|
|
Post by racket on Oct 6, 2017 19:09:15 GMT -5
Hi Matt Because you need to run a bit cooler than the Gt6041 , your division of hole areas will need to be altered a tad , you could use the same Primary and Secondary as the 6041 to burn the "same" fuel load, but just increase the Tertiary hole area to cool things a bit more . According to the drawing the 6 vapourisers have their bore centre on a 23mm radius for the inlet and a 65 mm radius centre for the outlets , in a 180 mm ID flametube , the centre of the vapouriser outlet is 25 mm from the wall, the length of the various legs appear to be different to what I told you earlier , the drawing gives 100, 65, 60 mm . If we look at the FT pic jetandturbineowners.proboards.com/attachment/download/178Top row of holes has 2 X 7mm dia holes, 15 mm centre distances, adjacent to each vapouriser outlet , 13 mm down the side wall , the 12 X7 mm holes have 42 X 2mm holes spaced between on that 13mm line , the second row of 54 X 2 mm holes (staggered between the first row holes) are 37 mm down the wall and are just for wall cooling . The third row of holes are still Primary holes , there are 6 X 9mm positioned between vaporisers and have been "bent" upwards to produce a jet of air going towards the FT cap to promote recirculation between the vapourisers , the 12 X 5mm holes , one each side of the 9mm holes need to be ositioned so as not to discharge ono the vapouriser . The fourth row is wall cooling so 54 X 2mm dia at 82 mm down the wall . The fifth row at 110mm down the wall are Secondary holes and discharge below the vapouriser radial leg , and consist of a mixture of hole sizes to produce jets of varying depth penetration (hopefully),some larger holes have been bellmouthed to aid airjet flow , there are 3 X 14 mm positioned directly under every second vapouriser leg, and 3 X 10 mm under every second leg ,in between these there are 12 X 8mm and 18 X 6mm dia . It would appear that the parallel side wall is only 143 mm long , 588 mm long with a 10 mm overlap for spot welding . I had 18 X 16.5 mm bellmouthed Tertiary holes and 25 X 4 mm at the bottom for wall cooling , you would probably go for 17.5 mm holes to provide a tad more cooling and flow potential Hope this helps Cheers John |
|
CH3NO2
Senior Member
Joined: March 2017
Posts: 455 
|
Post by CH3NO2 on Oct 6, 2017 21:42:08 GMT -5
Hi John,
There is something I have been wondering about on the hole inlet coefficient. You wrote "bellmouthed to aid airjet flow."
Bellmouthing increases the inlet coefficient but does it change anything else?
Say if for example a sharp edge hole has an inlet coefficent of 0.6 and a bell mouth has an inlet coefficent of 0.9.
If the sharp edge orifice diameter is sized 0.3 larger than the diameter of the bellmouth (assuming they both end up with the same mass flow rate) is there any difference as to what happens inside of the flame can?
Thanks,
Tony
|
|
|
|
Post by racket on Oct 6, 2017 23:33:00 GMT -5
Hi Tony
Yep , theres big differences between bellmouthed and plain holes , but to be quite honest I don't think I've read anything on comparing a large plain vs a smaller bellmouthed , could be wrong there.
If one is using 1.2mm stainless sheeting it gets a bit hard to bellmouth small holes , so I only do it on the "bigger" holes, generally I allow a bit more percentage of total hole area to the Primary and Secondary than is strictly necessary , but considering most of the holes in those zones are plain holes its probably warranted .
Flametubes are going to be full of compromises and unknowns , I guess the best we can hope for is that it burns the fuel fairly efficiently and in the right place , an "original" Member of the Yahoo DIY Gas Turbines Group , Mike Early , an ex Garrett Aero engineer , made >50 flametubes before he was happy with the one he used on his engine , most of us have too many other things to consider with our builds like fitting them with a freepower to mount in a vehicle to spend a lot of time developing a flametube to perfection, its not high on our priority list , as long as it "works", it'll do , the guys over on the GTBA who make engines for their RC aircraft are more particular about efficient burn rates for obvious reasons, the rest of us can make the flametube a tad more generously proportioned which compensates for the less sophisticated design .
