Compressor design

lilrex · Post by **lilrex** » Wed May 30, 2007 3:51 pm

I am designing a compressor wheel much like Kurt Shreckling but I have a few questions that might be answered here. First I will describe the project.

The wheel will be about 9 inches in diameter with retro curve impellers. I will cast a few of them in different configurations for my experiments. I will be working with low compression ratios in the order of 1.2 to 1.4 maybe a little higher if I can get away with it. I have found formula that describes the exit conditions of the wheel but not much on the actual wheel construction dimensions.

My fist question, what relationship does the inlet area have relative to the exit area along the periphery of the compressor: the inlet area being A1 and the outlet area being A2

Is it close to A1 = A2? Or is it more like A2 = A1 ∙ (1/π)
Or something else all together?

Is the power transfer ability of the compressor a relationship of the inducer radius and exducer radius? Or the mean inducer radius and the exducer radius? Or am I lost?

Kurt has done many experiment to ascertain the eff. of the compressor wheel he made and found the results to be satisfactory even for seemingly inefficient flow channels so I am not really worried about the fine details of the channel geometry I will just intuitively attack that problem. Any more comments or help will be appreciated and I will Post my progress as I make it.

cheers! :D

racketmotorman · Post by **racketmotorman** » Thu May 31, 2007 6:28 am

Hi
A reasonable inducer area is ~1.5 times the exducer area of the comp wheel .
Inducer diameter is ~2/3rds of the exducer diameter ,
So you'd be looking at a 6 inch inducer of ~28 sq inches area, with a 9 inch exducer dia of ~28 inch circumference , so 66% of 28 is ~18 sq inches for the exducer , or whatever a flow passageway of ~0.66 inches produces.

Mass flow at the low pressure ratio and tip speed you'll be using will possibly be ~3 lbs/sec

Power transfer by the comp depends on exducer tip speed only, if there is no inlet pre swirl. The faster it spins the more pressure you'll get :-))

Cheers
John

lilrex · Post by **lilrex** » Thu May 31, 2007 1:57 pm

what is the reasoning behind the rules of thumb outlined? I can respect a good rule of thumb but I am more curios then that. I have Kamps book and Schrecklings book but they dont have allot of info on the subject.

see I am thinking about things like this:

V = Volume
D = Mean Diameter of inducer and exducer
r = Rotations per second
o = Density of air at mean duct pressure
p = pressure
A2 = Exducer Area

p = (D * r * .5 * V * o)/ A2 'before diffuser'

Power = p * Flow 'not includeing diffuser'

I just pulled this fomula out of the keyborad just now and it is my no means a proffesional fluid dynamics formula but it gives you an idea as to what I am after.

by the way thanks for the information thus far it will help in getting this project off the ground.

Ash Powers · Post by **Ash Powers** » Thu May 31, 2007 2:30 pm

I wish someone would design/construct a 9" diameter centrifugal comp wheel like that in turbochargers. There are some aircraft-quality axial turbine wheels available out there that could be used on the business end. :)

The RR250 turbine wheels are almost always available on ebay for decent money....

racketmotorman · Post by **racketmotorman** » Thu May 31, 2007 6:08 pm

Hi Lilrex
The numbers come from measuring comp wheels , both big and small , just accept them :-))
Unless you want to read thru a whole heap of texts on compressor deign , and form your own opinions :-))

Cheers
John

racketmotorman · Post by **racketmotorman** » Thu May 31, 2007 6:11 pm

Hi Ash
Look for first stage Garrett TPE331 comp wheels , titanium and good for >40K at more than 9 inch dia , with >6 inch inducer .
Cheers
John

lilrex · Post by **lilrex** » Thu May 31, 2007 11:09 pm

Racketmotorman,

There is more to life then just accepting things, for instance people in this forum are not excepting the fact that it is extremely not economical to build a jet engine when you can buy one for less of a headache, not to mention the fact it would probably be allot better then the ones we make too. That said, we can all be into it for different reasons: one, we are all crazy! Two, we were dropped when we were a kid (or maybe that just me). Three, we like the challenge of doing something unique and difficult. The level of difficulty varies but its still there, and finally a sense of accomplishment. If this describes any one here raise your hand... if not oh well that’s ok too...

In closing of this attempt to justify my obsessions, I will yield to accepting that I will have to use the rule of thumb for now... but if anyone has more information I am all ears! Thanks for the ratios Racketmotorman

I will post my progress.

