Fan... tastic. ;-). I`ve made my diploma Work about the Fan Blade Alloy, TI AL 6V4 bending fatigue stress at 200°C and 400°C. 30 years ago. We designed a tast bed for this. In the beginning we tested the fatigue of the test equipment. Science is going forward. Besides, this ORENDA badge triggered me. In 1978 i had a flight with a guy, runnung a Tiger Moth at Lahr Canadian Airbase/Starfighter in Germany. He was an employee of Orenda Companie. Cheers Axel
Fantastic! For abrasion, (and apparently strength) metalize the leading edge like the GE9X Maybe even on the Fir Tree/root? From GE- "The new material incorporates a higher stiffness carbon fiber and a new epoxy resin. The leading edge material will also be modified from titanium to a steel alloy to further enhance the blade's strength."
5 месяцев назад+2
I'm wondering that kind of processing uses a type of electrostatic inconel vapor deposition like ESAVD.
In an industrial application, with intake filtration, erosion by of the aerofoil by dust, etc, shouldn't be a problem. What might happen to the dovetail root, in terms of wear, is probably the biggest uncertainty for me, with the integrity of the composite structure with time a close second. The problem will, therefore, be demonstrating to prospective operators that the blades will last tens of thousands of hours between overhauls. That will mean some form of representative endurance testing - and just driving a bladed stage 1 disc at max engine rpm in isolation will not be representative.
I've had yet another thought about a rig test of a bladed disc. There have been suggestions of a small number of composite blades spaced around the row. No, it really needs a full set to be representative. One clever way of detuning a problem blade row is to have blades of different natural frquencies spaced around the row. The first stage of a small turbofan with which I was familiar some years ago had a mixture of 'A' and 'B' blades to do this.
11:34 Sounds like a great idea for the long term fatigue testing to have it populated with a minimum number of blades and driven by an electric motor. One question, does it make sense to have the test blade to have a leading companion blade to try to setup the airflow field that is as reasonably comparable to a fully populated assembly? If the airflow fields are too different it might produce different vibrational forcing that could affect the fatigue results. Just wondering, but I don’t know what I am talking about. Also with a more experimental test rig it might be easier to get high speed video or get some strain gauges incorporated into the setup to check on some of the dynamics going on. Also a possibility is to use a pair of optical reflective probes, one at the root and one looking radially inward towards the blade tip to get some dynamic behavior monitoring up on a digital scope.
I was able to experience a full aug F-16 engine run at night. It's like a white noise generator at 140db, your entire body just being blasted with concussive sound the whole time it's up over 80% throttle. It's something to experience first-hand for sure.
In electronics for military use there is HALT (highly accelerated life testing) and HAST (highly accelerated stress testing). MIL-HDBK-217 might be worth a cursory glance and I think is currently at revision F (public domain on the internet). If it can be done for electronics it can be done for mechanical systems is my guess. It could potentially shorten your test time.
And now a comment on my comments. As a young and very inexperienced designer, in the days when we had things called drawing boards, and dinosaurs roamed the earth, I was told that, "You think with a pencil, you design with an eraser." This is effectively what I've been doing mentally with my thoughts about composite blade testing. However, I've changed one word for the benefit of my more delicate North American readers. What was actually said to me was, "you design with a rubber." I'm aware that the word commonly used for a pencil eraser on this side of the pond means something different over there.
So I would also think you would need the Stator as well for a electric test stand for the interrupted airflow and what not. Make for one hell of a shop fan though
After I suggested using a complete J79 compressor for a rig test, with all of the blades/vanes removed after stage 1, I started to have doubts. So, I thought I'd better remind myself of the actual layout of the compressor. On looking at a cross-section, it is plainly obvious that the barely compressed flow from stage 1 will be constricted long before it reaches the mid frame, and I've now realised that I must now screw my designer's head on more firmly. Consequently, I'm now thinking in terms of cutting some large ports in the casing to exhaust the flow, which might need some external support or reinforcement as a result. Or perhaps even more major surgery might be considered, with both the casing and the rotor 'truncated'? However, this would mean numerous new parts having to be designed and made. It could be a great challenge for a designer. I'm not volunteering, but I'll stand on the sidelines to offer support (and constructive criticism, if necessary).
The described test rig sounds like a cool project. It is just for centrifugal loads w/o compression loads and only 4 blades it would certainly seem doable with a smallish electric motor. You could set up a live stream on your Patreon with a timer.
If you've been paying attention, "smallish" may or may not be applicable. The bets are between 50 HP and 1000 HP motor will be needed. What's your recommendation?
@@AgentJayZ as you say, the engineers can figure that out. That said, my back of a pizza box thoughts are maximum would 4/21*1000=~190HP since you would be running 4 of 21 blades (I think I counted those correctly) of a compressor that nominally takes 1000HP to drive. I think it would be less since the blades wouldn't be compressing anything although still moving air. Definitely need some sort of gears or belts to get it to your desired RPM since pretty much the fastest A/C motors are 3600 RPM. That's 3-6% power loss. If I was ordering one w/o consulting an application engineer, I would order a 200HP one and be confident it was enough. Best guess as to what will actually be required? 100HP , so smallish
21 blades normally on a wheel, use every 7th position. i.e. 3 blades. Spinning would only identify creep or failure due to centripetal effect, not combined pressure action with centripetal. Methinks you will need some perforated disc behind the wheel to simulate (somewhat) compressor action by the blades. Maybe spare stators?
@BerndFelsche I've recommended what I would describe as an engine parts rig, with a front frame, VIGVs, and stage 1 VSVs. Driving a bladed stage 1 disc in fresh air just wouldn't be representative of engine conditions.
Please see my reply to Mr BerndFelsche. I've recommended what I describe as an engine parts rig, with a front frame, VIGVs, and stage 1 VSVs. Maybe the easiest way to achieve this is to have a complete compressor assembly, including a mid frame, but with all the blades and vanes removed downstream of stage 1. PS I've gone on to post further thoughts.
