rolandsean
Member
Joined: September 2020
Posts: 13
|
Post by rolandsean on Oct 5, 2020 23:15:19 GMT -5
Thank you, Chris! I greatly appreciate you taking the time to expand on those concepts more. It will take me some time to put that into application but I understand the theory more thoroughly now.
-Sean
|
|
|
|
Post by finiteparts on Dec 21, 2020 19:04:59 GMT -5
In case you haven't seen this, it is a great description of the Marx generator....that could easily be modified to build a great ignitor driver.... www.youtube.com/watch?v=dje7uhyW23oEnjoy! Chris |
|
|
|
Post by wannabebuilderuk on Dec 22, 2020 5:00:58 GMT -5
|
|
|
|
Post by mecarloschavez on Jan 5, 2021 22:21:15 GMT -5
hello, i find this very interesting. i havea question theorically, air mas flow inlet can burn x amount of fuel how can i calculate how much fuel is needed to properly size an atomizer for a chamber in case i want to make a liquid fuel atomized combustion chamber ?,, also based on this how much power can it produce in optimum eficency, i want to have a reference for that because i want to build a turboshaft engine. thank you !! Ok, back to combustor design. I think I will skip right into the calculations and then discuss the theory as we go along. I will be working through the calculations for a can style combustor for my Borg Warner EFR6758 turbos. I gave some of the design point specs on the previous page, but here is a refresher: Liner outside diameter = 5.0 inch
Combustor casing inside diameter = 6.0 inch
Compressor discharge temperature, T3 = 340 F
Compressor discharge pressure, P3 = 47 psia
Pressure drop = 5%The approach that I am using is from Arthur Lefebvre's book, "Gas Turbine Combustion", but I will also shamelessly steal techniques from other books. I highly recommend these books in addition to Lefebvre's book: Aircraft Engine Design, 2nd edition, Jack Mattingly, William Heiser and David Pratt, AIAA Education Series, 2002
Design of Modern Turbine Combustors, A.M. Mellor, Academic Press, 1990
Ok, so I will just start with the basics. Once we have some design point numbers and a target combustor pressure drop, how do we figure out how many holes to put in the liner? I am a strong believer in showing by example....so I will work these out for my combustor and you can just change the relevant data to work your design. Also, I highly recommend that you always check your units! They beat this into you in every engineering class and it really does help to catch errors...so you will see in my calcs how I was taught to check units. First, Lefebvre developed a few reference parameters. The first one is a reference area and it should be noted that it is just the open area inside the combustor casing...we don't have to concern ourselves with the liner size yet. The properties are the compressor discharge temperature and pressure, not the hot gas properties in the liner. This is because what we are basically trying to do is ise the orifice area that produces the required pressure drop across the liner. The hot gases have a pressure loss of their own, called the fundamental pressure loss, which should be very small if the Mach number in the liner is low. We will ignore these for the moment, but may touch back on these later.  Now that we have the reference properties, we can calculate what is termed the pressure loss factor. The PLF is used to get the effective flow area requirements from the reference area.  So we now know that we want to end up with a total effective area across all the liner holes and any other leakage paths to be around 1.838 in^2. Now, there needs to be some logic on how that effective area is distributed between the various zones of the combustor. I will start with the primary zone. Since I have a primary zone equivalence ratio target already, this is relatively straightforward on how much air I will need in the primary zone.  With the slightly rich head end, I will need to put 21.3% of the engine mass flow into the primary zone. Now, it should be understood that this is a rough estimate. When we have recirculation zones, some of the secondary air, the burning products, etc., can get pulled back into the primary zone and change this equivalence ratio. The slightly rich head end is a result of following the work of Bragg and others that have shown that the maximum consumption rate of incoming fuel-air occurs in a slightly rich mixture, not in a stoichiometric mixture. If you were trying to put a stoichiometric ratio in the head-end, you would target around 27.8% of the air flow to be in the primary zone. If you put 30%, you are at slightly lean condition, around a phi = 0.93 (phi is the symbol for the equivalence ratio). Definitely a safe spot...anywhere near a phi = 1 is ok and I will show some of the stability curves next time as well as discussing how other factors play into sizing the primary zone. We also will need to size the secondary and dilution zones, and then work out how we get the effective areas translated into geometric areas so that we can start drilling holes. Stay tuned! Chris |
