Sweetenough
Veteran Member
Joined: April 2016
Posts: 121
|
Post by Sweetenough on Mar 28, 2019 2:32:26 GMT -5
Thanks John! It is really strange, the fuel control for the EBK is kept constant. The EBK fuel pressure is fairly constant the red line in the bottom of the chart The slight increase in EBK fuel pressure during the run I assume comes from that the engine is adding more fuel to keep the rpm set point. 2,1 to 2,3 bar Even without EBK it behaves strange See run below, at higher rpm almost all parameters is increasing with constant speed. The back pressure in the EBK can bee seen as "EBK fuel pressure" with no EBK fuel the pressure transmitter meassure the back pressure from the EBK fuel nozzles.
|
|
|
Post by racket on Mar 28, 2019 3:23:48 GMT -5
Ah , electronics .............not my thing :-(
|
|
|
Post by smithy1 on Mar 28, 2019 19:10:49 GMT -5
Ah , electronics .............not my thing :-( Likewise.....I much prefer manual/mechanical things I can physically change and adjust myself. Smithy .
|
|
|
Post by finiteparts on Mar 28, 2019 21:37:33 GMT -5
Thomas,
Some things are not clear to me...BK fuel is the combustor fuel flow and EBK is the afterburner fuel flow?
Are you commanding a rotor speed and letting the controller drive the fuel pump to achieve this within some limits (EGT, etc.)?
You mentioned that T52 is the turbine inlet temperature...is that a relatively large thermocouple? If so, you may have a slow TC response time that may drive some false behavior, since the TC is not following the actual gas temperature well. If you look at the T3, T52 and T7 responses after the fuel pressure is reduced and the GG speed reduces, the temps stay flat...seems odd. A quick fix may be to reduce the ramp rates to see if the engine and temp follow more closely.
Also looking at the T52 and T7 plots, does the temperature drop match what you expect from what is calculated from a compressor map? Without any scale on many of the curves, it is very difficult for us to make any decent calcs.
Also the plots might be more clear if you plot the combustor fuel pressure minus the P3 pressure...that may give a better trend..maybe? It does look like there are moments in the data when this would actually give zero or negative flow...so this might not actually work.
Good luck!
Chris
|
|
Sweetenough
Veteran Member
Joined: April 2016
Posts: 121
|
Post by Sweetenough on Apr 8, 2019 6:11:51 GMT -5
Hi, Electronics that are so nice :-) I use an ABB AC70 controller that fortnuattly have a lot of useful logic blocks used for larger turbines. Used to be a commissioning engineer for 50MW gas turbines with ABB control systems so that is why I am using with this system. So it is not electronics down to componet soldering level, just programing in an graphical interface with logger capability :-)
Thanks for your input Chris! Sorry for my late response. BK fuel is the pressure to the fuel injectors in bar EBK fuel is the same but the pressure for the EBK fuel nozzles. Yes, I am commanding a rotor speed and letting the controller achieve this exactly as you describe The T52 probe is a 5mm type K probe from china so probably not the most accurate or fastest Have not made much calculations at all, it is more or less a copy of Anders and Olovs engines they made for their dual engine jetkick. Have all the data is in excel format as well but to analyse it I imported it into a trend view program. Delta pressure over the fuel nozzles can be checked in excel easy but not in the trend viewer I think there is possitive fuel pressure "steady state" but at some transients with the long time scale it looks funny.
I can not understand why fuel/p3/T52/T7 is increasing at constant rotor speed.... My plans for next test is to verify if the speed probe is giving a correct value at high speeds. Will paint a white spot on the compressor and try a handheld meter and also use an sound app in my phone with FFT to verify if the frequency is constant at higher rpm. Also I will do some additional meassurements on the oil temperature to the bearing, do not really think it has anything to do with the problem but I need to strike from the list of causes in my mind.