And now my thoughts about wear in the dovetail root, with a suggestion of electroplating being offered elsewhere. My starting point must be that I learnt as a very new and very 'green' designer that fretting and wear was problematical in industrial and marine engines, as compared to their aero equivalents. They run the whole time in air at ground/sea level and consequently experience much higher cumulative aerodynamic and vibratory loads, hence fretting and wear being such a problem. Modifications to introduce 'hard' wear-resistant coatings were necessary to give acceptable lives. Another mod which was commonly used (and in aero applications) was the introduction of anti-fret liners into compressor vane fixings. Perhaps something similar might be needed for these blades - but first quantify the problem. A static rig test, vibrating a sample root in a disc slot might be a starting point, before going to a rotating rig test, as previously suggested. PS This has kept me occupied as the German countryside has rolled by.
G'day Jay, Yay Team ! I'm very impressed with the performance of those Blade-Root Attachments..., great stuff. Regards the Electric Fan Test... Unless the Compressor Blades have a Ring of Stator-Vanes in front and behind them, then they won't be Compressing Anything, merely Displacing, and accelerating a Constant flow of Air moving back through the Disc. Actual Axial-Flow Compressor Blades function by kind of Scraping and squeezing the "bit of Air" which the Blade's Leading Edge cuts off from Airmass ahead of the Upstream Stator's Trailing Edge..., Off against and under the Downstream Stator Vane's Leading-Edge, with the Blade's pasding Trailing Edge... No Stators to "work against"..., with Each Blade generating a Leading and Trailing Compression/Decompression Wave Against EVERY Stator Vane in the Ring ; would mean that you'd be testing a Fan-Blades rather than Compressor Blades - and ideally the Inlet Guide-Vanes should be Adjustable, to check for Cyclic Stress-Concentrations across the entire range of actual operationally working Inlet-Vane Angles... And, to actuarially duplicate what a Working Blade experiences..., Would not one "Need" to have ALL the Blades on the Compressor Wheel, to be able to Replicate the environment of Turbulent Eddies being encountered by the Composite Leading-Edge - resulting from the "Compression-Scrapings" left in it's Path, from at least the Trailing Edge of the Blade ahead of the one to be tested... So, how about 2 Rows of Stators, 2 Composite Test Blades, each with a Metal Blade ahead of it on the Wheel, with the Pairs trying to be 180° across the Shaft from each other....? If the 12-blade Compressor 14:17 Array indeed uses 1,000 Hp of Twistiness at 7,000 RPM...(?) ; then 4 Blades between 2 Rings of Vanes should suck a "Mere" 333.3 Hp @ 7,000 RPM, and if your Grid runs at 50 Cycles/second then Electric Motors run happiest at 3,000 RPM, and 3,600 if the Grid has 60 Sine Waves per second.... You'll want a 2.35 or 1.95 to 1 Step-Up Gear Drive-Train between your 800 Hp Electric Motor. And at Half a Columbus worth of Wattzies (1492÷2 = 746...) At per each Horsie-Power of Twistiez, Then 799.2 Hp x 746 Watts = "Only" 596.2032 KillerWhats of Torque emerging from your Motor - Which might be 90% Efficient, at optimum RPM & Load. Which will be only happening at about 60% of it's Rated Output. So it might be Wisest to begin by Aiming at a 1,000 Hp Motor, Running at 60% Torque, To spin your Gear-Train which Delivers a Reliable 350 Hp or so at 7,000 RPM, To test a third of a Compressor Wheel worth of Blades... Half of which will be Composite Test Objects - Each with ut's Control In front of it on the Wheel... That strikes me as being a Shit-Ton of Electromechanical Engineering, But afterwards the Rig would still exist, ready to conduct further tests for years/decades to come. Burning the Fuel might be cheaper, and faster, Once, And it will cost the same in burnt Propane, every time after that. How many times will The nameless Client be wanting to Repeat the Test, on Similar - but different, "More Promising" Blade Pairs...? It appears to be a case of Do it Properly, or Burn (waste) Shitpots of Fuel Every time the Boffins have a sequential upgrade to their last Brainwave...(?) ! You'll work it out... Such is life, Have a good one... Stay safe. ;-p Ciao !
Dang... I hadn't considered the effect of a blade breaking and FOD'ing out the engine if you are standing down wind ("down stream"?). Does the shrapnel go directly to the rear, or could a blade fragment catch the air and deflect sideways? I think I'd stick to the control room and not tempt fate. 😄
@SkyhawkSteve This is a nitpick from this old pedant: if nothing 'foreign' (ie, from outside the engine) has caused the failure, then it can't be FOD (foreign object damage). It's not a commonly used term, but DOD was used in my day. I'll leave you to guess what the 'D' stands for.
@@grahamj9101 fair enough. Maybe "entropy based damage"? "Internal Object Damage"? I'm still curious about how much stuff comes flying out the back of the engine, though!
Graham is correct, as usual. Since these are first stage compressor blades, and they are more likely to break up than steel blades, I think it would be very likely for debris to come out the front of the engine. Any damage to the later stages of the compressor would likely cause a stall, helping to blow any composite or metal debris forward. It may not be too noticeable in the videos, but everybody was more careful than usual to not be in front of or behind the engine when it was running.
@grahamj9101 FOD is also commonly used as in Foreign Object or Debris Damage as a catch all. I agree with you though as I had the terms FOD and DOD through the RAF and it causes the brain to read it as Foreign Object Damage Damage WHATNOW!