Cheers
John
|
|
CH3NO2
Senior Member
Joined: March 2017
Posts: 455
|
Post by CH3NO2 on Oct 7, 2017 23:03:02 GMT -5
...... According to the drawing the 6 vapourisers have their bore centre on a 23mm radius for the inlet and a 65 mm radius centre for the outlets , in a 180 mm ID flametube , the centre of the vapouriser outlet is 25 mm from the wall, the length of the various legs appear to be different to what I told you earlier , the drawing gives 100, 65, 60 mm . If we look at the FT pic jetandturbineowners.proboards.com/attachment/download/178Top row of holes has 2 X 7mm dia holes, 15 mm centre distances, adjacent to each vapouriser outlet , 13 mm down the side wall , the 12 X7 mm holes have 42 X 2mm holes spaced between on that 13mm line , the second row of 54 X 2 mm holes (staggered between the first row holes) are 37 mm down the wall and are just for wall cooling . The third row of holes are still Primary holes , there are 6 X 9mm positioned between vaporisers and have been "bent" upwards to produce a jet of air going towards the FT cap to promote recirculation between the vapourisers , the 12 X 5mm holes , one each side of the 9mm holes need to be ositioned so as not to discharge ono the vapouriser . The fourth row is wall cooling so 54 X 2mm dia at 82 mm down the wall . The fifth row at 110mm down the wall are Secondary holes and discharge below the vapouriser radial leg , and consist of a mixture of hole sizes to produce jets of varying depth penetration (hopefully),some larger holes have been bellmouthed to aid airjet flow , there are 3 X 14 mm positioned directly under every second vapouriser leg, and 3 X 10 mm under every second leg ,in between these there are 12 X 8mm and 18 X 6mm dia . It would appear that the parallel side wall is only 143 mm long , 588 mm long with a 10 mm overlap for spot welding . I had 18 X 16.5 mm bellmouthed Tertiary holes and 25 X 4 mm at the bottom for wall cooling , you would probably go for 17.5 mm holes to provide a tad more cooling and flow potential Hope this helps Cheers John Hi John, Thanks for posting your flame can details. Since I will be building a system using the GT6041BL the best thing I can probably do, as a beginner, is to try to shamelessly duplicate your flame can design and performance results. Since part of what I want to do is inject rocket propellants for fuel, I could baseline the system performance on kerosene using your flame can design. If my baseline testing closely matches up with your test results, that would be ideal. At least I will be starting with a known/working design that experiments can be extrapolated from. I'll still do the sore thumb can with swirler. That's mandatory too.  Thank you, Tony PS - I'll make the vaporizers from inconel. |
|
|
|
Post by racket on Oct 8, 2017 0:28:11 GMT -5
|
|
CH3NO2
Senior Member
Joined: March 2017
Posts: 455
|
Post by CH3NO2 on Oct 8, 2017 10:29:59 GMT -5
Hi John, I have a few questions about your flame can. In this picture here I notice the various different parts are put together using lap joints and and the joints are only sparsely tack welded. Is there a reason to not use fully sealed and welded butt joints? Another question is about the boundary layer coolant holes just before the turbine inlet. What is the purpose of additional cooling there? Is it to reduce thermal gradients at the weld next to the flange joint? postimg.org/image/swe3ec3fp/ |
|
CH3NO2
Senior Member
Joined: March 2017
Posts: 455
|
Post by CH3NO2 on Oct 8, 2017 12:06:14 GMT -5
Could you give some clairification of the fuel injector bird cage section? Can't see what is going on inside of it. What is the 1/4" tubing going into the top right of the lid here? Thank you, Tony postimg.org/image/o6qc9e1ln/ |
|
CH3NO2
Senior Member
Joined: March 2017
Posts: 455
|
Post by CH3NO2 on Oct 8, 2017 13:01:56 GMT -5
|
|
|
|
Post by racket on Oct 8, 2017 15:31:08 GMT -5
Hi Tony
The FT parts are only tack welded mainly because I didn't want to contend with buckling from a full weld , also tack welds can be easily ground out if I needed o change parts of the FT , because both the top cap and tapered tertiary pieces have their "laps" bent over a steel die , the straight wall section is seated firmly and being on the "inside" will expand outwards into the lap for a decent seal, its irrelevant if theres a few leaks here and there , the whole FT is full of holes, also having a lap means the weld is easier to do than a butt weld where the chances of blowing a big hole are always prevalent...........unless you're a very competent welder like Anders ;-)
Those small Tertiary holes at the outlet of the FT are to keep the slip joint cool so that it doesn't bind in the joint as things start to heat up , the stainless FT will expand a lot whereas the mild steel can won't .
The 1/4" tube in the top of the FT is Smithies hot streak .
Inside the "round house" there are a couple of manifolds , one for kero and one for propane preheat, along with their syringe injectors , the roundhouse is simply there to allow air to get into the ends of the vapourisers , hence all the holes in its side wall , it also has a threaded tube( 1/2" BSP thread ) at the end which carries the kero and propane lines, as well a nut on the exterior of the outer can locks the FT in place against the inside of the can dome, the heads of those cap screws hit the inside of the cap, the FT is then allowed to expand axially towards the turb scroll .