On another not does anyone have any suggestions on measuring the properties of the wheel? Setups, measurements, mathematics, ect.

lilrex · Post by **lilrex** » Thu May 31, 2007 11:21 pm

Ash,

that is my intent, I don’t plan on making it easy on myself, I want a really challenging casting project, I will also construct my turbine wheel using forged inconel blades and a ss hub using a Christmas tree attachments. The hardest part for me at this point now that I have a good rule of thumb for the compressor is figuring the strength for the parts. I am planning on using A356 and when treated to a T6 should have good strength. I am not sure what super alloy to use for the blades though. They all look good on the surface. Does any one know what the best super alloy to use for turbines is? And a good supplier for it, I have some sources but they are horribly expensive.

lilrex · Post by **lilrex** » Thu May 31, 2007 11:26 pm

I am planning to put allot of work into this thing and was wondering based on your experience, what compression ratio should I go for? It would be nice if the thing would run but I wouldn’t mind if it performed either.

Mark · Post by **Mark** » Fri Jun 01, 2007 12:13 am

"That said, we can all be into it for different reasons: one, we are all crazy! Two, we were dropped when we were a kid (or maybe that just me).

I was reading in "The New Yorker Magazine" the other day a cartoon. It had a businessman on the phone I think it was saying something like, "I would have gotten that package to you sooner, but I wasn't held enough as a child." The above reminded me of it.
Mark

lilrex · Post by **lilrex** » Fri Jun 01, 2007 3:03 am

Ha! seems it is pretty common to blame our childhood for our selves huh!
:lol:

lilrex · Post by **lilrex** » Mon Jun 04, 2007 4:04 am

here is a rendering of my current designe I think I am going to change the inducer angle.

racketmotorman · Post by **racketmotorman** » Mon Jun 04, 2007 5:42 am

Hi Lilrex

Sorry about the hurried reply the other morning :( ........more time now to answer :D

Your compressor wheels inducer tip angle is "normally" in the 30-35 degree range , but depending on your inlet velocity , again a function of rpm and mass flow rates , that angle could be a bit "steeper" at ~40-45 degrees

Inconell 713 is the prefered alloy for turbocharger turbine wheels as well as the axial micro turbine wheels . GMR235 is also used for turbo wheels .

Having spent the last 17 years playing with home made turbines , I'm well aware of their expense and difficulties , but having ridden and driven turbine powered vehicles (shaft drive) I also appreciate the "sense of accomplishment" .
Heres some "hard numbers" from the first stage comp wheel I have from a Garrett TPE331 (T76) aero engine .
Inducer dia 6.6 inches with a 35deg tip angle and 17 full length blades at 0.075" thick (titanium) .
Hub is 2.4 inch dia .
Trim is ~52
Overall comp wheel dia 9.2 inches with an exducer tip height of 0.5 inches , blades at an angle of ~60-65 degrees sweepback .
This comp wheel spins to 41,730 rpm producing a tip speed of ~1675 ft/sec , very high !!
Inducer inlet to comp outlet area ratio is ~2:1 with an inducer diameter ~72% of the OD.
As this comp is the first stage of a two stage system , its design is "different" to other compressor wheels but very similar to the comp wheel in my Garrett TV84 turb that I use in my turbinebike .

As pressure produced varies as the square of the comps tip speed, you'll be needing probably 15,000 rpm as a minimum to get some working pressure to make the engine run .

There are large centrifical compressor wheels,( 11 inch dia ) used as superchargers on radial aero engines, with inducers of ~6 inches in dia .
The large overall dia to inducer dia is so that rpm can be kept "low" to minimise transmission difficulties , whilst still producing reasonable boost to the engine .

Cheers
John

lilrex · Post by **lilrex** » Mon Jun 04, 2007 5:02 pm

Hi Racketmotorman

It seems to me that the relationship between the inlet area A2 and outlet area A1 is simply to match the inlet velocity (correct me if I am wrong) against the outlet velocity.

Where A2 is the inlet area, A1 is the outlet area, Vi is the inlet gas velocity, and o is the increase percentage in gas density at the outlet.

Vn = (A1/A2) Vi (1-o)

For my wheel using the before mentioned logic:

674.67 = (18.66/12.44) 500 (1 - .10) = gas velocity

Of course that puts the Compressor at 17180 RPM. This does not sound right as a running speed (674 FPS) so am I looking at this wrong? About the inlet velocity, you suggested 500 Fps on another thread of mine, I want the flow to be subsonic but how close to sonic should I try to take it? I am thinking that I might change the A1 / A2 ratio in order to bump up the exit speed a little. Maybe closer to the wheel you were talking about 1:2 instead 1:3 like I have it now.

Currently the specs on my compressor are:

Major Diameter of 9in

Inducer Diameter of 6 in

Hub Diameter of 3.5 in

Inlet area of 18.66 square inches

Outlet area of 12.44 square inches

I have not decided on a specific retro curve yet.

Why exactly do allot of compressor I see have the blades pitched forward like illustrated in my design? I figure it is to reduce gap losses creating a lower pressure area at the leading edge of the blade but if you have any input about that I would appreciate it. Actually I appreciate anything I can get.

lilrex · Post by **lilrex** » Mon Jun 04, 2007 5:08 pm

any ideas about how to estimate the max speed of a compressor of this shape?