|
Post by Johansson on Jan 27, 2024 6:51:58 GMT -5
Yup progress is being made swiftly so far, there are some parts that I haven´t figured out how to make yet like the engine cover seal but it will work itself out once I am there.
During the testing I won´t slam the throttle up to 10.000rpm and hope for the best, but take everything in increments after a set test plan.
I am deep into uncharted waters with this engine so I will do this very scientific with a 2 minute run at every 1000rpm with every temp and pressure monitored, and a check for wear and tear after. When every kink is sorted out I will make 5 minute endurance runs at 8000, 9000 and 10.000rpm.
|
|
|
Post by andym on Jan 27, 2024 7:09:43 GMT -5
Yup progress is being made swiftly so far, there are some parts that I haven´t figured out how to make yet like the engine cover seal but it will work itself out once I am there. During the testing I won´t slam the throttle up to 10.000rpm and hope for the best, but take everything in increments after a set test plan. I am deep into uncharted waters with this engine so I will do this very scientific with a 2 minute run at every 1000rpm with every temp and pressure monitored, and a check for wear and tear after. When every kink is sorted out I will make 5 minute endurance runs at 8000, 9000 and 10.000rpm. Sounds like a plan lots of video please
|
|
richardm
Senior Member
Joined: June 2022
Posts: 338
|
Post by richardm on Jan 27, 2024 9:25:23 GMT -5
|
|
|
Post by Johansson on Jan 28, 2024 10:57:44 GMT -5
One blade section weighs 453g and will be held by 5 M6 bolts in 12.9 quality.
If it turns out that the centrifugal forces is too high for the M6 bolts I will use these blades up to their safe RPM and then make a solid blade ring later that holds up for 10.000rpm. It might be possible to make them from thinner 2.5mm or even 2mm sheet to save weight.
|
|
richardm
Senior Member
Joined: June 2022
Posts: 338
|
Post by richardm on Jan 28, 2024 13:20:12 GMT -5
Did you try the calculator ?
|
|
|
Post by finiteparts on Jan 28, 2024 14:13:44 GMT -5
So I ran a few quick calcs. At 10krpms, the blades will be exerting a pull load of nearly 87 kN (19,543 lbf) each! Spread that across the 5 bolts, that are in double shear, and that puts a shear stress through the stainless steel of 210 ksi. I guessed the radius to the mass center of the blade segment was 175 mm...you can let me know if that seems reasonable. The blade segment is subjected to approximately 20,000 g's when it is rotating at 10krpm. You can replicate this in the online calculator provided above... If I assume that they are 410 stainless, the ultimate yield stress (at room temperature, not even hot) is roughly 130 ksi and using distortion-energy theory's value of 0.577*Sy to calculate a reasonable ultimate shear strength, that puts it around 75 ksi for the failure point in shear. So, unfortunately your part is loaded 2.8 times larger than what we would expect for failure. I went to McMaster Carr and got a weight for a M5-0.8 x 12 bolt (~ 29.5 g (.065 lbm)). Each of the bolts produce a dead-load on the disk of 4849 N (1090 lbf). So for each segment, the rotor disk is being pulled by ~ 111 kN (~25000 lbf). If I am seeing it correctly, you have eight of these segments...so that is a total disk dead-load of 889 kN (200,000 lbf) trying to tear that metal disk to pieces. I hate to be the bearer of bad news, but the numbers seem a bit upsidedown on this. I will try to think of what you could do to fix this. In the mean time, can you give me the dimensions of the main disk, hub, thickness, etc... from that we can get a sense of where you have to be. - Chris
|
|
|
Post by Johansson on Jan 28, 2024 14:37:36 GMT -5
I´ve been busy this weekend Richard so I didn´t have time to try it out, thanks for the link though! Thanks a lot for the help Chris, I´d rather get the bad news like this than IRL when the turbine disc disinterates on me... I will check the measurements and get back to you, meanwhile what is the max safe rpm with these blades? If I can safely run them a bit past idle I will use them for the initial tests and change them for a more suitable set of blades later. What if you run the calcs with 2mm thick blades? 302g per segment. Now when I think about it I really need to know before making the hot end if it is at all possible to run a 500mm dia turbine wheel, or if the centrifugal forces get too damn high and I need to make a smaller diameter wheel.
|
|
|
Post by finiteparts on Jan 28, 2024 14:42:08 GMT -5
Oh yeah...also, if you can, what it the material of the turbine hub and primary disk?
|
|
|
Post by Johansson on Jan 28, 2024 14:45:46 GMT -5
Oh yeah...also, if you can, what it the material of the turbine hub and primary disk? They are S355 steel.
|
|
|
Post by Johansson on Jan 28, 2024 14:51:12 GMT -5
The turbine disc is 290mm in diameter and 6.5mm thick. The hub is 110mm in diameter and 30mm thick with a bore of 80mm where the clamping element sits.
|
|
|
Post by finiteparts on Jan 28, 2024 14:58:23 GMT -5
On a bit of good news, the lower rotor speeds keeps the kinetic energy down to a lower level, if you were to throw a blade segment off. Your current blade segment would only have 10% of the kinetic energy of a .357 Magnum (250 grain) round. But my guess is that it would not fail spectacularly, it would yield out and rub the case before it burst....like the old Schreckling engines use to do with the lower capability metal wheels.
But, it is still a large force. Equivalent to getting hit by a 1.065 kg mass going 109 km/hr (2.347 lbm at 68 mph) or a 92.5 kg mass at 16 km/hr (204 lbm at 10 mph). Doesn't sound too fun.
|
|
|
Post by finiteparts on Jan 28, 2024 14:59:39 GMT -5
Cool....I will give it a go on those dimensions and see where an allowable condition would be.
|
|
|
Post by Johansson on Jan 28, 2024 15:09:30 GMT -5
That is good to know, a contained failure is preferable. I think that the way forward is to figure out the limits for this engine, with the "least bad" rotor construction. I have no idea what kind of revs the sheet metal compressor is capable of and it is difficult to calculate with the welded blades and stuff. My goal with this engine project is to make a 100% sheet metal DIY gas turbine that can produce some decent thrust, if I have to downrate the top RPM so be it. Even at 6-7000rpm it will still send a small hurricane over to my neighbours yard.
|
|
richardm
Senior Member
Joined: June 2022
Posts: 338
|
Post by richardm on Jan 28, 2024 15:21:30 GMT -5
May I suggest that you run have the same calculations don for the compressor wheel. Also I would build a containment ring made of thick enough steel around those rotating parts for safety. I ve seen GTCP series APU with " bullet holes " trough their turbine housing resulting from turbine blades disintegration. Must have been hell around those things when it happened ...
|
|
|
Post by finiteparts on Jan 28, 2024 15:44:15 GMT -5
From the material properties that I can find for S355 steel, you current disk is likely not even able to carry it's own weight at 10krpm....and that is using room temperature 0.2% yield stress. So what this means is that you probably can't get away with the flat disk. If you look at most turbine disks, they are wider at the bore. This is because the material at the bore has to "carry" the weight of all the material outboard of it. So what would probably need to be done is that the turbine disk will need to be machined to have a profile. Now, because you do have a very thick bore piece, maybe if can carry the load.
I will try to do a quick FEA to check the load carrying ability of the hub portion. Is the disk centered over the wider bore piece or is it pushed toward one end?
|
|