|
|
|
Post by racket on Jan 5, 2021 22:51:16 GMT -5
|
|
|
|
Post by finiteparts on Mar 6, 2021 17:48:20 GMT -5
Hi Chris Would you like to do a worked example for us as its often easier for us DIY'ers to understand than a theoretical non dimensional design. I used journals.sagepub.com/doi/abs/10.1243/PIME_CONF_1968_183_245_02 when designing the TV84 combustor , not that I knew what I was doing at the time , but it'd be interesting to see what mistakes I made , I have a fair bit of info on the outcomes that might be able to be cross referenced  1.8 lbs/sec at 3.8:1 PR at 70% calculated/measured comp effic Flametube 145 mm dia by 400 mm long , total volume ~ 1/4 cubic foot Swirl vane passageways were 9 X 5 mm X 12 of Cap had 36 X 3mm dia plain holes Primary 12 X 8 mm plain holes Secondary 15 X 11mm dia plunged/bellmouthed Tertiary 10 X 16.5 mm plunged/bellmouthed The wall cooling louvres were equivalent to 23 X 3 mm holes at each of the 3 bands . It ended up with ~29% primary 26% secondary and 45% tertiary . At a latter date I added 23 X 5 mm dia holes at the bottom of the Primary Zone to help the louvres cool the wall as there was heat damage to the 0.5mm stainless on the louvres where they projected into the flametube. TIT ~900 C Jetpipe total pressure ~12 psig Cheers John John, I finally got a few minutes to take a look at your TV84 combustor. I ran it through my program to see how my design process may differ and here are the results. Now, I will say that my program is still in progress and there are many things that I still need to incorporate and I will try to point them out as we come across them. OK, so in this first screen grab you can see some of the inputs that I used. First off, I had to take a guess at your combustor casing size, which I put as 6.2 inches. We can see that the length to diameter ratio is above 2 so, we are in a good design area and the combustor delta temp is not very large. Now moving to the reference quantities, nothing seems to stand out as crazy...  Moving on, now we are looking at some of the "performance" drivers. We can see that we have a fairly low entrance velocity to the head end...now I also would have to do some estimating here, but instead, I just assumed a 4 inch inlet tube diameter, so that is what is reflected in the "Inlet" parameters. The most striking value is the flow speed in the annulus surrounding the liner. Now again, I assumed a 6.25 inch inside diameter, so there might be some error introduced due to that....but it is a good point of discussion. As you can see in the note, it is recommended to keep the velocity in the annulus below ~ 200 ft/s due to the incurred pressure losses and the poor feed of the liner holes. Unfortunately, the total pressure loss due to friction shown is only for the inlet tube, I haven't coded up the total pressure loss in the annulus yet...so we can't bound what that loss is at this time, but we know that slower is better and this should be the reason that you try to have a wide annulus between the outer casing and the liner. The loading parameter is given in the two commonly presented units. The space heat rate is pretty low, so there should not be a real challenge on cooling the liner. The residence times are relatively "long", so there should be plenty of time to vaporize and react well before leaving the combustor.  Here we see the chemical proportioning...the overall equivalence ratio (phi) is just the previously show Fuel to Air ratio (FAR) over the FAR that would be required to achieve a stoichiometric reaction...thus you can see that only about a quarter of the air going through the engine participates in the reaction. The PZ phi is slightly rich, but not so rich that we might get soot forming in the headend reactions. The PZ and SZ phi's are set upfront in the program as a user input and this is just where I choose to set them for full power conditions. After that I break out the individual flows in lbm/s and as a percent of plane 31 (combustor diffuser exit) flow. Now I did notice an error, the PZ flows are not a percentage of W31, but rather as a percentage of the flow entering the PZ. Here is an area that John and my calculations differ. As John stated, he was trying to put 29% of W31 ("W" means weight flow, and 31 means it is at Plane 31) into the PZ, while I would suggest putting in around 20%. I do have to check how the cooling in the PZ was booked, we might have a small change in the 20.4% due to a potential addition of some local cooling.  