Agent Z, You are a “legend” in our A&P students’ community! Reference this experimental path, my vote is for: a. Populating the compressor with FULL (21 I guess) set of new blades (to get the relatively high "n" required for statistical analysis) at 125% of maximum operational RPM; for 1500 hours (slightly over two months 24/7) using an electrically driven motor. b. Stop experiment, remove and completely inspect 3 blade only at a time (7 blade apart) very thoroughly by all NDT techniques available and note deficiencies, then inspect another set of three blade the same way, and proceed until all 21 blades have been inspected (to get good data for statistical analysis on how to, and how many blades to inspect and how often to do it). c. If still serviceable, repopulate the compressor shafts with the FULL set of blades again; with the test compressor stage enclosed in vacuum sealed chamber and spun to produce 200% of design centrifugal load for 6 hours. d. Stop experiment and retest blades as in "b" above. e. If still serviceable, start again as set up for "c" above, run at 200% of design centrifugal load for one more hour; then without stopping the experiment, gradually reduce the vacuum (to reintroduce aerodynamic load) at steps of doubling such aerodynamic load in each stage of vacuum reduction. and remain at such load for one hour, until arriving at ambient pressure level. It will take some resources and cost to do it; but in the long run will arrive at practical implementation and certification quicker on the overall project path. In addition, you may be jumping ahead in development of test rigs and protocols for servicing and inspecting these blades once accepted into routine operation service. Al
You are putting a lot of faith in the disc. I would not want to be in the same building with that thing at 15,000 rpm. These blades will be tested for wear at 100% design speed for up to a year of running. In service, the engines never overspeed.
Fan blades only with inlaid titanium sheet leading and training edge reinforcement. All the gas path blades are fully metallic alloys the vary for location (based on temperature, size and speed).
If you added a stator you would introduce minute pressure differences as the blades came together and passed. So (rough math) 24 blade fan at 6000 rpm=144,000 pressure changes per minute. MotoGP has banned the use of CF rims on race bikes because the of the unpredictable catastrophic failures that can occur on the wheels. There is a wheel, or there is shrapnel, there is no in-between. I am sure engineering can figure this out, but get lots of testing done.
As I've suggested elsewhere, a rig test of a bladed disc would need a front frame and a VIGV row in front of it, and a stage 1 VSV behind it, so as to present a representative flow field to the blades. I'm going to leave the guys at S&S Turbines to decide how to do this.
That is both the beauty and the curse of carbon composites. While incredibly strong--you'd need a micrometer to check for any elongation--I tend to liken CC parts to bridge girders. They will do their job, and do it well, for dozens of years without fail. However as they age (mostly environmental impact, notably oxidation, heat fatigue and weather exposure) they can fail in the most strikingly catastrophic way without warning. In that moment, they lose a ton of energy though and the parts don't travel quite as freakishly shrapnel-esque like metallic components. The one saving grace so long as you're fifteen feet away for so. None of that is a knock on this project, rather a generalization of CC parts in hard use universal to the material. I wish I'd had high-speed camera footage during vibrational testing a few years back when I got a set of parts absolutely singing at 65K rpm and found the upper threshold of my bearings. Pow!
They will not be put in service until a "life" is determined. If they go at 4 years, the replacement will take place at two.... Easy. Private owners can do what they like. How many cars do you see every day with bald tires on the freeway?
Depends upon the power setting. At full power, the 1500 core will burn the equivalent of about 35 gallons of jet fuel or diesel per minute. We don't actually measure fuel consumption on gaseous fueled engines, because they are used on natural gas pipeline compressor stations, where their fuel is taken directly from the pipeline.
@@AgentJayZ Thanks! At first I was thinking that but thought it couldn't be as pounds are usually abbreviated as lb and In my country the gas that goes to power plants is usually measured in m2/h. Will check those books.
Cheaper to manufacture. Non OEMs can go mad with old engines as the operators are not contractually bound to use manufacturer's parts and overhaul facilities. Plus industrial engines are not bound by the same compliance regs as aero so it's massively easier to get alternate parts into them rather than genuine if they are still manufactured.
Watch the previous videos. In short, they're aiming at 1% or so fuel savings. 30MW (or so) thermal is about a kilogram of fuel a second, ten grams a second saved. For an industrial engine, that's 300 tons of fuel saved a year of continuous operation, which is a largeish slice of a million dollars.
The hope is to be able to increase performance or economy by introducing different airfoil shapes... not necessarily for 1500s, but for other engines. I tried to explain all that, but I'm kinda used to people not listening to the details...
Hot stuff!❤️🔥 Who might take interest in this hmmmmm, GE maybe 🤔 Lockheed Martin maybe 🤔 S&S Turbines Maybe 🤔 and maybe PierCarlo and friends @ “ Starfighter’s Space “ as old Skool TV “ Laugh In” Would say “Very Interesting “🧐 OOPS I might have let the J79 out of the bag 💼 er cat 🐱.
Here are my further thoughts, and I continue to maintain that the only truly representative test will be a field trials installation. However, I also maintain that a lot more testing than a few hours engine running in the test cell with even a full set of blades will be necessary before an operator agrees to a field trials exercise. So what can you do that's near 'representative', without tying up the test cell? I'm sorry, but just spinning a bladed stage 1 disc up to engine speed will not be 'representative'. The very fact that 50 to 200 HP might be sufficient to do the job should tell you that it's not. The composite blades need to be absorbing the power they would in the engine, which means they need to be subject to an equivalent of the engine installation, but without the rest of the engine. They need a front frame with its eight struts and a row of VSVs in front of them, together with a row of stage 1 vanes behind them. I imagine you and S&S Turbines have enough bits to build such an engine parts rig. However, it would probably need more than a couple of hundred HP to drive it, which starts to sound expensive, but far cheaper than engine running - and it frees up the test cell.
