TV94 Turboshaft Project

[ashpowers]

I've been visiting this site recently and it seems there is a bit more member involvement here vs. over on the PJ site. Good projects going on here at this forum!

I have been working on this new engine build for a few years - a lot of what you see in the images below was done back in 2007. Amazing to think its been that long, LOL. I've made pretty much all of the cold section components, bearing assembly, and am working towards completing the hot section of this engine.

Here are a few pics of where she's at:

Solidworks model:








RR Allison 250 4th stage wheels.


I recently obtained SolidWorks 2010 and have been diving head first into learning this application and have modeled the hot section components. Still a few things left to draw up - waiting on a TV94 turbine heat shield so I can incorporate it into these parts - that should be here next week sometime. Here's the new models:







I have a local cnc shop that my business uses for various parts, however, they have rather antiquated mazak machines - like 1984 model units which complicate their ability to import a dxf file. Their machines cannot even save programs - they literally have to reprogram them for each new part they run.

I'd really like to find a cnc shop that could produce these parts. Are there any machinists on this forum who would like to take this on? Not asking for any handouts here - I have the 316SS rounds to make these parts - it machines well and has corrosion resistance to temps well above what it will experience. I simply do not have the time to hand fabricate these parts like I used to. Not to mention, the radial vane section would be a huge challenge to do on a manual mill.

As for the turbine exducer ring that shows the axial swirl vanes, the vanes are not machined - these will be welded to the ring after it is turned. I'll be picking up some 316ss thinwall tube to cut those pieces from and attach them.

Happy to be here and a part of this community! I look forward to participating more with the members of this forum! ;D

[racket]

Hi Ash

Some nice bits you've machined up there :-)

Yeh , a bit more turbine activity here , this Site is excellent for Build Threads .

Looking forward to more pics as the build progresses ......................that'll be 3 different TV94 builds on here now , lots of different approaches makes for good problem solving .

Cheers
John

[ashpowers]

Hi John!

I've been reading through Anders thread and am green with envy. Something that went through my mind was a recosideration of using the reverse-flow combustor design and going to the layout you and he are using. I'm just really surprised at how small the combustor is with your layout and that it appears to work.

I'm not going to change direction now though - too far in. Nice thing about the reverse flow configuration though is the bearing tube isn't surrounded by combustion. It also allows for a full diffusion of gases produced by the comp section, and because the air is moving in opposite directions on the inner and outer walls of the combustor, it should create a strong swirl at the top end of the combustor and improve combustion.

Way back in the day when I built that turbojet engine using a stock nissan 300zx turbocharger I placed the compressor discharge entry point into the combustion chamber housing at the bottom closest to the turbine housing inlet.


This actually created an unexpected result: the top plate of the combustor housing would literally start glowing red hot. I had to use a cone-shape "cap" for the flame tube to control the combustion area to drop it a little further down the tube.

Original:


Final:


With the centrally located flametube like you and Anders are using, the flow of air is all moving towards the rear. This makes it difficult to create swirl in the primary zone and reduces the volume of air moving into this zone. With the space constraints that layout has and the short length of the flame tube, combustion is really not occuring in the primary zone - it appears to occur around the secondary zone and dilution zone. My T04 engine showed the same evidence:



I've conservatively sized the combustor in my TV94 engine but I'm really starting to lean towards shortening it quite a bit - at the most, the same length as the combustor in your engine.... maybe even shorter. I'm thinking the flow pattern of the reverse-flow configuration is going to produce the same effect I saw in the MK2.5 Z turbocharger engine.

It would probably be a good approach to install a few viewing windows into the casing - 3 of these at each zone would allow you to easily see where the burn is. Might even be able to make them such that you cut a 1" port into the side of the flame tube and install a joining tube onto the backside of the viewing port that can be installed/removed. This way, your view port into the flame tube would not create flow disruption/distribution issues. You would be able to get some very useful insight into the flow patterns by how combustion is occuring as well as verifying combustion location, etc etc and make alterations to the flame tube design from there...

McMaster has high pressure glass sights for $13.38, 536F max temp, 1250psi rated.
www.mcmaster.com/#sight-ports/=g3vrl8

[racket]

Hi Ash

The flametube on my 9/94 engine ( 9" dia outer can compared to the 10/98's 10" dia can ) was considerably smaller in volume again even though both use the TV94 comp, pics 75 - 82 in my Rackets Photo Folder over on Yahoo DIY .

