"Fat Boy" 12/118 thrust engine

[ripp]

Hi John,
is it possible to measure the distance between compressor wheel and turbine wheel?




Cheers
Ralph

translate.google.com

[racket]

Hi Ralph

Approximately 124 mm, there is slight variations due to the inconsistency of the turbine wheel casting.

This distance can be increased by ~6 mm by adding a spacer behind the comp wheel as theres plenty of thread available as standard, but any changes to the positioning of the wheel would come with changes to rotor dynamics ...........I had the comp wheel as far forward as I could when building my ball bearing equipped engines .

A "better" turbine wheel for a gas turbine build is the Garrett GT6041 wheel as the distance between comp and turb is ~40 mm larger if my memory serves me well , but this wheel is EXPENSIVE and it requires a screw on boreless comp wheel with a very limited range of sizes compared to the bored ones we are using.

Cheers
John

[turboron]

John, I have never seen the logic behind the extended tip impeller. I assume that some clever aerodynamist predicted a benefit in an improved velocity profile to vaned diffuser. I doubt it makes a difference in a vaneless diffuser. My thought is that is mostly marketing bullshit for the general application. You may not see the speed improvement from simply running a small diameter. I suggest more homework on the benefits of the extended tip, if any.

Thanks, Ron

[finiteparts]

Ron,

The challenge of turbocharger compressor design is that you are trying to maximize the mass flow, pressure ratio and efficiency while simultaneously trying to keep the rotor inertia and stresses as small as possible.

Pressure ratio is driven by the change in the absolute whirl between the inducer and exducer. If we don't have any inlet prewhirl and assume a constant mass flow and rotor speed, then the change in absolute whirl is due to the tip velocity and the blade backsweep. Backsweep helps to stabilize the compressor and enhances efficiency by reducing the relative discharge Mach number. Reducing the relative discharge Mach number reduces the friction generated in the impeller by the simple fact that the relative shear flow between the impeller surfaces and the bulk flow is reduced. It also helps to move some of the diffusion of the kinetic energy imparted by the rotor, from the diffuser to the rotor (i.e. increases the diffusion ratio of the impeller).

If we assume a constant backsweep, then the only variable left to change the pressure ratio (with the above assumptions) is the tip velocity. Since we assumed a fixed rotor speed, then the only way to change the tip speed is to increase the radius. But increasing tip diameter increases the rotor polar moment of inertia and the bore stress because both are driven by the amount of mass farthest away from the axis of rotation. Turbochargers are on a mission to reduce the throttle lag, which is a function of the polar moment of inertia, bearing drag, etc.. Scallops in the turbine wheel are used to reduce the mass of the rotor that is the farthest from the axis of rotation, but scallops would wreck the airflow coming out of the compressor.

The extended tip allows the tip diameter to increase with a minimal increase additional tip mass. The minimal increase in tip mass also minimizes the change to blade natural frequency and the pull stress imparted at the blade root and the impeller bore. Really, all these benefits are almost synonymous with the scallops on the turbine wheel, but you get an increase in stage pressure ratio to boot. So I wouldn't say that is is just "marketing bullshit"...it's just basic physics.

So just for comparison, I ran John's impeller at the two different tip diameters to see how things change. Now, take this with a grain of salt, because I didn't have all the geometry and flow parameters...but it should give a rough idea of the change in parameters due to just a change in tip radius. Also, I took the meanline tip diameter of the extended tip design. I ran the case at 65,000 rpm, because I wanted the lower tip radius to be at a pressure ratio o f around 3.5. Also, I did not include the diffuser pressure recovery in the calcs, I only looked at the gain in total pressure through the impeller.

Ok, so here is the results...The low tip diameter is the 150 mm that John wants to cut it to and the higher diameter is the average of the 151 mm and 163.11 mm, so 157.055 mm.

The first thing that comes out is that the power absorbed by the impeller drops by about 12% when he turns the extended tips off, this is from only a 4.7% reduction in tip speed. The absolute discharge angle is decreased 1.8 degrees, due to the reduction in absolute whirl by about 6.9%. Finally, the decrease in total pressure ratio is around 14%. Notice that I tried to use on percentage changes due to the fact that the absolute values will change with the B-width, backsweep angle, etc., of which all were assumed. So this case gives a representative case, not full on absolute numbers, but I think it does illustrate how a relatively small tip diameter change can have a large impact on the performance of the impeller stage.

I think John's logic is solid and think this is an interesting approach to re-matching the compressor to the turbine available power.

I hope this helps.