Great to see you here adding your many years of jet engine design experience to the discussion. I would find it very educational to have some input on what types of instrumentation are used on blade development. The test rig that MAN uses for compressor development (youtube video of surge testing) shows images of a tip timing system. I assume strain gauges on rotating parts are possible (though I don’t know how this is done). I am relieved that my “intuition” that a depopulated stage would not be representative of the minimal requirements for fatigue testing. But then again, I don’t have any experience with these issues.
@Mentaculus42 I've been in retirement for 20 years now, so my knowledge of instrumentation is dated. In my early days, the signals from straingauged blades were taken via slip rings. In the later stages of my career, the use of a contactless radio system had come in. On larger compressor blades, the measurement of tip deflections was also used, with a grid embedded in the rotor path above the blade row. However, I do have contact with more recently retired colleagues who specialised in instrumentation design. I can ask them.
@grahamj9101 current rotating s/g telemetry set ups are still the same as the radio/antennae methods used 20 years ago. Certainly atm there are new telemetry units in design or moving to F status. Do you recognise DIGBERT and ALBERT at all?
@@paulpower3766 Could you expand on your comment, I find it rather cryptic with the acronyms that don’t seem to have google search results. If you know something about the topic then I am interested in learning but I couldn’t seem to understand what you were saying. grahamj has always provided very educational information that always makes sense in the context of my own engineering background.
realizing we're still a way from aviation...are you allowed to comment on grain structure and how they go about 'forging' the blades? As shown narrow neck, potential different quick airfoil just seems 'wow'!
I'm just a guest commenter. I have met these guys. They are materials engineers, and have previous experience making internal engine parts for top fuel cars. Yeah, so let's all go a bit easier on our armchair worries.
Coming from a performance boat background, I'll give an opinion, inaccurate as it may be. With that said, my money is on the carbon blades being vacuum infused construction, with cure time controlled via vaiable heatl.
@AgentJayZ As I've been telling you, running just three or four blades will tell you relatively little. You really need a full set with a front frame, VIGVs, and stage 1 VSVs to do some meaningful testing
They need to do some testing to see what wear effects repeated starting and stopping does to the root. I would imagine the repeated loading and unloading of centripetal forces would be a lot harder on the resin than it would on steel.
A fiber plate will be as strong as metal. Until it gets a nic or microscopic scratch. Then it will shatter. Will also have terrible erosion qualities in sand or salty air
I watched them and I remember an explanation on the economical benefits but I don't remember in which one and in what minute. might be useful for all viewers to highlight the information in the descriptions. thanks.
I still think that these Carbon-fiber blades should be electroplated.. Some type of chrome-nickel (let the engineers figure that out).. 'cause I know they want these blades airborne.. well just for the heavy dust..
So who exactly wants them airborne? There must be just a handful of J79 powered aicraft flying, which will do just a few tens of hours a year at air shows. Surely the market that Robin Sipe and S&S Turbines are targeting is the industrial gas generator repair and overhaul market? Otherwise why bother?
One only needs to provide the power to turn the one stage (assuming it’s just that one disk mount in a containment rig of some kind). At the main OEMs, the containment test stands use large electric motors to turn single disks for testing.
NotA: You get nothing. A turbine is not a compressor, They are in fact exactly opposite in function, One is the bat, and the other is the glove. The air is the ball.
![I.N "HALLUCINATION" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/n5B5q1Hwt_U/mqdefault.jpg)
Extremely exciting to see this process Jay. Thanks for taking us along.
What a unique opportunity to participate in a potential game changing advancement in the turbine world ! I look forward to see the results.
Fan... tastic. ;-). I`ve made my diploma Work about the Fan Blade Alloy, TI AL 6V4 bending fatigue stress at 200°C and 400°C. 30 years ago. We designed a tast bed for this. In the beginning we tested the fatigue of the test equipment. Science is going forward. Besides, this ORENDA badge triggered me. In 1978 i had a flight with a guy, runnung a Tiger Moth at Lahr Canadian Airbase/Starfighter in Germany. He was an employee of Orenda Companie. Cheers Axel
There is so much cool engineering going on here that it is scary exciting!!!
Fantastic! For abrasion, (and apparently strength) metalize the leading edge like the GE9X Maybe even on the Fir Tree/root? From GE- "The new material incorporates a higher stiffness carbon fiber and a new epoxy resin. The leading edge material will also be modified from titanium to a steel alloy to further enhance the blade's strength."
I'm wondering that kind of processing uses a type of electrostatic inconel vapor deposition like ESAVD.
In an industrial application, with intake filtration, erosion by of the aerofoil by dust, etc, shouldn't be a problem. What might happen to the dovetail root, in terms of wear, is probably the biggest uncertainty for me, with the integrity of the composite structure with time a close second.
The problem will, therefore, be demonstrating to prospective operators that the blades will last tens of thousands of hours between overhauls. That will mean some form of representative endurance testing - and just driving a bladed stage 1 disc at max engine rpm in isolation will not be representative.
This would be very exciting to be a part of. I'm glad these guys are letting you document this.
Jay, that is impressive !!
Long time see. So glad yall still cranking. Thanks
I have wondered for years and asked around with no good answer back to why this haven't been tested before, and now finally I see it.
I can see the excitement. Heck I’m excited. I run LM6000’s and I can imagine the future for the power generation industry.
I've had yet another thought about a rig test of a bladed disc. There have been suggestions of a small number of composite blades spaced around the row. No, it really needs a full set to be representative. One clever way of detuning a problem blade row is to have blades of different natural frquencies spaced around the row.
The first stage of a small turbofan with which I was familiar some years ago had a mixture of 'A' and 'B' blades to do this.
11:34 Sounds like a great idea for the long term fatigue testing to have it populated with a minimum number of blades and driven by an electric motor. One question, does it make sense to have the test blade to have a leading companion blade to try to setup the airflow field that is as reasonably comparable to a fully populated assembly? If the airflow fields are too different it might produce different vibrational forcing that could affect the fatigue results. Just wondering, but I don’t know what I am talking about. Also with a more experimental test rig it might be easier to get high speed video or get some strain gauges incorporated into the setup to check on some of the dynamics going on.
