My turbine wheel

[skyfrog]

I've got another wax pattern, so I can test the wax turbine again, this time I use my index finger functioning as a brake to prevent the wheel from going wild. It works pretty good. Let me show you the video clip.

[skyfrog]

This is the Inconel 713 round bar I've got. Mostly Nickel, but much more expensive than Nickel. Give me one or two weeks, I'll convert this bar into turbine wheels.

[cudabean]

Why do they melt and cast the inconel in a vacuum? Anyone know? Does that effect the crystalization?

Another thing I was wondering: When you operate a turbine wheel, it obviously gets pretty hot and somewhat close to the melting point. Doesn't this change the temper of the material? Maybe temper is meaningless because normal operating temp is so hot...

cudabean

[JetSet]

Why do they melt and cast the inconel in a vacuum? Anyone know? Does that effect the crystalization?

I'm only guessing - but i think so that the metal doesnt oxidise and become a little 'impure' in places, giving it weak spots.

[Al Belli]

Hi Cudabean,

Vacuum melting preserves the alloy's chemical properties, and prevents the adsorption of gasses, which can cause reduction of the alloy's physical properties ( and cause cracks to form ).
Vacuum casting allows the molten alloy to fill the mold properly, especially the thin areas, since there is no air or gas in the mold that needs to be displaced by the metal.
Both melting and casting are done in the same chamber which has an internal door that can be closed to isolate the melting area so that the completed casting can be removed while the next batch of metal is being vacuum melted. This speeds up the process since the melting area is being evacuated while the next casting mold is being loaded into the casting part of the chamber. The access door is closed, and the casting area is pumped down by the " roughing pump " which is a very large Roots-type rotary pump fed through an oil diffusion pump. Two sets of pumps are used; one set for each section of the chamber, so the vacuum levels can be equalised before opening the internal door. The melt must be brought to pouring temperature before the casting mold is removed from the oven, to prevent the cooling of the mold, which will cause voids and cold-shuts due to the molten metal cooling excessively when it enters the thinner areas of the mold. For turbine blades with thin trailing edges, the molten metal appears as a white-hot molten stream that can be quite viscous even at that temperature. The evacuated mold allows the metal to flow easily into the thinner areas. For single high performance blades, a one-at-a-time process is used that has a mold that is open on the bottom, sitting on a gasketed knurled copper chill that causes nucleation of the molten metal at each point on the knurl. This produces directional solidification of the alloy, which provides longitudinal grains that run from the bottom to the top of the blade. Since metals fail at the grain boundary, these blades resist the rotationally induced forces better due to the grain boundaries being parallel to the tensile forces.

I'll stop at this point, since I probably told You more than You wanted to know!!!!!!!

Al Belli

[hinote]

The turbine wheel doesn't actually get that hot, in a properly running turbine. It's not unusal to have the turbine scroll be glowing very slightly, but with the turbine still black at the outlet because of the pressure drop.

It will explode well before it gets anywhere near melting.

Plus, many turbines employ hollow-core blades to allow internal air cooling.

Bill H.
Acoustic Propulsion Concepts

".......some day soon we'll be flying airplanes powered by pulsejets."

[Tom]

I have seen some of the turbine blades from a Rolls Royce engine a month or two ago on a trip to the Cavendish Labs in Cambridge, about 10 mins away. I cant imagine how they cast those internal channels, they are a sight to behold. Al, Bill? Anyone know?

Tom

[skyfrog]

I have seen some of the turbine blades from a Rolls Royce engine a month or two ago on a trip to the Cavendish Labs in Cambridge, about 10 mins away. I cant imagine how they cast those internal channels, they are a sight to behold. Al, Bill? Anyone know?

Tom

I read from a japanese investment casting book saying that to get the internal channels for cooling we can use ceramic core. Sometimes those castings get extremely expensive and would cost US$ 800 a kilogram. Fortunately we don't need to tangle with that monster (ceramic core), because in our case the typical temperatrue of the combustion gas entering tubine stage is around 700 deg. C only, very well under the melting point (1400 deg. C).