Kind Regards Thomas
|
|
Sweetenough
Veteran Member
Joined: April 2016
Posts: 121
|
Post by Sweetenough on Jul 1, 2019 3:47:51 GMT -5
I think I might have found the reason for the strange behavior through some data analyse Not proven in operation yet, but it seems like a reasonable explanation at the moment at least. Plotted P3 vs speed for several runs and created a compressor line from the "best run" from behavior "2019-02-25 13:32" That also was the second run, an earlier run was done an half hour before so the engine was warm. When operating at constant measured speed the problem has been that P3 & fuel and other variables keeps on increasing. The speed measurement probe has been a suspect for some time but now with the plot it seems likely that there is some kind of "slippage" between real and measured speed. Will try to do some comparison measurement with and handheld speed meter and also an sound app in the phone with FFT analyse. One coincidence is that during the "best run" above the oil cooling did not work as planed, the temperature was increasing to higher levels than normal So first before any more test run, I will install an Earls oil thermostat to get better control over the oil temperature. Currently I use an oil cooler with fan but I only measure the temperature in the oil tank and not after the cooler to the bearings. And of course some more temperature sensors will be added Still amazes me how much heat that is transferred to the oil, from the start I thought that the oil tank in aluminium would be enough to cool the oil. The a small cooler was added with no fan, then a temporary stand still fan was added. Then as I hope would be the final solution a bigger cooler with permanent installed fan was added. I am also thinking about building a new after burner to get some use of my newly bought really old sheet metal rolling machine. Got some good ideas from Andys "special" about afterburners But what is the limiting factor for fuel to the after burner? Can burn as much fuel as there is oxygen? Had an idea to use a slight conical shape on the pipe from the turbine exhaust into the afterburner to act as an pressure recovery diffuser. Anybody tried that? Cheers Thomas
|
|
jetric
Veteran Member
Joined: December 2014
Posts: 132
|
Post by jetric on Jul 1, 2019 10:14:43 GMT -5
Hi Thomas, If you want the reheat/afterburner to work then do not deviate from my original design plans and scale it up or down to suit your turbine outlet diameter, It took me four years of testing various afterburner design configurations ( including completely burning all of the grass in my back garden and seriously annoying my neighbours with the noise when testing ) in order to finally arrive at the best stable design. Do not put a taper on the parallel tube from the turbine, you need to keep the gas velocity high so that when the gas exits the end of the parallel tube into the larger volume of the afterburner body the gases curl back on themselves on the outside of the parallel tube forming a donut shaped toroidal vortex ( same as a smoke ring ) the fuel/air gases trapped in this vortex are burning when the afterburner is ignited and this forms a flame anchor that keeps the reheat lit. I did try a vee shaped vapour gutter flame anchor in an early design but this was not very stable and was unsuccessful, the vee gutter system tends to work better on larger diameter reheats for engines like the size of a RR Viper. The end of the parallel tube needs a straight edge in order to create a stable toroidal vortex, do not put a wavy edge on the end of the tube as in the design modifications that Andy has done as this miss shapes the vortex and makes it unstable ( one of the main reasons that Andys reheats always blow out during a run ) I did warn Andy of this. With regards to fuel flow the reheat on my design drawing was supplied with fuel from a Bosch 044 electric fuel pump, The fuel rail discharge hole size and number of holes was calculated from the cross sectional area of the inside bore of the fuel rail tube, the calculated area of all of the discharge holes has to be equal to the area of the tube bore, this is to try to equalise the flow out of each discharge hole and to prevent the lower holes from flowing slightly more fuel due to the effects of gravity. Having the discharge holes in the fuel rail facing towards the turbine wheel greatly improves the vaporisation of the fuel, as with the tests that John conducted, although there was two inaccuracies with Johns tests ,First, he was using water and not kerosene, and second, he was using cold air supplied from the leaf blower and not using gas that was at 700 to 800 oC as in the exhaust gas from a turbine engine. Hope this helps, Richard Staveley (Jetric) I think I might have found the reason for the strange behavior through some data analyse Not proven in operation yet, but it seems like a reasonable explanation at the moment at least. Plotted P3 vs speed for several runs and created a compressor line from the "best run" from behavior "2019-02-25 13:32" That also was the second run, an earlier run was done an half hour before so the engine was warm. When operating at constant measured speed the problem has been that P3 & fuel and other variables keeps on increasing. The speed measurement probe has been a suspect for some time but now with the plot it seems likely that there is some kind of "slippage" between real and measured speed. Will try to do some comparison measurement with and handheld speed meter and also an sound app in the phone with FFT analyse. One coincidence is that during the "best run" above the oil cooling did not work as planed, the temperature was increasing to higher levels than normal So first before any more test run, I will install an Earls oil thermostat to get better control over the oil temperature. Currently I use an oil cooler with fan but I only measure the temperature in the oil tank and not after the cooler to the bearings. And of course some more temperature sensors will be added Still amazes me how much heat that is transferred to the oil, from the start I thought that the oil tank in aluminium would be enough to cool the oil. The a small cooler was added with no fan, then a temporary stand still fan was added. Then as I hope would be the final solution a bigger cooler with permanent installed fan was added. I am also thinking about building a new after burner to get some use of my newly bought really old sheet metal rolling machine. Got some good ideas from Andys "special" about afterburners But what is the limiting factor for fuel to the after burner? Can burn as much fuel as there is oxygen? Had an idea to use a slight conical shape on the pipe from the turbine exhaust into the afterburner to act as an pressure recovery diffuser. Anybody tried that? Cheers Thomas