Also a possibility is to use a pair of optical reflective probes, one at the root and one looking radially inward towards the blade tip to get some dynamic behavior monitoring up on a digital scope.
Glad to see the exuberance in your work. Will be watching
this whole series is amazing and awesome! keep going!!
THANK YOU AGENTJAYZ
I was able to experience a full aug F-16 engine run at night. It's like a white noise generator at 140db, your entire body just being blasted with concussive sound the whole time it's up over 80% throttle. It's something to experience first-hand for sure.
Cheers from Edmonton.
Don't tell anybody, but I was born in Edmonton. I will deny that.
I am pre-denying that.
Still, better than Calgary.
@@AgentJayZ 😁😁😁
Holy smokes 🤯
In electronics for military use there is HALT (highly accelerated life testing) and HAST (highly accelerated stress testing). MIL-HDBK-217 might be worth a cursory glance and I think is currently at revision F (public domain on the internet). If it can be done for electronics it can be done for mechanical systems is my guess. It could potentially shorten your test time.
And now a comment on my comments.
As a young and very inexperienced designer, in the days when we had things called drawing boards, and dinosaurs roamed the earth, I was told that, "You think with a pencil, you design with an eraser."
This is effectively what I've been doing mentally with my thoughts about composite blade testing.
However, I've changed one word for the benefit of my more delicate North American readers.
What was actually said to me was, "you design with a rubber." I'm aware that the word commonly used for a pencil eraser on this side of the pond means something different over there.
lovely thing that is!!!
Another great video.. Thanks for sharing. I wonder how well the blades do with small FOD impacts. My gut feeling is they'd do really well.
I think you're about due for some more fill for Jet Wash Alley.
So I would also think you would need the Stator as well for a electric test stand for the interrupted airflow and what not. Make for one hell of a shop fan though
After I suggested using a complete J79 compressor for a rig test, with all of the blades/vanes removed after stage 1, I started to have doubts. So, I thought I'd better remind myself of the actual layout of the compressor.
On looking at a cross-section, it is plainly obvious that the barely compressed flow from stage 1 will be constricted long before it reaches the mid frame, and I've now realised that I must now screw my designer's head on more firmly.
Consequently, I'm now thinking in terms of cutting some large ports in the casing to exhaust the flow, which might need some external support or reinforcement as a result.
Or perhaps even more major surgery might be considered, with both the casing and the rotor 'truncated'? However, this would mean numerous new parts having to be designed and made. It could be a great challenge for a designer.
I'm not volunteering, but I'll stand on the sidelines to offer support (and constructive criticism, if necessary).
The described test rig sounds like a cool project. It is just for centrifugal loads w/o compression loads and only 4 blades it would certainly seem doable with a smallish electric motor.
You could set up a live stream on your Patreon with a timer.
If you've been paying attention, "smallish" may or may not be applicable. The bets are between 50 HP and 1000 HP motor will be needed. What's your recommendation?
@@AgentJayZ as you say, the engineers can figure that out. That said, my back of a pizza box thoughts are maximum would 4/21*1000=~190HP since you would be running 4 of 21 blades (I think I counted those correctly) of a compressor that nominally takes 1000HP to drive.
I think it would be less since the blades wouldn't be compressing anything although still moving air. Definitely need some sort of gears or belts to get it to your desired RPM since pretty much the fastest A/C motors are 3600 RPM. That's 3-6% power loss.
If I was ordering one w/o consulting an application engineer, I would order a 200HP one and be confident it was enough.
Best guess as to what will actually be required? 100HP , so smallish
I think you are right on, there.
21 blades normally on a wheel, use every 7th position. i.e. 3 blades. Spinning would only identify creep or failure due to centripetal effect, not combined pressure action with centripetal. Methinks you will need some perforated disc behind the wheel to simulate (somewhat) compressor action by the blades. Maybe spare stators?
The engine is actually running, so there is some pressure load. I'd have to go back and see if there are stages missing or anything, but...
@BerndFelsche I've recommended what I would describe as an engine parts rig, with a front frame, VIGVs, and stage 1 VSVs. Driving a bladed stage 1 disc in fresh air just wouldn't be representative of engine conditions.
Please see my reply to Mr BerndFelsche. I've recommended what I describe as an engine parts rig, with a front frame, VIGVs, and stage 1 VSVs. Maybe the easiest way to achieve this is to have a complete compressor assembly, including a mid frame, but with all the blades and vanes removed downstream of stage 1.
PS I've gone on to post further thoughts.
rofl... go back, yes. Then go away.
And now my thoughts about wear in the dovetail root, with a suggestion of electroplating being offered elsewhere.
My starting point must be that I learnt as a very new and very 'green' designer that fretting and wear was problematical in industrial and marine engines, as compared to their aero equivalents. They run the whole time in air at ground/sea level and consequently experience much higher cumulative aerodynamic and vibratory loads, hence fretting and wear being such a problem. Modifications to introduce 'hard' wear-resistant coatings were necessary to give acceptable lives. Another mod which was commonly used (and in aero applications) was the introduction of anti-fret liners into compressor vane fixings. Perhaps something similar might be needed for these blades - but first quantify the problem. A static rig test, vibrating a sample root in a disc slot might be a starting point, before going to a rotating rig test, as previously suggested.
PS This has kept me occupied as the German countryside has rolled by.
@@grahamj9101 anti fret liners are still in use in wide body designs
G'day Jay,
Yay Team !
I'm very impressed with the performance of those Blade-Root Attachments..., great stuff.
Regards the Electric Fan Test...
