100-150 Hp turboshaft thoughts

[monty]

Hi guys. I've been lurking for a while and decided to surface and pop the hatch. Used to be a member of the GTBA a long time ago. I've been sidetracked by life for a while, but the old turbine bug is gnawing at me again!

I'm wanting to build a 100-150 hp direct drive turbine. I'd settle for 50-70... but we all know too much power is almost enough! It would be great if it could burn diesel without causing problems. Jet A isn't cheap!

What I have in mind is the rotating assembly, derived from a large turbocharger. All the other components will be fabricated by yours truly. I am going to try to combine the best of the model engine world with some automotive stuff for reliability. For instance, I'm thinking of using hydrodynamic bearings so I don't have to replace expensive ceramic ball bearings every 20 hrs.

I know the limitations of a direct drive turbo-shaft, but one will work just fine with my application. I could just hunt around for a Solar T62, but it has a lot of extraneous stuff on it, that I don't need or want. I'd like to use heavy duty truck turbocharger components if possible. Mainly to be able to take advantage of newer compressor/turbine technology, and also economies of scale. Plus I don't want parts that I have to hunt and scrounge for, just to pay thousands of dollars for something with unknown cycles and use/abuse history. I want to be able to buy the components when I need them without restriction (bearings, turbine wheel and compressor). Maybe the larger Holset stuff-HX60?? It would be nice to find something with a large enough family of parts to mix an match. I know I need a smaller compressor, and larger turbine for a turbo-shaft, but I'd like to know your thoughts.

I can run the numbers, but I've learned through the years to ask more knowledgeable people just to make sure I'm not way out in left field. I've been impressed with all the things you guys are doing. Any thoughts/recommendations?

Inducer and turbine size?

Vaporizers? Yay/Nay

Ball bearings or hydrodynamic?? (either way I plan to have a closed oiling system)

Any good model engine or open source FADECs out there? I haven't looked in a long while.

Looking forward to contributing to the forum. I'll start a build thread when I get going.

[racket]

Hi Monty

Welcome to the Group :-)

How do you intend taking the power off the shaft ??

This has always been a stumbling block to using a turbo rotative , especially with "brass" bearings , theres a lot of "movement" , not to mention the high rpm which makes the gearing difficult , a two shaft arrangement is a lot simpler .

Getting your required horsepower won't be a problem ,.............LOL, plenty of examples in the DIY Engine section jetandturbineowners.proboards.com/board/2/diy-turbine-engine-discussion-builds.

Is there a particular reason why you only want a single shaft engine ?

Cheers
John

[monty]

Hi Racket,

I plan on taking power off the front of the shaft, just like most of the APU engines. I am still scratching my head about the Hydrodynamic bearing location vs gear shaft situation. Calculated torques and forces at the power levels I'm after are very low, so some type of high misalignment coupling/spline is a possibility. If I use an electric pump, I can pressurize the bearings before start and during shutdown. I'll also look at ball bearing's properly sized, with a closed oil system. You did get right to the sticky point though! ;-)

I'm not too worried about the gearing. I'm fairly handy in a machine shop. Plus if the model helicopter guys can pull it off, it's humanly possible! The load for this project has a power consumption curve just like the turbine output curve, and I can trim it to match, and also limit over-speed...so I think all that is tameable...If not exactly easy.

The desire for single shaft is mostly packaging and wt. Both important for what I want to achieve.

Thanks for the reply!

Monty

[racket]

Hi Monty

OK , single shaft it is .

You need to start at the "back end " to make sure you have enough area when the gases are fully expanded , so go for the Holset HX82 size if you want to go Holset , or large frame Garrett in G Trim for TV94 , both are 129 mm dia inducer and ~110 mm exducer , get one without any clipping of the exducer as you need max gas deflection .

To keep turb tip speeds within reason , you'll be needing a very low Trim compressor wheel so that you can get your Pressure Ratios high enough for reasonable power output as well as SFC .

The Holset HX82 comp wheels with 152 mm exducers will allow good tip speeds , but finding one with a smallish inducer of ~92-95 mm max might be difficult.............checkout the KTS Site for suitable billet wheels .

"Suitable power curve"..........mmmm, maybe a boat?

