100-150 Hp turboshaft thoughts

[monty]

John,

My diffuser starts at 172.5mm dia. My numbers on this wheel are 11.8 at the tip, and I was planning about an 11mm ht because of the issues you raised. I measured the actual wheel. Haven't put it on the inspection plate yet, but I will.

I'll post the design as it is now tomorrow.

Monty

[racket]

Hi Monty

Thats the X856 wheel , it has the bigger tip height and a bit less superback than the X846 wheel , but probably a better wheel for a 4:1 PR than the X846, the old GT6041( 106 mm inducer 141 mm exducer) had a 12 mm tip height, the vaneless diffuser had a "pinch" just outboard of the exducer where the axial height of the vaneless space rolled over to an ~9mm height over a distance of ~15 mm radially , this then forced the air to a greater radius quicker reducing its flow path and improving efficiency .

Your exducer is a similar height at 11.8 mm to Andy Ms X858 wheel he's now using jetandturbineowners.proboards.com/thread/1025/hx-102-255-money-pit , the inducer is 110 mm so just a tad more flow, but it has an "advertised??" exducer angle of 45 degrees so I've gone for an extra few thousand RPM ( 67K) to get his 4:1 PR , diffuser tip at 87 mm "R" ( 174 mm "D" ) with an air angle of ~20-22 degrees on a ~9.0 mm diffuser height ...............I tried all different heights but ended up coming back to ~9mm

Cheers
John

[monty]

John,

My thinking: on the lower PR to up the tip ht. because of the lower discharge density. I need around 64-66Krpm for the fan with the gearbox ratio I have. The smaller inducer should be good for the lower flow I'm after in the core. I'm sure there is an optimum trim, and diffuser that would work better....but I'm just trying to get close. Refinement can come later. Besides....too many guesses!! The lower PR does require a larger diffuser throat area though. I've tried torturing that to death, and I always come up with 17 channels. When I up the flow and go with a higher PR, I get 15 channels. If I reduce the ht to 9 or so, I'm going to need even more channels. I don't want to do that because then I don't have room for pass-throughs for the fuel/oil etc. I have a book called Centrifugal Compressor Design and Performance by David Japikse. It has a lot of empirical data on channel aspect ratio, L/D etc that I'm using to size things. I'm trying to keep everything in the ~80% efficiency range. Getting closer to 90-95 is possible, but at the expense of very difficult fabrication and no convenient pass-through/fastener locations....

I am deliberately under sizing the diffuser slightly for the compressor. In theory the inducer should be slightly smaller ~98-100mm for 3 lb/s.....but data is really the only way to get there... and I have none.

Monty

[monty]

This is where things are currently:

[racket]

Hi Monty

Yep , all we can do is keep juggling numbers until we have the least amount of compromises and "unknowns" , even the best design is only ever going to be "right" under very limited circumstances , and without any comp maps we're flying blind anyway.

Japikse has put out quite a few Papers .

Graphic is looking good :-)

Cheers
John

[monty]

Thanks John,

Yes...compromises!! The X855 or X850 wheel is probably better, but I have the X856 in hand! I'll design my castings so the X855 and X850 can be used.

Monty

[monty]

John,

What are the dimensions of your burner can? length specifically. I'm trying to figure out how much room I have for the diffuser before the can. I'd like use that dead space in an attempt to improve diffusion, rather than just having a straight dump diffuser with a lot of swirl component.

Monty

[racket]

Hi Monty

I used a length of 10" dia "industrial" stainless tube for the 10/98 engine jetandturbineowners.proboards.com/thread/19/10-98-engine , Andy M is doing the same for his 10/110 engine jetandturbineowners.proboards.com/thread/1025/hx-102-255-money-pit

My 10/98 FT inside cross section was 223 mm OD and 127 mm ID which gave me 3.6 inducer areas ( 98 mm inducer)

Now as for length ..........bugger , nothing with my hole calcs , better look for a drawing ..........., heh heh , can't find one :-( .......OK build journal , better luck , inner wall was 62 mm long , outer wall 100 mm wide strip jetandturbineowners.proboards.com/attachment/download/15 this flametube was ~10 mm shorter than a previous build due to the lube in/out castings on the back of the diffuser wall.

Cheers
John

[turbotom]

Monty, are you sure your reduction gearbox will be able to handle the power? To me, it appears to be very much on the small side. As coincidence goes, I recently also got involved in a turbofan project, very similar to yours. It will be based on a modified version of my "Neo" power head, utilizing the same rotating assembly. Since it's supposed to be quite a "fast" powerplant (heading for an airspeed of 500+ km/h of the craft, probably with a twin engine installation), the fan will be revolving at round about 16.5 krpm. At that low reduction ratio, a classic epicyclic gearbox isn't possible anymore since the planets would be too small, revolving much too fast to be suspended on off-the-shelf bearings that will be able to carry the radial loads. Since a coaxial drive shaft was mandatory, and I wanted to do without another set of high-speed bearings for the input pinion, I went for a design with three intermediate shafts, running at round about 17krpm. The calculation of the gear loadings resulted in a design bigger in diameter (186mm) than the impellers but considerably less than the outer diameter of the power section. The power rating of the gearbox is 150+ kW at 10k hours lifetime (at least) but it won't be able to handle more than 200kW. Compressor intake can still be arranged smoothly around the input end of the gearbox. Of course, the cantilever design of our rotating assembly helps here. The starter/generator will be arranged the same way as in your design (makes kind of sense that way...). Lubrication is also quite tricky, every intermeshing point of the gears and every bearing will have to have its individual lube oil supply. The oil/air separator is included in the output shaft by means of four radial bores -- at least, that's a no-brainer. Here are a few images:





And a sketch of the gearbox: Gearbox PDF

Please don't get me wrong, I don't want to interfere with your project or in any way compete against you, our designs follow completely different approaches and hence are not directly comparable, I just wouldn't want to see you to get into trouble with an external component, not directly related to the core engine, that may be inadequate for the task. So please just take it as a hint to check the design of your transmission again.