Here is a very interesting region. John's liner hole area sums up to around 8.45 in^2 and my total geometric liner hole area sums to roughly 8.08 in^2! We are pretty close on that! You can see that the high annulus velocities have suppressed the cold side local static pressure so much that the hole Cds are way down in the 0.3's! This is why all the work is done to lower the annulus flow speeds...slow speeds mean that the local static pressure and total pressure are much closer. When the annulus flow speeds are high, the local static pressures are low and thus they have lower force with which to push the flow through the liner holes. Many people get all wrapped around the axle on total and static pressures and bungle this up...the important thing to remember is that total pressure is imaginary...you can only reach it if you somehow magically bring the flow to rest without introducing any loss! So when you are thinking about forces and flow directions, think in static pressures. Total pressures are usually reserved for energy type calculations, such as trying to figure out how much power the turbine will extract due to the change in the TOTAL pressure. We can also see how plunged holes help fix jet penetration into the combustor somewhat, when the annulus velocity is too high. Here we see that a std hole gives a Cd ~ .33 and a plunged hole Cd~0.44. The low "cold dome velocity" suggests that the combustor liner diameter at the dome (PZ) could have been a bit smaller, the high residence times also suggest that the combustor could have been shorter and the low space heat rate also supports the idea of a larger than required combustor volume. But, it is better to lean to the "too big" side that to the "too small" side...one works, the other doesn't.  The secondary and dilution zone hole sizing is shown here. You might have also noticed that I do a hot to cold calculation on the hole size...I still have to code in for it to output the cold hole size, since all the numbers shown are for the required hot hole size. It is around a 3% for the assumption that the liners are running around 1400 F at all locations...which is obviously not realistic. It is really just more of a "bounding" number and likely, we can't control the effective areas of the holes to anything near +/- 3%!  The final section shows the zonal breakdown of the annulus flow. Since some portion of the flow goes into to the combustor, using a fixed velocity through the entire annulus would be incorrect...so I recalculate the passage velocity after each major flow split (i.e. after the PZ flow is pulled off, after the SZ flow is pulled off, etc.). I also assume the flow is from the dome end and flows toward the DZ end, which looks to probably be incorrect for John's design. I may code in an option to switch that now that I see this issue. So Zone 1 is the larger open area near the plane of the swirler inlet, Zone 2 is the annulus upstream of the PZ holes, Zone 3 is between the PZ holes and the SZ holes and finally, Zone 4 is between the SZ and DZ holes. You can see that due to the reduced mass flow in the aft end of the annulus, there are lower flow velocities and thus higher static pressures helping to feed the DZ holes. As you might have guessed, this translated to a slight repartitioning of the flows as engine mass flow changes. Since lower mass flows will have better hole Cd's, the holes will be more evenly fed. As the mass flow increases, the high flow speeds in the annulus will cause a maldistribution of the static pressure in the annulus and actually bias more flow to go out the DZ holes. The Cd's shown here do a great job illustrating this point.  Now, I have future goals to improve this program, but I thought that it might be interesting to run through this on John's successful combustor design. As I said before, some of this might be incorrect due to the assumptions that I made, but I hope it is at least informative. Also, as I always state, all of this work is covered in Lefebrve's " Gas Turbine Combustion" book and/or Mellor's "Design of Modern Turbine Combustors"...both are highly recommended. - Chris |
|
|
|
Post by racket on Mar 6, 2021 20:29:34 GMT -5
Hi Chris
Thanks for running the numbers :-)
A few extra bits of info that I've been able to find .........
The upper portion of the outer can was constructed from a 7" OD fire extinguisher ~177mm , with ~1mm thickness , so ~6.9" ID -175mm which allowed 15mm clearance between FT and outer can for the Primary and Secondary holes.
The lower section of the outer can is ~205 mm providing ~30 mm clearance to the large Tertiary holes being fed from the air delivery tube's "diffusing funnel" , the 30mm clearance also hopefully allowed the air to disperse more easily around the flametube .
My thoughts at the time were that 50% of the air would have been swallowed by those Tertiary holes making it easier for the Secondary and Primary holes to cope with only 15mm passageway clearance , theres a "bellmouth" between the 205mm and 175mm sections at the Vband connecting flange .
From my ....LOL, "notes??", it appears I worked on an air speed of ~100 ft/sec at the outlet of that bellmouth through the ~0.066 sq feet of gap for the remaining half of the air , at that time I was hopeful of flowing ~2 lbs/sec from the comp , so "calculated??" for ~1lb/sec to the secondary and primary holes .