[hinote]

Sometimes those castings get extremely expensive and would cost US$ 800 a kilogram.

My brother went to work for a company in the Midwest (some years back) that was casting turbine blades for one of the big turbofan engines. They were valued at something like $35,000 each.

There was a reject barrel with over 100 of these, scheduled for scrap. My brother devised a repair procedure that saved all of them, and saved the company something in excess of $3.5 million in the process.

Bill H.
Acoustic Propulsion Concepts


".......some day soon we'll be flying airplanes powered by pulsejets."

[skyfrog]

There was a reject barrel with over 100 of these, scheduled for scrap. My brother devised a repair procedure that saved all of them, and saved the company something in excess of $3.5 million in the process.

Wow, most companies here are SMEs(small and medium sized enterprise), they don't even worth $3.5 million. Can I meet your brother ? ;-)

Manufacturing of big jet engines are technology intensive as well as capital intensive. So glad I am into "toy engine" category only, instead of into commercial or military engine category.

[Al Belli]

Hi,
As an example I'll describe what I remember about a first stage hot zone blade for a P&W JT9-D. ( I'm thinking back ~32 Yrs. )
There are 7 round cooling passages formed by 0.032" Dia. quartz rods that have been bent to conform to the centerline of the blade's airfiol chord. They extend from the top and bottom of the investment wax pattern, and are retained by the investment ceramic. after casting, the quartz rods are dissolved in a high pressure autoclave containing a solution of sodium hydroxide in water. For other blades, specially shaped internal passages are formed with a silicon dioxide based molded ceramic. Very thin core sections for the trailing edge cooling passages were made of punched silicon ( leaving round buttons that tied the upper and lower airfoils together ) that was subsequently placed in a high temperature oxidizing furnace to convert the silicon to SiO2. These thin cores were attached to the main cores using a ceramic cement. After casting these cores were also dissolved out using the autoclaves.
The castings were then X-rayed to detect any broken cores which would prevent proper flow of cooling air. The castings were then given a waterflow test to determine sufficient coolant flowrate. The trailing edge coolant flow holes were then " drilled " through to the main coolant flow passage using E.D.M. or electrochemical drilling. This was necessary due to the small contact angle with the blade's airfoil. this drilling was subsequently done by P&W, not by the casting facility.
As You might imagine, many castings were consigned to the remelt scrap barrels. The unmachined casting for the JT9-D was sold to P&W for about $54.00 in 1973 . The raw alloy cost was about $27.00 per pound, so the cost of the casting was quite reasonable considering the amount of manufacturing cost that was required.

Al Belli

[skyfrog]

Thanks for sharing, Al.

Internal cooling, thermal barrier coating, directional solidified structure, single crystal, ceramic blades..., I wish these advanced technologies won't be used in my engine. Let simplicity and low cost be always with homebuilders.

Don't get me wrong, I am not saying that I don't wish one day things like spaceshipone can be homebuild.

[skyfrog]

I know one day homebuilders will have serious confront with those monsters mentioned above.

Sorry for the casting delay, it takes time optimizing the wax injection parameters and mold tunning. I and those foundry guys worked together closely and finally we've got perfect wax patterns ready for casting. Due to traffic jam in their production line (yes, recession has gone), mine will be sent casting by next week, out of 304 only.

Two stage turbines for homebuilders ? There will be one day soon.

[skyfrog]

Phew! The key component of my jet engine is born !

I went to the foundry this afternoon and eye witnessed the birth of my very first turbine wheel. What an exciting day ! After this my hearing loss a bit, because in which environment it was so noisy we had to shout to each other when communicating. And fortunately it isn't summer or I'd be like roasted frog. No the foundry environment doesn't really like hell, but I wouldn't chose it as my working place if I have chance.

I'll show you guys the picture tomorrow.

[skyfrog]

This is the first casting sample out of 304. Since it's not heat-resistant material, it will be used for static display only. Nothing innovative I must admit, but to get hold of the knowhow for building small turbine wheel, IMO, for a small company like us it's definitely a historical milestone. After fine tunning the casting process, I'll use that stock Inconel 713 material to make the wheel.