|
|
Deleted
Joined: January 1970
Posts: 0
|
Post by Deleted on Jul 1, 2019 12:28:10 GMT -5
Hi.... sorry Thomas, just high jack your thread one minute Jetric, do you see the afterburner blowing out in any of these videos An issue started, and had a few blow outs, but that has been sorted.... and some of the time it was operator error As for the hole size, i have never follow that calculation forum, as it dose not take into account flow and pressure of the pump against the pressure in the afterburner tube. No Richard you did not tell me any thing about unstable vortex, what you said was the edges would melt off, and have not. To think in my last video i gave you credit for your work on afterburners Lets see what happens with the new afterburner on the new engine
|
|
|
Post by racket on Jul 1, 2019 17:35:36 GMT -5
LOL................I'll have to jump in as well ..................."inaccuracies" in my testing ?? ..............please explain !! Certainly I used water rather than kero and cold air rather than hot gases , but unless you can produce data on how your injected fluid stream is being affected by the hot gas stream , and how its "destruction" differs with my findings , you cannot claim mine were "inaccuracies " . You will have undoubtedly read the A/B injection and design data for the AiResearch engine of the 1970's arc.aiaa.org/doi/abs/10.2514/6.1978-965 with its excellent data Cheers John
|
|
jetric
Veteran Member
Joined: December 2014
Posts: 132
|
Post by jetric on Jul 2, 2019 4:44:27 GMT -5
John/Andy, Yes john there was some inaccuracies in your testing!!! So you are trying to tell me that the Kerosene that will be coming out of the discharge holes in the fuel distribution ring in the reheat will not be flash boiled into a vapour by the 700 to 800 oC exhaust gas from the engine!!!!!. No I have not seen the paper that you were referring too as when I was doing my R&D on my reheat design back in 1999 there was not much information available on reheats for DIY turbine engines, certainly there was no chat groups such as this one and the Yahoo groups one. John it would be nice if I was shown just a little respect, I have been doing this hobby since 1995 and have done quite a few developments for the Turbocharger based DIY turbine engines, It just seems to me that as soon as I put a post up on this group then I have you and Andy rip into anything that I say, Does this chat room not promote freedom of speech and opinion? John I hold you with the upmost respect, the projects that you have done have been amazing ( I especially liked the turbine bike and loved watching the on board video ) perhaps I might get time to do more projects like yours when I am retired!! Andy did you not quote on your video of Tom Bagnalls land speed run at elvington that the reheat blew out mid way along the run when the record was set and said `There's more to come' and wanted to do a full run without it blowing out to see the full potential of the kart, Oh and by the way when are me and martyn going to get invited to Elvington?, did you not get a full mard on when me and Martyn went to Elvington but did not invite you!!! I couldn't invite you because it was Zef isenburg the owner of the Y2K turbine bike that arranged it for me as a thank you for doing the work on his BMW/RR T312 turbine engine, but you phoned me and your first words were `I AM NOT IMPRESSED ' ,not well done mate for setting a new record!!. Andy yes I did warn you of the unstable vortex and the tips probably melting on the wavy edge on the parallel tube, you only thought of the idea after I showed you the picture from my works RR magazine showing the saw tooth design trailing edge on the bypass duct on a RR turbofan, at the time you were trying to improve the propulsive efficiency of the reheat by using the coanda diffuser pipe on the reheat outlet. All I was doing was advising a member on here about reheat design and telling them a design that is guaranteed to work, what people do with my reheat design once they have it working is up to them, just like you with the wavy edge, I am only trying to save people the time and wasted material of a reheat that does not work!!! Andy, over the years you have caused me huge amounts of stress such as shunning me from the UK DIY gas turbine group, maybe its because you see me as the Competition, The DIY turbine hobby was a very big part of my life that I used to enjoy! not so now. I used to love giving people help and advice but not when they use it against me. Rich. LOL................I'll have to jump in as well ..................."inaccuracies" in my testing ?? ..............please explain !! Certainly I used water rather than kero and cold air rather than hot gases , but unless you can produce data on how your injected fluid stream is being affected by the hot gas stream , and how its "destruction" differs with my findings , you cannot claim mine were "inaccuracies " . You will have undoubtedly read the A/B injection and design data for the AiResearch engine of the 1970's arc.aiaa.org/doi/abs/10.2514/6.1978-965 with its excellent data Cheers John
|
|
Deleted
Joined: January 1970
Posts: 0
|
Post by Deleted on Jul 2, 2019 9:55:29 GMT -5
Sorry Thomas
For Richard
“Andy did you not quote on your video of Tom Bagnalls land speed run at elvington that the reheat blew out mid way along the run when the record was set and said `There's more to come' and wanted to do a full run without it blowing out to see the full potential of the kart”
Yes Richard, I did say that, but like already said, and show in videos….. it does not always blow out “(one of the main reasons that Andys reheats always blow out during a run ) I did warn Andy of this” and it did not melt the edge off the wavey bit. No you did not tell me about the vortex, but that it works clearly proves something ? Yes you did show me a picture of the saw tooth design on the trailing edge on the R/R turbofan bypass duck, perhaps you or R/R should get credit for me trying it on my afterburner that don’t work
“Oh and by the way when are me and Martyn going to get invited to Elvington?, did you not get a full mard on when me and Martyn went to Elvington but did not invite you!!! I couldn't invite you because it was Zef Isenburg the owner of the Y2K turbine bike that arranged it for me as a thank you for doing the work on his BMW/RR T312 turbine engine, but you phoned me and your first words were `I AM NOT IMPRESSED ' ,not well done mate for setting a new record!!.”