At 60-65,000 rpm you'll only be having ~12 ft lbs of torque extraction , so a fairly lightweight ( thin) PTO shaft required , I've often felt that taking the power out via the rear end might be "better" due to the very heavy turb wheel being less sensitive to any PTO shaft inbalance , its "flywheel" effect and the fairly thick hub might make for easier attachment of the shafting .

Where in the world are you situated ??

Cheers
John

[monty]

Sept 19, 2018 20:00:32 GMT -5 racket said:

"Suitable power curve"..........mmmm, maybe a boat?

Close....but I think it wise to keep my ultimate level of insanity to myself, until my work reveals I'm not a complete nutter...  

At 60-65,000 rpm you'll only be having ~12 ft lbs of torque extraction , so a fairly lightweight ( thin) PTO shaft required , I've often felt that taking the power out via the rear end might be "better" due to the very heavy turb wheel being less sensitive to any PTO shaft inbalance , its "flywheel" effect and the fairly thick hub might make for easier attachment of the shafting .

Just like all your previous arguments, completely correct and accurate, but I'm going to do the opposite, because...packaging. And I have chosen the way of pain.

Where in the world are you situated ??

36.009749 N 94.165040 W.....A very long way from NSW!

[monty]

Sept 19, 2018 20:00:32 GMT -5 racket said:

To keep turb tip speeds within reason , you'll be needing a very low Trim compressor wheel so that you can get your Pressure Ratios high enough for reasonable power output as well as SFC .

The Holset HX82 comp wheels with 152 mm exducers will allow good tip speeds , but finding one with a smallish inducer of ~92-95 mm max might be difficult.............checkout the KTS Site for suitable billet wheels .

Actually I'd probably prefer to go with Garrett...Any reason I can't trim a commercially available wheel myself? I have a CNC lathe.Thanks for the food for thought....I'll do some digging.

[racket]

Hi Monty

OK , Middle America , so no problems getting whatever bits you need , that'll make life easier than for some of our Members , me included .

Front power takeoff it is then , I'll be interested to see how you manage it .

Probable max flow rate ~2.0- 2.2 lbs/sec depending on your Pressure Ratio , enough for 150 hp .

Cheers
John

[racket]

Yep , Garrett would be more available where you are .

Some of the newer comp wheels have "complex" shroud curves , but nothing a CNC couldn't machine.

[monty]

I did some digging and found part numbers for the TV94. Only turbines I've found are the -001, -003 and -008 wheels. The -003 looks like it's been clipped. The -001 has a 106mm exducer, and looks un-clipped. That's the one I was planning to order unless somebody has better numbers. I've learned I need to just call and order part numbers like I know what I'm doing. Otherwise they get suspicious and sometimes won't sell to you. The hot rod shops don't care, but they charge a lot more than the industrial places around here.

I found conflicting info about the diameter of the through shaft. Is it 11.2 or 11.12mm? Probably a typo, but I don't want to order the wrong compressor wheel.

Bthw John, I've been down the rabbit hole with a couple of your builds on here. First class stuff! Any idea why so many pics have gone missing?

Monty

[racket]

Hi Monty

I've been using the 442208-0001 turb wheel which is the F Trim wheel with the 4.165" exducer, the G Trim wheel has a 4.365" exducer , the -0001 wheel is a common and relatively inexpensive one and was the reason I first started using it, though I've had it come with varying degrees of clipping .

The quill is 11.12 mm , 7/16" ..........though theres slight variations in diameter along its length , with a tad larger near the oil slinger which makes it a force fit into the superback of the comp wheel for accurate alignment, the billet comp wheels we use from KTS have 11.12 mm bores

The pics in my build threads should all have been fixed recently , could you point me to any you can't find and I'll see what I can do .

Cheers
John

[monty]

John,

The missing pics are about mid way through your 666 engine build, but there a lot of pics missing everywhere I look. Maybe something on my end??