Unless the Compressor Blades have a Ring of Stator-Vanes in front and behind them, then they won't be
Compressing
Anything, merely
Displacing, and accelerating a
Constant flow of Air moving back through the Disc.
Actual Axial-Flow Compressor Blades function by kind of
Scraping and squeezing the "bit of Air" which the Blade's
Leading Edge cuts off from Airmass ahead of the Upstream Stator's Trailing Edge...,
Off against and under the Downstream Stator Vane's Leading-Edge, with the Blade's pasding Trailing Edge...
No Stators to "work against"..., with
Each Blade generating a
Leading and Trailing Compression/Decompression Wave
Against
EVERY
Stator Vane in the Ring ; would mean that you'd be testing a Fan-Blades rather than Compressor Blades - and ideally the Inlet Guide-Vanes should be
Adjustable, to check for
Cyclic Stress-Concentrations across the entire range of actual operationally working Inlet-Vane Angles...
And, to actuarially duplicate what a
Working Blade experiences...,
Would not one
"Need" to have
ALL the Blades on the Compressor Wheel, to be able to
Replicate the environment of
Turbulent Eddies being encountered by the Composite Leading-Edge - resulting from the
"Compression-Scrapings" left in it's Path, from at least the
Trailing Edge of the
Blade ahead of the one to be tested...
So, how about 2 Rows of Stators, 2 Composite Test Blades, each with a Metal Blade ahead of it on the Wheel, with the
Pairs trying to be 180° across the Shaft from each other....?
If the 12-blade Compressor 14:17 Array indeed uses 1,000 Hp of Twistiness at 7,000 RPM...(?) ; then 4 Blades between 2 Rings of Vanes should suck a
"Mere" 333.3 Hp @ 7,000 RPM,
and if your Grid runs at 50 Cycles/second then Electric Motors run happiest at 3,000 RPM, and 3,600 if the Grid has 60 Sine Waves per second....
You'll want a 2.35 or 1.95 to 1 Step-Up Gear Drive-Train between your
800 Hp
Electric Motor.
And at
Half a Columbus worth of
Wattzies
(1492÷2 = 746...)
At per each
Horsie-Power of Twistiez,
Then
799.2 Hp x 746 Watts =
"Only"
596.2032
KillerWhats of
Torque emerging from your
Motor -
Which might be 90%
Efficient, at optimum RPM & Load.
Which will be only happening at about 60% of it's
Rated
Output.
So it might be
Wisest to begin by
Aiming at a
1,000 Hp Motor,
Running at 60% Torque,
To spin your Gear-Train which
Delivers a
Reliable
350 Hp or so at
7,000 RPM,
To test a third of a
Compressor Wheel worth of Blades...
Half of which will be
Composite Test Objects -
Each with ut's
Control
In front of it on the Wheel...
That strikes me as being a
Shit-Ton of
Electromechanical Engineering,
But afterwards the Rig would still exist, ready to conduct further tests for years/decades to come.
Burning the Fuel might be cheaper, and faster,
Once,
And it will cost the same in burnt Propane, every time after that.
How many times will
The nameless
Client be wanting to
Repeat the Test, on
Similar - but different,
"More Promising"
Blade Pairs...?
It appears to be a case of
Do it
Properly, or
Burn (waste)
Shitpots of Fuel
Every time the
Boffins have a sequential upgrade to their last
Brainwave...(?) !
You'll work it out...
Such is life,
Have a good one...
Stay safe.
;-p
Ciao !
Good luck 🤞🏼
Very cool.
With the change in mass from having a full set these carbon comp blades bring any other advantages over the steel ones?
We'll see. The weight difference is not significant for engine performance.
Are OEM parts still in production for the J79?
No, but there are many NOS parts still available.
You could do with a partner in a very hot country, lots of people would enjoy the breeze coming out of that test rig.
Any estimates on how much more efficient an engine would be with those lighter blades?
The weight won't make any difference. The ultimate goal is a more efficient airfoil shape. These are just the first step in a long process.
Dang... I hadn't considered the effect of a blade breaking and FOD'ing out the engine if you are standing down wind ("down stream"?). Does the shrapnel go directly to the rear, or could a blade fragment catch the air and deflect sideways? I think I'd stick to the control room and not tempt fate. 😄
@SkyhawkSteve This is a nitpick from this old pedant: if nothing 'foreign' (ie, from outside the engine) has caused the failure, then it can't be FOD (foreign object damage).
It's not a commonly used term, but DOD was used in my day.
I'll leave you to guess what the 'D' stands for.
@@grahamj9101 fair enough. Maybe "entropy based damage"? "Internal Object Damage"? I'm still curious about how much stuff comes flying out the back of the engine, though!
Graham is correct, as usual. Since these are first stage compressor blades, and they are more likely to break up than steel blades, I think it would be very likely for debris to come out the front of the engine.
Any damage to the later stages of the compressor would likely cause a stall, helping to blow any composite or metal debris forward.
It may not be too noticeable in the videos, but everybody was more careful than usual to not be in front of or behind the engine when it was running.
@grahamj9101 FOD is also commonly used as in Foreign Object or Debris Damage as a catch all. I agree with you though as I had the terms FOD and DOD through the RAF and it causes the brain to read it as Foreign Object Damage Damage WHATNOW!
Agent Z,
You are a “legend” in our A&P students’ community!
Reference this experimental path, my vote is for:
a. Populating the compressor with FULL (21 I guess) set of new blades (to get the relatively high "n" required for statistical analysis) at 125% of maximum operational RPM; for 1500 hours (slightly over two months 24/7) using an electrically driven motor.
b. Stop experiment, remove and completely inspect 3 blade only at a time (7 blade apart) very thoroughly by all NDT techniques available and note deficiencies, then inspect another set of three blade the same way, and proceed until all 21 blades have been inspected (to get good data for statistical analysis on how to, and how many blades to inspect and how often to do it).
c. If still serviceable, repopulate the compressor shafts with the FULL set of blades again; with the test compressor stage enclosed in vacuum sealed chamber and spun to produce 200% of design centrifugal load for 6 hours.
d. Stop experiment and retest blades as in "b" above.
e. If still serviceable, start again as set up for "c" above, run at 200% of design centrifugal load for one more hour; then without stopping the experiment, gradually reduce the vacuum (to reintroduce aerodynamic load) at steps of doubling such aerodynamic load in each stage of vacuum reduction. and remain at such load for one hour, until arriving at ambient pressure level.