Chris

[racket]

Hi Ron

LOL.................I think your correct about the extended tip , its got more to do with "marketing" than anything else , theres are some benefits with the "moment of inertia" and a slight increase in PR compared to a similar sized rear walled wheel without extension, ..............but we don't really need to be concerned with moments of inertia to the same degree as with an automotive setup , and the PR can be easily replaced by a few thousand more rpm .

My concern is to get the diameter down a bit so that the size difference between comp and turb is less , a comp ~10% bigger in dia than the turb is probably about ideal , my current ratio ( 163 vs 129 mm ) is getting a tad too much , ..........torque required vs torque produced , and because I'm trying to stuff a lot of gases into the inducer of the turb wheel their approach angle has to be "high" which reduces the torque producing tangential component :-(

Cheers
John

[turboron]

Here's a link to a Garrett (Honeywell) patent paper that discusses the attributes, uses and benefits of ETT. It does not quantify the benefits.

www.faqs.org/patents/app/20080229742

Thanks, Ron

[racket]

Hi Ron

Heh heh .....That just about covers everything ;-)

Cheers
John

[finiteparts]

Ron,

The primary feature of that patent is the "Extended Leading Edge" and as such it is somewhat different. It does throw in the comment about the "trailing edge establishes a positive reverse clip angle...", but it fails to expound on the benefit from such a feature. That patent seems to come up routinely on the turbocharger forums, but unfortunately does not really address the extended tip technologies.

I have searched quite a bit for the Borg Warner's patent on the extended tip technology, and I am not sure that they actually have an issued patent on that. Since so many other manufacturers also use it and more recent BW publications seem to have dropped the term "patented" verbiage when talking about the ETT, it really is unlikely that they have any patent coverage.

It is interesting that a feature that is used on so many turbochargers out there, seriously lacks any technical literature expounding on it's benefits.

~ Chris

[turboron]

Chris/John, I have worked on centrifugal compressors for industrial compressors and helicopter gas turbines for many years. I have never seen this technology in those fields.

Thanks, Ron

[racket]

Hi Ron

Theres no need for it other than for automotive usage where milliseconds of acceleration time for the rotor are of any interest , for a more steady state usage you simply add a "full" bit of tip rather than a triangle along with the extra back wall and live with the few millisecond of extra acceleration rate , .............which on most gas turbines is irrelevant considering their rotor acceleration rates are measured in seconds.

The technology is squarely aimed at marketing to the performance/race car guys, and probably more so towards the wannabe racers :-)

Cheers
John

[enginewhisperer]

I think it lets them squeeze a little bit more performance out of an existing design without having to change the housing or noticeably reducing the transient response.
It'll act like a slightly larger wheel with less mass tradeoff. All the performance turbos seem to have it now.

[turboron]

John, my only concern is that if you just remove the extended tip you will not achieve the planned speed increase as if it were a diameter decrease. I would somehow discount the head increase at a given speed by the extended tip. For example, 10 - 20 percent of the head increase by a diameter increase. Once I estimated this contribution I would then reduce the remaining diameter to achieve the speed increase desired. Since it is a function of the speed squared the diameter decrease should be not be large.

Thanks, Ron

[racket]

Hi Ron

Theres gotta be a reduction in power required if its removed , the average dia of the extended tip is 157 mm , now 157 squard, divided by the new dia of 150 squared , gives a 9.5% difference .

157mm dia at 60,000 rpm which is my current 100% N1 would need to be increased to 62,800 rpm to restore the tip speed .

Turbine power is a function of mean blade speed times gas deflection , those extra 2,800 rpm mean more turbine power being produced .

My IGV stator should also require more rpm to restore my 3.5:1 PR compared to pre IGV , its a cumulative benefit I'm attempting here , a little bit here , a little bit there , hopefully just enough to get the engine up to full power , as it currently stands its a beast of an engine , burning lotsa fuel and making heaps of noise , but can/should/will do better .

I can't keep reducing the exducer diameter very far as I then run into inducer Mach problems once up to anything more than ~90% N1 , before fitting the IGV the relative Mach number at the inducer shroud at 60,000 rpm was >1.2 , this should have been reduced by the IGV , but probably restored once the rpm are increased after the extended tip is removed ............high Trim comp wheels aren't at their best once past ~2.5:1 PR :-(

Cheers
John

[racket]

Hi Guys

Done , ............she's now a 62 Trim wheel



Next job, impingement start.

Cheers
John

[turboron]

John, your use of the average diameter of the extended tip makes a lot of sense to me. Good point on the inducer Mach number.

Thanks, Ron