I now wish I'd been more dilligent with my book keeping at the time, theres not much recorded , guess I wasn't planning on still being at it 30 years latter .................I got hooked on turbines :-(
I'll sit down after lunch for a Sunday afternoon read and "digest " your calcs , thanks again :-)
Cheers
John
|
|
philip111
Member
Joined: September 2022
Posts: 11
|
Post by philip111 on Oct 24, 2022 10:36:09 GMT -5
|
|
philip111
Member
Joined: September 2022
Posts: 11
|
Post by philip111 on May 17, 2023 12:12:13 GMT -5
I was curious if there is a difference between using propane or butane, or are they effectively the same for DIY gas turbines.
|
|
|
|
Post by racket on May 17, 2023 18:40:09 GMT -5
Propane has a higher vapour pressure at lower temperatures so is preferable to butane .
As we draw off fuel the liquid cools as it boils off the gas , eventually the vapour pressure drops to a point where it can't overcome the combustor air pressure and the engine slows down and hopefully is shutdown before any damage is done .
Stick to propane for a GT
Cheers
John
|
|
|
|
Post by finiteparts on May 28, 2023 16:53:00 GMT -5
I had a member reach out to me about the reference area calculation (Eqn 4.12 in 1st edition) given in Lefebvre's book (or one of his papers), producing unrealistic cross-sectional area requirements. The provided information appeared to contain inconsistent units and that is likely the culprit of the excessively large area results. I worked out the equation using his provided data and a few assumptions, primarily that the specific gas constant for air is 53.35 ft*lbf / (lbm*R), which is usually sufficiently accurate for our calculation purposes. Here is how I worked it out:  The one thing that they drill into you in any good engineering class is that you have to show and work the units in addition to the variable values. If you look close, you will see that the units are placed beside their respective variable value and crossed out when they cancel out. Doing this, allows you to find errors in your equations or other parts of your work because you may get a value, but if the units don't work out, you know you have an error somewhere. I have found errors in textbooks by using this method, including Dixon's "Fluid Mechanics and Thermodynamics of Turbomachinery", Baskharone's "Principles of Turbomachinery in Air-Breathing Engines", several NASA papers among others. Luckily, they already were aware of the errors and were planning to fix in their next editions...because engineers and engineering students are very well empowered to recognize these errors by the above cited method. I hope that helps others out there too, Chris |
|
|
|
Post by britishrocket on Jun 14, 2023 21:11:23 GMT -5
I have now got to the end of this thread and I am far more informed and educated than I was. Thank you to finiteparts and racket for their extensive contributions, and thank you to everyone else who has taken their time to share information. Especially the links to technical papers, all of which I have found extremely useful.
I have come to a few tentative conclusions after reading the material presented here, as well as Lefebvre and quite a few other texts on gas turbine and combustor design. I'd be interested to hear what others think.
There is an old adage in engineering - and in other fields of endeavor too - that an ounce of practice is worth a ton of theory. Do not read that sentence and immediately think that I am decrying theoretical knowledge. I am not - I have always tried to base the design of anything I make on sound theoretical principles.
What I am saying is that when it comes to building a gas turbine from an automotive turbocharger, there are so many uncertainties and variables that we may as well use a broad brush tool like JetSpecs and not get too hung up on the minutiae. It seems to me that we are dealing with something of a "black art". For example, I have heard about the "20% rule" when it comes to selecting a turbocharger as a suitable candidate for gas turbine conversion. I understand well the thermodynamics as to why this would be the case. And yet the internet is full of units that do not obey this rule but run well. This state of affairs reminds me of another old saying that an aerodynamics lecturer I once had was fond of. That bees are aerodynamically incapable of flight, it is just that no one has told them.
Similarly, Chris's analysis of John's combustor, which has clearly been built in accordance with good principles in terms of airflow regime shows that it is less than ideal in many respects, with low Cd values caused by too high an annulus flow velocity for some of the flame tube holes. And yet, this combustor clearly works and works extremely well, and is rightly held up as "best practice" on this site.
I had always felt that the "typical" amateur gas turbine combustor design that is seen on this site and online in various guises was poor, in terms of the compressor entry being tangential and thereby giving rise to a swirling flow inside the can which will reduce static pressure at the very point where we want to increase it. That said, looking at some of the CFD plots showing different arrangements given in this thread, tangential entry, albeit improved by the addition of a diffuser, would seem to be a more than viable option.
