control. Other routines for startup procedures can also be implemented and controlled as well. A full-color touchscreen will make for quick and easy manipulation of run-time variables as well as setting up datalogging and graphing functions for displaying engine information over durations of time.

I am most interested in setting up thermocouples both before and after the turbine and compressor so as to calculate efficiencies of these components on the fly - being able to make educated decisions on component modifications to improve engine performance will be ultimately what I'm looking to do here.

Ash Powers
Posts: 176
Joined: Sat Sep 04, 2004 7:17 am
Antipspambot question: 0
Location: Edgewater, FL
Contact:

re: Been a while - new project, T04 Gas Turbine.

Post by Ash Powers » Thu Dec 14, 2006 9:11 pm

Made some more progress last night.. I cut out the compressor diffuser vanes and inserted them into the diffuser plate and both the compressor diffuser vanes and NG vanes were tig welded to their plates.

Both items were chucked into the lathe and I proceeded to machine the contours into them to fit the housings. The process went really well - no mistakes were made with the exception of slight burn-through on the NGV backplate in a couple of spots which were cleaned up with a dremel tool and small carbide bit. Here's the pics:

Image

Image

Image

Image

Image

Image

Image

Image

I also recieved both the Bluetooth to RS232 wireless adapter and the B&B Serial DAQ device today. Still waiting on the PDA to show up, but in the meantime I've started with the coding, putting together the layout of the interface and the various configuration pages.

The software will have three primary operations: Startup procedure, runtime procedure, shutdown procedure. Each procedure has its own user-configurable operating conditions (conditions "x" specified by user).

The layout is as such:

Startup Procedure:
Check that RPM is zero, if not zero, error message
Apply power to startup motor
At "X" RPM, turn on igniter circuit
Turn on fuel pump at "X" startup fuel delivery voltage
Watch TIT to exceed "X" value (do we have ignition?)
If ignition, turn off igniter circuit, then
if TIT falls below "X" value, error message and halt (flameout)
else increase fuel delivery to "X" over "X" rate to reach idle condition (RPM="X")
if TIT exceeds max, error message and halt
at "X" RPM, disengage startup motor, maintain fuel flow rate
Display RPM, pressure, and TIT during startup procedure
Goto Running Procedure
End Startup Procedure

Running Procedure:
Get RPM
Get TIT
Get CC pressure
Get throttle position
Display data
Output throttle position to pump
Check for thresholds breach, if over, error message and halt
Loop

Shutdown Procedure:
Shutdown fuel pump
Get RPM data
At "X" RPM, engage startup motor
Run startup motor for "X" time for cooldown
pulse pump for "X" duration every "X" time (keep bearings lubed)
End Shutdown


Should be pretty fun getting the code up and running - I think I will build an engine emulator though to simulate the gas turbine to test the code before I strap it to the motor and bet the farm, if you know what I mean. :)

Johansson
Posts: 1161
Joined: Sun Nov 21, 2004 9:42 pm
Antipspambot question: 0
Location: Northern Sweden

re: Been a while - new project, T04 Gas Turbine.

Post by Johansson » Fri Dec 15, 2006 7:41 am

Great work!

You mentioned that you will pulse the fuel pump to lube the bearings while cooling the engine down, have you thought about other ways of delivering lubrication as well?

In your engine the fuel pressure won´t be nearly as high as in my bike, so bleeding some fuel off the main line won´t be a problem, but if you use a separate lubrication vessel pressurised with P2 you won´t have to mix oil in the fuel and the lubrication will flow as long as the compressor wheel is turning.

//Anders

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

re: Been a while - new project, T04 Gas Turbine.

Post by larry cottrill » Fri Dec 15, 2006 3:08 pm

Man, that is so beautiful. What a delicious project!

L Cottrill

Ash Powers
Posts: 176
Joined: Sat Sep 04, 2004 7:17 am
Antipspambot question: 0
Location: Edgewater, FL
Contact:

Re: re: Been a while - new project, T04 Gas Turbine.

Post by Ash Powers » Sat Dec 16, 2006 4:42 am

Thanks for the feedback! With this design of engine, I was planning to just mix 2-stroke oil in with the kero and have the fuel pump provide a small, metered amount of kero/oil into the bearing tube. A small needle valve and sight window was planned to be used to get the delivery to the bearings just right.

I haven't ever seen a setup as you mention using a seperate reservoir for the oil being pressurized by the compressor. I was speaking with Thomas Baumgart about how these engines are lubricated and he suggested to mix the oil into the primary fuel and feed that into the bearing tube as the kero will vaporize and carry a good bit of heat away with it. The kero would help to keep things cooled down and the residual oil left over once the engine is shut down wont evaporate, thereby providing some lubrication for the next startup process.

The code to pulse small quantities of fuel would be only for the bearings - to keep them lubed and help bring the temps down, although, this will also mean small quantities of fuel will also be delivered to the primary fuel ring to the combustor, which isn't so desirable. But I figure that while the engine is still hot after shutdown and continuing to drive the rotating group will just vaporize off the small amounts of kero/oil that are injected into the combustion chamber so as not to have a runaway condition from pooled fuel in the CC lighting off at next startup. I'll probably also incorporate a drain plug in the main body to prevent this condition from occuring, if it does.
Johansson wrote:Great work!

You mentioned that you will pulse the fuel pump to lube the bearings while cooling the engine down, have you thought about other ways of delivering lubrication as well?

In your engine the fuel pressure won´t be nearly as high as in my bike, so bleeding some fuel off the main line won´t be a problem, but if you use a separate lubrication vessel pressurised with P2 you won´t have to mix oil in the fuel and the lubrication will flow as long as the compressor wheel is turning.

//Anders

Post Reply