442208-0003 is the G wheel I think, roughly $500US.

looking at X895 KTS comp wheel 152mm exducer 95mm inducer. Only thing is, the wheel is made from 7075. 7075 is really strong...at room temp. Not so great at elevated temps. Plus the fatigue characteristics aren't great. Good stuff, just not so sure a comp wheel is the best place for it. Thoughts? Wish I could find a TI one in larger diameter, and MAR M turbine wheel....maybe for version 2.0

Bearing wise, I've been thinking about a couple possibilities. One that I currently favor is to use a deep groove ceramic hybrid at the pinion gear for the PTO. In this arrangement I could eliminate the stock hydro-thrust bearing. I read that it caused you some issues, and isn't the best design anyway. The nice thing about the deep groove ball, is a high angular misalignment tolerance. The rotating assembly could orbit on the bushings without causing problems at the ball. Whirl could be an issue, but the bushings should do a good job of damping that. I would mount The ball so that the outer race is oil damped, and use a wave spring to pre-load in the primary thrust direction. This would allow for some radial displacement and add damping, while preventing ball skid. Ceramics should really help with centrifugal loading. Since I'm using a straight cut planetary to take the power out, there is minimal thrust load from the pinion, and no real radial load. All the ball bearing would do is take the main thrust load, and locate the pinion. Planets could be on hydrodynamic bearings, and it could all float to find its happy place under load. Prelim life, strength calcs on the gears look pretty good. Probably going to dig into #s heavily this weekend.

Monty

[racket]

Hi Monty

The pics all seem to be there in the 666 thread , the Group had some problems with where pics were kept , each contributor needs to fix theirs, I spent several hours on mine :-(

If you could verify the -0003 wheel is the G Trim I'd appreciate it in case someone ever needs one .

Yep , all alloy wheels are crap at elevated temps , but Ti isn't needed until running past ~5.5 :1 PR , the ship turbos run alloy at >5:1 for tens of thousands of hours , Ti has its place but not really needed unless the comp has to cope with "special" needs , Holset make a Ti wheel for their HX83 turbos , theres also some wheels made by KTS , if you'd be a tad more forthcoming with your requirements I might be able to provide solutions ;-)

I think Turbonetics use a " ball and brass " setup with their turbos.

Cheers
John

[monty]

John,

I talked to one guy today who said the -0003 wheel was a 74 trim with a 111.something mm exducer. It is clipped unfortunately. It's not the G trim wheel you have in mind. I'm still searching. I need to look through an old hard drive. Somewhere I think I have one of Garrett's catalogs. It seems to be getting harder to find good info online. Garrett's info for the automotive world is easy to find, but the industrial stuff-not so much.

I'd like to hit a 5/1 PR with reliability, but 4 will suffice if it's required to keep things from getting out of hand cost and weight wise. I'll post the layout I have in mind a bit later this weekend. That KTS spreadsheet is fantastic! I have settled on the X897 compressor, and possibly the -0003 turbine wheel, but I'm trying to find one that isn't clipped before I give up.

For now my requirements are to keep me entertained and perhaps make lots of fire smoke and interesting noises while wasting copious amounts of dinosaur juice. We'll see how it goes from there, though careful study of my EXACT location will give some indication of one of my more serious addictions and where you can find me a lot of weekends. ;-)

I checked out some of your videos. That one where your test stand started bucking around and all the bushes were acting like they were in a tornado was quite entertaining!

Thanks again,

Monty

[racket]

Hi Monty

Yep , very hard to get dimensional info on most turbo stuff.

Andy is using a Holset wheel with 111mm exducer jetandturbineowners.proboards.com/thread/1025/hx-102-255-money-pit?page=2 it has a clipped exducer as well .

With regards running at a 5:1 PR , I'm not certain that its possible to efficiently fully expand down from that in a single shaft engine using a turbo turbine wheel, if theres a 5% drop through the flametube we'd have a 4.75 PR going into the NGV , with a choked NGV theres another 1.9:1 PR used up , leaving a 2.5 : 1 PR across the wheel, 5:1 PR with a two shaft engine is no problem .

Precision turbo run a H Trim wheel in their PT118 www.youtube.com/watch?v=-w3Xc5UbBVI , again its been clipped .

Once you have the turb wheel , a quick measure up of its exducer throat area will allow you to reverse engineer for the comp inducer size

LOL...........that video had an ~ 94 mm inducered comp wheel running at ~3.5 PR .

Cheers
John

[monty]

John,

This should clarify what I'm up to:

Fan flow rate 30 lb/s PR=1.08

Mixed flow, which isn't the best, but I'm going for smiles per gallon, not miles per gallon.

I'll take all the core PR I can get, whatever the turbine can't handle the exhaust nozzle can use.

Variable fan nozzle to trim the fan PR in order to keep the core PR up.

This is not the most efficient way to do this, but it is the most cost effective way!