All the best,
Thomas

[monty]

Oct 30, 2018 9:00:22 GMT -5 turbotom said:

Monty, are you sure your reduction gearbox will be able to handle the power? To me, it appears to be very much on the small side. As coincidence goes, I recently also got involved in a turbofan project, very similar to yours. It will be based on a modified version of my "Neo" power head, utilizing the same rotating assembly. Since it's supposed to be quite a "fast" powerplant (heading for an airspeed of 500+ km/h of the craft, probably with a twin engine installation), the fan will be revolving at round about 16.5 krpm. At that low reduction ratio, a classic epicyclic gearbox isn't possible anymore since the planets would be too small, revolving much too fast to be suspended on off-the-shelf bearings that will be able to carry the radial loads. Since a coaxial drive shaft was mandatory, and I wanted to do without another set of high-speed bearings for the input pinion, I went for a design with three intermediate shafts, running at round about 17krpm. The calculation of the gear loadings resulted in a design bigger in diameter (186mm) than the impellers but considerably less than the outer diameter of the power section. The power rating of the gearbox is 150+ kW at 10k hours lifetime (at least) but it won't be able to handle more than 200kW. Compressor intake can still be arranged smoothly around the input end of the gearbox. Of course, the cantilever design of our rotating assembly helps here. The starter/generator will be arranged the same way as in your design (makes kind of sense that way...). Lubrication is also quite tricky, every intermeshing point of the gears and every bearing will have to have its individual lube oil supply. The oil/air separator is included in the output shaft by means of four radial bores -- at least, that's a no-brainer. Here are a few images:





And a sketch of the gearbox: Gearbox PDF

Please don't get me wrong, I don't want to interfere with your project or in any way compete against you, our designs follow completely different approaches and hence are not directly comparable, I just wouldn't want to see you to get into trouble with an external component, not directly related to the core engine, that may be inadequate for the task. So please just take it as a hint to check the design of your transmission again.

All the best,
Thomas

Hi Thomas!

That's a gorgeous piece of work!

Thanks for the hint. ....and nope, I'm not sure....but I am only looking for 1000 hrs of life, and I'd be happy with 500. I will definitely check the numbers again. I am not using rolling element bearings on the planets. That and the life requirement are very different about my design.  If it doesn't work, I don't have a lot of money in this...so no biggie. Plus it keeps me occupied. Just get yours going so I can buy one if mine doesn't work!

My goal is to see just how much on the cheap I can get away with. I'm willing to push the envelope a bit. My pitch line velocities are not THAT high, at least not for high speed gearing. Without the modified addendum it wouldn't work at all.

There is definitely some design risk in the gears with FS close to 1 on surface life, but the bending loads are all well over 2. I have room to go with a larger gear set if I need to. It started out larger. The nice thing about not having investors or a boss breathing down my neck- I can push the envelope and see how far it bends. If it breaks....I'm out some work, but this is all just an entertaining way to spend some time for me, so no worries. But, yes....I will check the numbers again.

Monty

[monty]

This gear can handle 22000+KW!!!

Monty

[turbotom]

Oct 30, 2018 22:42:44 GMT -5 monty said:

This gear can handle 22000+KW!!!

Monty

Yes that's really amazing. I have no idea how it's possible to make a reduction gear that small to withstand that amount of power continuously. I think the basic design is an epicyclic gearbox with a multitude of planets. The real art is distributing the torque evenly over all the planets. While with three planets and a floating sun gear, this happens all by itself, with a higher planetary gear count, the planet axles (or their mounts) must provide some amount of flexibility to even out the machining tolerances. Maybe also some "esoteric" surface coating / processing is utilized and probably very special lubricants are used.

When using "typical" materials like 16MnCr7-6 for the gears and ground surfaces (gear quality 5), there isn't that much margin for the gearbox between the "will run forever" and the "breaks down in the glimpse of an eye" loadings. The relevant figure is usually the surface pressure at which fretting occurs. Due to the high speed / high intermeshing frequency, once surface breakdown has started, it will quickly lead to complete failure.

I think even the "big players" still have their share of trouble with these new designs. IIRC, the RR TP400 (for the A400M military transport aircraft) was delayed considerably due to prop gearbox problems (...though the TP400 isn't a geared fan).

I'ld recommend to use some "proper" gear design software in order to verify that the design will provide at least some endurance... Kisssoft for example offer free demo- and test versions of their software.

All the best,
Thomas

[monty]

John,

You guys are all using 10in seamless schedule 5 tubing??

Is there anything wrong with rolling and welding a casing? It would be easier to just cut out a piece of stainless and roll it using a slip roll for me. I've got to do that for the tailcone and a bunch of other stuff. Why did you guys settle on the method you are using?

Monty

[racket]

Hi Monty

Using standard tubing meant we could be confident it could hold the pressure , though for my FM-1 engine as well as the 12/118 I got a bit of 1.6 mm stainless sheet rolled and welded.

The 9/94 engine used a thin walled stainless container , but I had some problems with it when I silver soldered in the ignitor boss, its ideal if we can find a suitable "can" before machining up engine parts , I've looked at some stainless propane tanks for marine use that would make nice cans but getting just the right size is difficult..

Cheers
John

[monty]

Thanks John,

Unfortunately my TV94 rebuild kit came without the front thrust washer, and I am not having any luck finding one. It has me thinking of using the upgraded HX85 thrust arrangement. Any thoughts?

Monty