Thats a nice looking tool :-)
Cheers
John
|
|
|
|
Post by finiteparts on Oct 8, 2017 15:39:11 GMT -5
Hi Matt, John and Tony,
I wanted to throw my two cents in on a few comments posted here.
John, I am not sure what you are suggesting to Matt is actually going to help. The fuel flow requirement is set by the system demand for input energy and how well we can convert the fuel into heat. The combustor flow effective area is sized relative to the pressure drop that we are willing to take across the liner, the mass flow and the casing flow conditions.
Changing the dilution hole size (and thus increasing the overall liner flow area) will reduce the liner pressure drop and that may potentially make a small impact to the system overall energy demand (due to the reduced pressure loss), ie, a slight temperature reduction at the NGV plane, but you might also impact combustor stability. Again, the require temperature is set by the system energy demand, so there is only so much that you can reduce the turbine inlet temperature anyway.
The main point that I want to be clear here is that you cannot just drill extra holes in the liner and somehow the turbine inlet temperatures will go down. I am not sure that is what you were suggesting, but I was concerned that the way your comments read may have led some readers to think this.
If you have efficient combustion, the turbine inlet temperature is primarily a function of the overall air and fuel flow, i.e. the fuel/air ratio (FAR). All you are doing when you place the holes in the liner is controlling where the reaction takes place. Now if you don't have efficient combustion, then you probably have all kinds of hot streaks and other malformed flows that will wreck the hardware anyway, and larger dilution holes won't help that either. The primary zone airflow is critical, the secondary is important and the dilution zone is basically what is left over. This cascade of importance is recognized because you are sizing your combustor based on the time to complete the reaction and if the reaction starts too late, it will not be completed in the combustor as desired.
Tony,
The primary reason for using a sharp edge hole and the "bellmouth" or chutes is based on the control that you need on where the dilution jet goes. A sharp edge hole works great for the primary zone and secondary zone because the jet is injecting into a relatively slow flow, or put in another way, is injecting into a low momentum stream. So the jets momentum will be high enough that it can penetrate down into the flow. The thin sheet metal with a hole drilled in to does not offer any flow control of that jet...the momentum of the stream that the jet is injecting into will effect the jet right as it enters. Additionally, the jet momentum will be reduced because of the flow losses entering and exiting the hole...some of the flow energy is consumed making the turns in and out of the hole.
So now we look at the chuted hole. It exhibits much more flow control on the injecting jet and as such can push the flow deeper in to the core flow. Obviously this is a more desired feature on the tertiary dilution holes because there is much more bulk flow in the liner at the aft section and as such, there is much more flow momentum. The reduced entrance loss also helps to convert the pressure energy more efficiently to flow, using the assigned combustor pressure drop more effectively.
Finally, I am curious why you would want to inject rocket fuels into a gas turbine? Because most rocket fuels bring in their own oxygen, they are usually heavier than kerosene for the same amount of energy...and we also have plenty of free oxygen to use that we don't have to transport along with us. The USAF has done tons of advanced turbine fuel studies since the 1950s and guess what, they are still using kerosene based fuels.
Matt,
The use of the 90 degree elbow does bring in some strong secondary flows that will stir things up as the flow goes through the turn, which is not as much of a deal on standard turbine scroll housings since they are essentially long turning passages themselves. But because your housing is split in a different way, you need to make sure that everything is mixed well before entering your housing. You might want to make sure that you have a longer than normal tertiary zone.
I am not going to sugar coat it, that turbine housing looks awful. The flows will be coming from different directions and so the entrance angle to the nozzle guide vanes will be different relative to the which side they are coming up. One will be clockwise and the other will be counterclockwise relative to the NGVs. There is probably a reason that you don't see that style housing much. The NGV nearest tot he entrance will see a relatively straight on flow and the one completely opposite the entrance looks like there might be a dead spot.
It would be interesting to see the NGV and if they have accounted for that in the design.
Good luck!
Chris
|
|
|
|
Post by racket on Oct 8, 2017 17:50:43 GMT -5
Hi Chris
My reply to Matt has to be taken in context with previous correspondence , Matt has a greater mass flow than the GT6041, 3 lbs/sec vs 2.75 lbs/sec , hence it would normally require a greater flametube wall hole area to maintain the same pressure drop , but because he is required to operate at lower turb temps he needs more dilution air , so my solution was simply to use the same Primary and Secondary as the GT6041 , a proven design , to burn the same amount of fuel as the GT6041 in the same volume at the same pressure drop across the FT wall , the "extra" left over mass flow is simply dumped through larger Tertiary holes to lower the combustion gas temp to the lower design limit .
I'm no combustor engineer , heh heh, or any other type of engineer for that matter,......... so guys can take my recommendations, or not , LOL.......I'd be only too happy to hand over combustor designing queries to anyone else who'd like the job,...... hint hint ;-) .