JetSpecs has been called an unscientific tool or maybe a sledgehammer to crack a nut. Given the broad tolerances we are dealing with when it comes to turbocharger gas turbine conversions, this may be the best approach. Get into the ballpark with JetSpecs and then refine based on performance. An empirical approach that is informed and advised by the theory.
I would be very interested to see the mathematical basis behind JetSpecs and the assumptions that it is based on. Would anyone have that information, and if so could it be published here?
I'd be very interested to hear what others think.
|
|
|
|
Post by racket on Jun 15, 2023 0:47:05 GMT -5
Hi Back before JATO came into being a decade ago , there was the Yahoo DIY Gas Turbines Group , there were a number of Members having problems with their engine builds because they didn't have any info to guide them . So I wrote up some guidelines using the experience I'd gained whilst helping get those engines sorted , those guidelines were shifted to JATO and are now contained in jetandturbineowners.proboards.com/thread/680/diy-turbines , Combustion Chamber Section . One of the DIY GT Members, Jesse , approached me about making a Program ..............JETSPECS , he could do the programming and I'd help with "numbers" , namely those I'd already written up and anything else that was needed. We fiddled around for quite a while trying to "improve" on my written guide , but in the end we decided that there were simply too many variables , so took the KISS approach . I've been approached over the years to do an update , but with there being thousands more configurations of turbos now available , theres an even greater need to keep it simple , some guys will want to use old style turbos with mediocre performances whilst others will spend the cash and use high flowing more efficient units , how to incorporate those unknowns into a program would be impossible unless comp and turb maps were available for those thousands of turbos so that the builder could punch those numbers into the program . LOL............yep , there are "exceptions to the rules " but just how well they perform without "adjustments" is debatable , heh heh , I'm guilty of trying :-) My first build using the Garrett TV84 turbo taught me the importance of having that large turb exducer after a couple of expensive turbo repairs , but a "LARGE " turb exducer doesn't always mean a larger exducer face area , its greatly influenced by the exducer throat area , but as most builders don't want to get into modifying their turbos exducer , we stick with the exducer face area being 20% larger than the comp inducer face area . But , we can come unstuck with another variable like the comp inducer tip angle , a normal comp has an angle ~24 degrees , but the high flow ones use 10 degrees more and will bite off a lot more air , necessitating an even larger turb exducer flow area ................variables , variables and more bloody variables :-( So to conclude , there isn't any mathematical basis behind Jetspecs , it was based on my gut feelings and the results obtained from helping a lot of guys to get their engines sorted, If anyone would like to do an updated version I wish them well . Hope this helps Cheers John |
|
|
|
Post by britishrocket on Jun 15, 2023 6:14:59 GMT -5
Hello John,
Thanks for that reply. I think we are broadly in agreement in terms of the number and different type of turbos that exist and that fact that this gives rise to a seemingly never ending list of interdependent variables for designers to cope with. I think that JetSpecs is an excellent tool and the results it gives do tend to agree reasonably well with calculations when the data for a specific turbo is known.
I've been visiting this forum now on and off for the best part of a decade. I have learned a vast amount from yourself and people like Chris. All of you give of your time and knowledge freely and for that I am extremely grateful.
I would not even attempt to update JetSpecs, that would be far too great a task for my level of knowledge. As I said in my earlier post, I think a combination of different design methods need to be adopted. That includes using JetSpecs, combustor design like Chris has shown and also "rule of thumb" best practice for flow regimes that are based on theoretical information.
Thanks again,
Carl.
|
|
|
|
Post by racket on Jun 15, 2023 19:48:33 GMT -5
Hi Carl If both comp and turb maps are available, as I had with my GT6041 build jetandturbineowners.proboards.com/thread/78/garrett-gt6041-powered-kart the job is much easier as the turb map gives potential corrected flows with the different scroll A/R housings , but there are still unknowns with regards losses getting air into/through the combustor and with the actual combustion process depending on fuel type and presentation . Yep , Jetspecs will get the engine going , but theres still opportunities for developmental changes to fine tune the engine . LOL .........even with a lot of work theres still the chance something "simply" holds back the performance, so its critically important to do thrust testing to cross check against the theoretical calcs , just because an engine makes lotsa noise and appears to be working well , doesn't mean its actually performing to its potential ................ how well I know :-( Cheers John |
|