Yes I did say “I was not impressed”, but not because you went to elvington, or that Zef had sorted it etc. But what I was unimpressed with was, some one I thought was a friend, that we had been working together to promote the hobby. Would lie to me, play games …. Come up with some story about it being a secret, when it was on facebook, and done in a public place ? But why you have brought this up now and here…. Well I will let others make there own conclusions. As for the record, you never called to tell me… it was days after you had been Elvington we spoke…. How could I know about the new record, it was a secret, in a public place
“Andy, over the years you have caused me huge amounts of stress such as shunning me from the UK DIY gas turbine group, maybe its because you see me as the Competition, The DIY turbine hobby was a very big part of my life that I used to enjoy! not so now. I used to love giving people help and advice but not when they use it against me.”
I have never shunned you. I have credited you in my videos for you work, for example mentioned you in both my last two videos. I held you in high regards, that is until…..
|
|
|
Post by racket on Jul 2, 2019 18:06:02 GMT -5
Hi Rich
Please don't take offence at my comments ...................you simply used the wrong wording , "inaccuracies" was not the correct one to use as it implied my readings were wrong , a better word would have been "differences" due to water vs kero and cold vs hot .
The AiResearch Paper has been around the DIY Turbine scene for a very long time , I think it was Mike Early a retire Garrett aero engineer who first bought it to our attention , it really is a very good Paper to read , the sudden expansion dump afterburner is only an offshoot from the ramjet combustors .
If you read the Paper you will understand that our fluid jets don't suddenly vapourise, they take time, exactly as it does within the gas producers combustor where even droplets at micron level still take milliseconds, a fluid stream in a gas stream at several hundred feet per second will take considerable distance to breakup and vapourise
Yes the 1990s was a difficult time for us DIY'ers , having started my bike project in 1991 I found it very hard to find any information on jet engine design/combustion here in Oz , you were fortunate to live in the "birthplace" of jet engines rather than 10,000 miles away .
As for respecting you , I will show you the same respect as I do everyone ....................but when someone questions the accuracy of my contributions , well ..............you be the judge.
Cheers John
|
|
Deleted
Joined: January 1970
Posts: 0
|
Post by Deleted on Jul 3, 2019 15:22:54 GMT -5
Once again Thomas sorry, and i should of perhaps left this a loan But Richard you said this "It just seems to me that as soon as I put a post up on this group then I have you and Andy rip into anything that I say, Does this chat room not promote freedom of speech and opinion?"Yes the forum is about sharing ideas, freedom of speech etc. We are not ripping into you, In your first post you was dismissive of both mine and John's comment and advise. Which yes you have a right to do. You have every right to express your opinion, and for me your help in the past has been respected. But am I not aloud to defend myself, express my opinion ??