It will take some resources and cost to do it; but in the long run will arrive at practical implementation and certification quicker on the overall project path. In addition, you may be jumping ahead in development of test rigs and protocols for servicing and inspecting these blades once accepted into routine operation service.
Al
You are putting a lot of faith in the disc. I would not want to be in the same building with that thing at 15,000 rpm.
These blades will be tested for wear at 100% design speed for up to a year of running. In service, the engines never overspeed.
@@AgentJayZ We LOVE your videos and our faculty show them in class.
I think they need to send me a hat, eh?
@@AgentJayZ Of course! [Estimated arrival Jul 11th - Jul 18th]
Al
@@AgentJayZ One item should arrive by the 11th, but the remaining items are expected to arrive by the 18th.
i know that maybe its not applicable in this case, but ice/water ingression could be a problem with this carbon fiber blades?
Not with industrial engines. This is the first step. Have patience, grasshopper.
Sweet video and great responses to some of these comments 😂
Full Power for a HOUR on 3 test no problem is GREAT NEWS for me and everyone on the EARTH... i just hope the longevity will be FLAWLESS...
This is how you handle "HOA Karens"..
i thought they already used composite blades in a gas turbine engine? GE90?
Fan blades only with inlaid titanium sheet leading and training edge reinforcement. All the gas path blades are fully metallic alloys the vary for location (based on temperature, size and speed).
But do they pass the toughness standard?
Pass the Bird strike test?
Not applicable to industrial engines.
I thought these tests might be for aviation application
Gotta watch and listen to what the cranky old host says...
If you added a stator you would introduce minute pressure differences as the blades came together and passed. So (rough math) 24 blade fan at 6000 rpm=144,000 pressure changes per minute. MotoGP has banned the use of CF rims on race bikes because the of the unpredictable catastrophic failures that can occur on the wheels. There is a wheel, or there is shrapnel, there is no in-between. I am sure engineering can figure this out, but get lots of testing done.
As I've suggested elsewhere, a rig test of a bladed disc would need a front frame and a VIGV row in front of it, and a stage 1 VSV behind it, so as to present a representative flow field to the blades. I'm going to leave the guys at S&S Turbines to decide how to do this.
Cool ill be back often again
That is both the beauty and the curse of carbon composites. While incredibly strong--you'd need a micrometer to check for any elongation--I tend to liken CC parts to bridge girders. They will do their job, and do it well, for dozens of years without fail. However as they age (mostly environmental impact, notably oxidation, heat fatigue and weather exposure) they can fail in the most strikingly catastrophic way without warning. In that moment, they lose a ton of energy though and the parts don't travel quite as freakishly shrapnel-esque like metallic components. The one saving grace so long as you're fifteen feet away for so.
None of that is a knock on this project, rather a generalization of CC parts in hard use universal to the material. I wish I'd had high-speed camera footage during vibrational testing a few years back when I got a set of parts absolutely singing at 65K rpm and found the upper threshold of my bearings. Pow!
They will not be put in service until a "life" is determined. If they go at 4 years, the replacement will take place at two.... Easy.
Private owners can do what they like.
How many cars do you see every day with bald tires on the freeway?
Hey J its been a while
what is the propane burn rate for one of these engines
Depends upon the power setting. At full power, the 1500 core will burn the equivalent of about 35 gallons of jet fuel or diesel per minute. We don't actually measure fuel consumption on gaseous fueled engines, because they are used on natural gas pipeline compressor stations, where their fuel is taken directly from the pipeline.
I know our lms100’s are burning though 55,000-56,000 pph natural gas at work.
@@JAMESWUERTELE What is pph?
Pounds per hour. It's a traditional measure of fuel flow. Gotta check some of those books I recommend.
@@AgentJayZ Thanks! At first I was thinking that but thought it couldn't be as pounds are usually abbreviated as lb and In my country the gas that goes to power plants is usually measured in m2/h. Will check those books.
What's the purpose of the CF fan blade on a ground running engine? Obv. weight in a flyer.
Cheaper to manufacture. Non OEMs can go mad with old engines as the operators are not contractually bound to use manufacturer's parts and overhaul facilities. Plus industrial engines are not bound by the same compliance regs as aero so it's massively easier to get alternate parts into them rather than genuine if they are still manufactured.
Watch the previous videos. In short, they're aiming at 1% or so fuel savings. 30MW (or so) thermal is about a kilogram of fuel a second, ten grams a second saved. For an industrial engine, that's 300 tons of fuel saved a year of continuous operation, which is a largeish slice of a million dollars.
Stronger (in specific ways) combined with lower mass lets them use different profiles, resulting in greater efficiency.
The hope is to be able to increase performance or economy by introducing different airfoil shapes... not necessarily for 1500s, but for other engines. I tried to explain all that, but I'm kinda used to people not listening to the details...
Many a bug died that day
Quite a few.
Hot stuff!❤️🔥 Who might take interest in this hmmmmm, GE maybe 🤔 Lockheed Martin maybe 🤔
S&S Turbines Maybe 🤔 and maybe PierCarlo and friends @ “ Starfighter’s Space “ as old Skool TV “ Laugh In”
Would say “Very Interesting “🧐 OOPS I might have let the J79 out of the bag 💼 er cat 🐱.