Cheers
John
|
|
|
|
Post by finiteparts on Oct 8, 2017 22:13:16 GMT -5
Hi John,
I didn't think that was what you were suggesting, but I also wanted to make sure it was clear to other readers.
One thing though...a 9% increase in mass flow can't be sustained on the same amount of fuel, well, depending on the system PR. Were you assuming that his engine would run to a lower design point? I would have just assumed scaling the combustor larger by 9% would be a good plan.
Don't diminish your expertise...you have experience and lots of knowledge. You and I talking this stuff out helps everyone else reading this to get answers to questions that maybe they didn't even know to ask.
I was hoping to provide the "tools" to help readers to design their own combustors, but free time is a bit scarce, so I will try to get back to that post as soon as I can.
Regards,
Chris
|
|
CH3NO2
Senior Member
Joined: March 2017
Posts: 455
|
Post by CH3NO2 on Oct 8, 2017 22:48:03 GMT -5
...... Tony, The primary reason for using a sharp edge hole and the "bellmouth" or chutes is based on the control that you need on where the dilution jet goes. A sharp edge hole works great for the primary zone and secondary zone because the jet is injecting into a relatively slow flow, or put in another way, is injecting into a low momentum stream. So the jets momentum will be high enough that it can penetrate down into the flow. The thin sheet metal with a hole drilled in to does not offer any flow control of that jet...the momentum of the stream that the jet is injecting into will effect the jet right as it enters. Additionally, the jet momentum will be reduced because of the flow losses entering and exiting the hole...some of the flow energy is consumed making the turns in and out of the hole. So now we look at the chuted hole. It exhibits much more flow control on the injecting jet and as such can push the flow deeper in to the core flow. Obviously this is a more desired feature on the tertiary dilution holes because there is much more bulk flow in the liner at the aft section and as such, there is much more flow momentum. The reduced entrance loss also helps to convert the pressure energy more efficiently to flow, using the assigned combustor pressure drop more effectively. Finally, I am curious why you would want to inject rocket fuels into a gas turbine? Because most rocket fuels bring in their own oxygen, they are usually heavier than kerosene for the same amount of energy...and we also have plenty of free oxygen to use that we don't have to transport along with us. The USAF has done tons of advanced turbine fuel studies since the 1950s and guess what, they are still using kerosene based fuels. ...... Good luck! Chris Hi Chris, Thanks for the feedback. Your input is always educational. On the orifice inlet, sharp edge Vs radiused inlet. Assuming equal mass flow, and I assume velocity, they should have the same momentum right? Do the different shape inlets create different velocity profiles or Reynolds numbers at their outlet? And yes, I am fully aware fuel chemistry has been experimented with in turbines. And yes, I agree Kerosene is the best fuel for 99.999% of applications. However, I am targeting a unique application. I am building my turbine for the purpose of maximizing air bleed. I was able to locate a GT6041BL turbo just like the one John used. Maybe its "slightly" different but it's close enough. The purpose or intent of using a rocket propellant(s) is so the combustion process wont require as much air to operate and power the turbine. The more mass flow and chemical energy that can be injected into the flame can and sucessfully reacted, will displace air that would otherwise be needed by the turbine. The more air that is displaced by the monopropellant is air that can be bleed from the system. Exactly how much monopropellant can be injected into the system and still keep everything working as intended.... I'll find out. I expect the propellant will be consumed very quickly but that's OK. I just want to see the numbers and potential. It's what I hope to try anyways.  I'll give it a shot and post results. Tony |
|
CH3NO2
Senior Member
Joined: March 2017
Posts: 455
|
Post by CH3NO2 on Oct 8, 2017 23:00:25 GMT -5
PS - Having pure air for bleed is one of the test conditions I want to try. In others, I'll be able to inject water before the compressor.
I'll test it every way I can imagine to see what I can get it to do. I have a LOT of chemistry experiments to test in this turbine.
|
|
|
|
Post by racket on Oct 9, 2017 3:21:27 GMT -5
Hi Chris
My thoughts were that Matts turbo has a lotta unknowns attached to it with regards component efficiencies , I feel his comp has a "better??" design than the Gt6041 and if the turb stage with its NGV picks up a few more percentage points over the 6041's 78% effic with the scroll, then the reduction in turb temps won't affect outcomes as badly as expected, he could end up producing a decent pressure ratio from those lowish temps .
Just aiming for a ballpark design at this stage , once Matt can measure up the NGV throat area we'll be in a better position to think about tweaking the design , a choked NGV will probably be a must have to control comp flow as the turb wheel exducer is far too "open" to help .
Yep , gotta throw things back and forth to shake out a decent design :-)
Cheers
John
|
|