|
|
|
Post by finiteparts on Jul 6, 2019 16:32:00 GMT -5
Wow...a bit of a heated post (pun definitely intended!!! Ha!!) Rich, I haven't seen where you posted your work and I apologize for this, but I always assumed that this dump style combustor evolved from the information that is out there in open literature. John posted one paper, but so everyone has access to the information, here are two other good sources of information on the development of this style afterburner... Airesearch's "Alternative Subsonic Low-Cost Engine" apps.dtic.mil/dtic/tr/fulltext/u2/a067277.pdfMajor Rami Dotan's Master Thesis, "Turbojet Thrust Augentation by Sudden Expansion Type Afterburner" apps.dtic.mil/dtic/tr/fulltext/u2/a081901.pdfI personally dislike this style AB because of the basic aerodynamic flaws it possesses. To start, it must be understood that there is a loss of total pressure associated with heat release in a confined duct, called the Rayleigh pressure loss, which is a strong function of the inlet stream Mach number. Also, as the duct inlet Mach number is increased, the amount of heat that can be added to the stream decreases, due to choking of the duct flow or downstream controlling areas...this is called thermal choking. Any compressible flow book covers this under the term Rayleigh Flow. This is the reason that combustors are preceded by diffusers, so that we can minimize this loss through the combustor. In afterburners, the local Mach number is usually much higher than the typical combustor and as such they suffer from higher total pressure losses. This pressure loss due to heat addition translates directly to a lower nozzle pressure ratio and thus a lower overall thrust. If we look at this AB system, the straight tube keeps the stream Mach number up till the flow hits this dump diffuser stage, which serves two purposes. The first purpose is to "help" reduce the stream Mach number from the high velocity in the small tube to the lower Mach number in the larger tube. Secondly, as you point out, the recirculation in the corner sets up a toroidial vortex to stabilize the flame in a similar manner to the vortex bubble in the main combustor, by providing low velocity regions that the flame can hang out in and the recirculation provides a means to refresh the air+fuel mixture at the flame front. So from the diffuser standpoint, dump diffusers are poor diffusers, but sometimes they "make sense" when length is limited and performance is not paramount. When you design a diffuser, you are trying very hard to keep the flow from separating from the surface. But by definition of a dump diffuser, you are creating a flow separation, which means a higher pressure loss and thus increased entropy. If you look at aircraft engines that use dump diffusers, they have a pre-diffuser section that is in itself a shallow annular diffuser. Then to create a stable vortex, many of them will extend the pre-diffuser past the dump wall, which could easily be done for these AB applications. The tube wouldn't "burn-off" because it has high velocity "cool" gas on the inside of the extension (assuming the extension thickness is small like a typical sheet metal). Back to the combustor comparison, there are many dump style diffusers that don't have the extended pre-diffusers, but they can get away with it because they have the downstream flow blockage of the combustor dome and other structures. This means that the pre-diffuser exit streamlines are not fully axial, but instead, they have a diverging nature as they swing around the dome, thus the vortex shear interface is pushing outward more than would occur in the AB application. Now, if we look at the heat release, it is likely that there is a significant amount happening in the high velocity region. In my opinion, it is better to put the flame anchoring in a known low velocity region after a proper diffuser. But this is a muddier point and could be a source of disagreement on where the bulk heat release is occurring, so I will not belabor this topic. Richard, John and Andy, don't take this as any statement about your work, your designs or experience. I am only giving my opinion on this heated topic. I do realize that this design is a very simple means to achieve an AB on these engines and as such fits very well into this hobby. My opinion is that I would encourage others to experiment on this design, especially since it is fundamentally flawed from an compressible aerodynamic design perspective. I agree with your statement about you getting one to work if the reader is someone that is just wanting to copy and run...but there is much more benefit to be had in my opinion and if they are willing to take on some risk, they can move this farther forward. I hope to someday soon, get some free time to get back out in the garage and build one of my own. I know the old adage is "Put up or shut-up"...so take this as my opinion only for now, maybe in the near future I can have video and data to support my claims. Some good reference texts on the subject include: Oates, G. (Editor), "Aerothermodynamics of Aircraft Engine Components" and the older, but still relevant, Zuyev and Skubachevskii, "Combustion Chambers for Jet Propulsion Engines" Enjoy! Chris
|
|
|
Post by racket on Jul 6, 2019 18:12:03 GMT -5
Hi Chris LOL......just a robust exchange of ideas ;-) When I did my A/B on the GT6041 I tried going down the diffuser first route, jetandturbineowners.proboards.com/thread/78/garrett-gt6041-powered-kart?page=9 , then having the spray ring and V gutter flameholder . I ran into problems mainly due I feel ( in hindsight) from attempting to use only a 6" dia A/B tube which meant the gas speeds after diffusion were still ~400 ft/sec , an increase up to a 7" or 8" tube with lower gas speeds would have been a wiser choice , but I wanted to push the boundaries as usual . When Smithy fitted an 8" tube the A/B started to perform as it should , with the 6 incher the V gutter area had to be kept to a minimum to kept gas speeds past them down, but with the extra cross section of the 8 incher things sorted themselves out nicely ..............Smithy did a beautiful job of getting her sorted .............thanks Smithy :-) I'll check out those couple of extra Links you've provided , thanks for that Cheers John
|
|