🤗🤗🤗🤗🤗💯💯💯💯💯
Here are my further thoughts, and I continue to maintain that the only truly representative test will be a field trials installation. However, I also maintain that a lot more testing than a few hours engine running in the test cell with even a full set of blades will be necessary before an operator agrees to a field trials exercise.
So what can you do that's near 'representative', without tying up the test cell? I'm sorry, but just spinning a bladed stage 1 disc up to engine speed will not be 'representative'. The very fact that 50 to 200 HP might be sufficient to do the job should tell you that it's not.
The composite blades need to be absorbing the power they would in the engine, which means they need to be subject to an equivalent of the engine installation, but without the rest of the engine.
They need a front frame with its eight struts and a row of VSVs in front of them, together with a row of stage 1 vanes behind them. I imagine you and S&S Turbines have enough bits to build such an engine parts rig. However, it would probably need more than a couple of hundred HP to drive it, which starts to sound expensive, but far cheaper than engine running - and it frees up the test cell.
Great to see you here adding your many years of jet engine design experience to the discussion. I would find it very educational to have some input on what types of instrumentation are used on blade development. The test rig that MAN uses for compressor development (youtube video of surge testing) shows images of a tip timing system. I assume strain gauges on rotating parts are possible (though I don’t know how this is done).
I am relieved that my “intuition” that a depopulated stage would not be representative of the minimal requirements for fatigue testing. But then again, I don’t have any experience with these issues.
@Mentaculus42 I've been in retirement for 20 years now, so my knowledge of instrumentation is dated. In my early days, the signals from straingauged blades were taken via slip rings. In the later stages of my career, the use of a contactless radio system had come in. On larger compressor blades, the measurement of tip deflections was also used, with a grid embedded in the rotor path above the blade row.
However, I do have contact with more recently retired colleagues who specialised in instrumentation design. I can ask them.
@@grahamj9101
Many thanks for responding. I was thinking along those lines. That is everything I wanted to know.
@grahamj9101 current rotating s/g telemetry set ups are still the same as the radio/antennae methods used 20 years ago. Certainly atm there are new telemetry units in design or moving to F status. Do you recognise DIGBERT and ALBERT at all?
@@paulpower3766
Could you expand on your comment, I find it rather cryptic with the acronyms that don’t seem to have google search results. If you know something about the topic then I am interested in learning but I couldn’t seem to understand what you were saying. grahamj has always provided very educational information that always makes sense in the context of my own engineering background.
realizing we're still a way from aviation...are you allowed to comment on grain structure and how they go about 'forging' the blades? As shown narrow neck, potential different quick airfoil just seems 'wow'!
I'm just a guest commenter. I have met these guys. They are materials engineers, and have previous experience making internal engine parts for top fuel cars. Yeah, so let's all go a bit easier on our armchair worries.
Coming from a performance boat background, I'll give an opinion, inaccurate as it may be. With that said, my money is on the carbon blades being vacuum infused construction, with cure time controlled via vaiable heatl.
11:18 If it is a 1000 hp fan, you won't test it at design speed with a 100 hp motor. You'll need 1200 hp or thereabouts. 1st law of thermodynamics.
It will only have 3 or 4 blades in it.
First law of paying attention.
@AgentJayZ As I've been telling you, running just three or four blades will tell you relatively little. You really need a full set with a front frame, VIGVs, and stage 1 VSVs to do some meaningful testing
The original idea was to test root durability. Your thoughts have merit, and the plan may change. Nothing has been built yet.
They need to do some testing to see what wear effects repeated starting and stopping does to the root. I would imagine the repeated loading and unloading of centripetal forces would be a lot harder on the resin than it would on steel.
Gee, if only somebody mentioned that in this series of three videos about this devlopmental process.
I really have no answer here.
A fiber plate will be as strong as metal. Until it gets a nic or microscopic scratch. Then it will shatter. Will also have terrible erosion qualities in sand or salty air
Industrial engines breathe filtered air. Problems can occur if the filtration system is faulty.
They will have to make sure they keep sand and salty air out of the power/gen room the engine will be running in.
Zorbak: So, completely innocent of any knowledge... even what I literally handed to you right there.... amazing.
CF blades are in commercial use in the GE90 already. They do have metal leading edges though.
@@Sonny_McMacsson
Is that the fan blades and / or compressor blades also?
can you please pin a comment with a link to your explanation on the benefits of the new carbon fiber blades over the old metal ones?
Watched all three vids?
These are the first few steps of a long process. These are not the finished product at all.
I watched them and I remember an explanation on the economical benefits but I don't remember in which one and in what minute. might be useful for all viewers to highlight the information in the descriptions. thanks.
🫂
I still think that these Carbon-fiber blades should be electroplated.. Some type of chrome-nickel (let the engineers figure that out).. 'cause I know they want these blades airborne.. well just for the heavy dust..
So who exactly wants them airborne? There must be just a handful of J79 powered aicraft flying, which will do just a few tens of hours a year at air shows. Surely the market that Robin Sipe and S&S Turbines are targeting is the industrial gas generator repair and overhaul market? Otherwise why bother?
To run a 1000hp turbine via a gearbox at max rpm, I suspect a 1000+hp electric motor would be needed. Unless you want to run it in a vacuum, unloaded.
Very common error. The function of a turbine is the exact opposite of a compressor.
One only needs to provide the power to turn the one stage (assuming it’s just that one disk mount in a containment rig of some kind). At the main OEMs, the containment test stands use large electric motors to turn single disks for testing.
Now that you're caught up, the bets are how powerful a motor we need...
@@AgentJayZ I get it now, the hp rating says what the turbine's power output can be. Spinning it up to it's operational rpm requires much less power.
NotA: You get nothing. A turbine is not a compressor, They are in fact exactly opposite in function, One is the bat, and the other is the glove. The air is the ball.