All the flametubes in my micro engines have been ~115 mm long , which is kinda short but they still seem to produce reasonable combustion but as you noted it tends to occur a little further downstream than the primary zone , same as with all the commercially available RC micro engines , the "front" end wall stays nice and shiny with little heat effects , I modify the first row of holes to try and induce some swirl into the airflow inside the flametube .

I'm looking forward to seeing how your reverse flow setup ends up , it certainly removes that necessity for a relatively short flametube , and I like the way you'll have plenty of room around the shaft tunnel for cooling, oil, etc .

All of my micros , FM-1 , 9/94 and 10/98 needed to be as short as possible for fitting into a bike frame , thankfully the combustion worked , though having to include 18 evaporators and fuel injectors to speedup combustion was a bit of a pain ....................strangely enough combustion ended up the least of my problems even though I thought it would be the greatest when I first started making them ....................LOL, such is life :-)

Cheers
John

[Johansson]

Hi Ash!

Good to hear that your turbofan build is up and running again, as John said it will be very interesting to have three different approaches to a TV94 based engine on the same forum.

Perhaps a shaft power competition between us three once all engines are running?

[ashpowers]

I've been having more fun with SolidWorks and modeling the entire engine. I've modeled every part I've already built as well as nearly all of the remaining parts for the design. I will model the compressor wheel this evening as well as the radial labyrinth seal that seats behind the compressor wheel, the injector nozzles, the bearing sleeve, shaft seals, and combustor zone ports. Once these remaining items are modeled I will be using the CFD package in SW to analyze flow characteristics of the engine; diffuser performance, combustor flow properties, and nozzle guide vane flow properties.

Another nice thing with SolidWorks is its ability to calculate masses of parts and assemblies. What you see here in total comes out to 40.56lbs and will probably add another 2-3lbs once the remaining parts are added. There are a number of pieces that can be modified to reduce this mass - I've overbuilt so many of the pieces with little consideration for weight. My target goal with this engine complete with free-power was 45lbs and it looks like I'll probably be right on the money.

[Johansson]

I wish I had those modeling skills, it sure beats the napkin drawings I am making. =)

How will you secure the jet pipe/power turbine coupling to the overhung combustor casing?

[ashpowers]

Hi Johansson,

I hear ya - I was doing 2D drawings in MS Paintbrush for several years, LOL. This program has definitely made things far easier to visualize the layouts and how parts will come together.

As for the jetpipe, the drawing currently just has the end face of the jetpipe mated to the outer casing's rear port which obviously isn't a solution. There are a few different approaches to that interface that I can think of - I'll just have to model them and see which one looks to offer ease of fabrication and good functionality.

I've spent the entirety of my evening modeling the compressor wheel - a little over 4 hours to get it right. Most difficult part was getting the two base sketch profiles correct to produce the proper blade shape using the loft function. That took 90% of the time - everything else was straight forward except for that but at this point I could do it in a fraction of the time. Check out the results.





Hope to have the rest of the parts modeled in the next couple of days so I can get to the CFD. This part beat me up pretty good tonight.

[Johansson]

Very impressive work on the comp wheel drawing Ash.

Can such a drawing be used somehow to CNC-mill a comp wheel or is that an entirely different process?

[ashpowers]

I dont see why it couldn't be used to machine a new wheel. SolidWorks can create a DXF file for any drawing which can be imported into CAM software packages.

[wolfdragon]

As long as you have a post processor that will make a dxf or direct machine file that the destination machine wants, most all 3D modeling programs can be used for CAM

it's usually well worth one's time/effort to find out what program the machine shop uses for it's own stuff and match it

[metiz]

CNC programming is usually pretty straight forward but all those curves will be a real challenge. Definitely don't program by hand

[Johansson]

I wonder what it could cost to have a series of, say, 10 comp wheels in the 130mm inducer class milled from solid? All drawings made etc. so just the manifacturing cost.

Group buy anyone?

[racket]

Hi Anders

Yes please :-)

Cheers
John

[metiz]

check out this video of an impeller being milled www.youtube.com/watch?v=AuH8OoC0JrI&feature=related