Combustion Design

[turboron]

Chris/Patty, under the heading of "better late than never" I recently read Chris comment on March 12, 2017 of this thread about my statement that the Critical Pressure Ratio for air was 2.4(my post on Fat Boy March 8, 2017). I have some comments on this and the Critical Pressure Ratio in general.

1. I worked for a major industrial compressor manufacturer in the early 1970s. We made compressors with compressor cases are up to 10 feet in diameter with impellers up to 5 feet in diameter for all types of gases from hydrogen, air, propane, etc. The source of the 2.4 Critical Pressure Ratio for air was the very senior Systems Engineering Supervisor. I do not know his basis for the number.

2.The formula that Chris posted has a typo. The 2/(k-1) term should be 2/(k+1) as shown online and several reference textbooks. The numbers you state are correct.

3. Crane's Flow Handbook shows that the formula we are discussing is for an orifice. The graph they provide in Appendix A page A-21 provides data expansion factors multipliers for nozzles. This reference is available online.

Thanks, Ron

[finiteparts]

Hi Ron,

Thanks for pointing out the mistake! I should stop trying to rely on my memory...it constantly lets me down. I will go back and fix it so no one uses the incorrect form...and try to check my formulas before I post.

The Critical pressure ratio of 2.4 still baffles me...if I crank down the nozzle efficiency to near 70% for a gamma of 1.33, I can get the critical pressure ratio to be around 2.4, but that would have to be one really poorly performing nozzle to only hit 70% efficiency! If you find out sometime, please follow up and let us know.

I am not sure what formula that you are referring to, but the one that I was citing is just the isentropic static to total pressure equation evaluated at M=1. We had to use Cranes's handbook in undergrad classes and it always bugged me that they didn't specifically call out static and total properties in their formulas...it was a constant source of error in homework problems. After working in the industry you get a better feel for things, but the initial problems that I had with Crane's has limited my use of it...well, in reality, at work we have actual, good test data from current or past designs to refer to, so those compendiums are of very little use and thus they sit on my book shelves at home..dusty. Ha!!!!

Thanks,

Chris

[turboron]

Chris, many of the industrial compressors I worked on had a balance piston that was balanced to the suction of the compressor. As the laby seals opened up with wear the seal leakage increased. My guess is that the system engineer had observed that the volumetric flow increased until the critical pressure ratio of 2.4 was reached. Of course, the mass flow increased when the upstream pressure continued to rise. As I surfed around to understand this issue I noted one paper said the upstream losses could effect the critical pressure ratio. This would imply that location of the pressure measurement be a factor.

Also, I designed a anti-surge blow off start valve for a new buz jet engine a few years ago. It was behind a single stage fan. I pointed out to the aero team that the critical pressure ratio of 2.4 would not be achieved. They scratched their heads for a week or two before commenting. They told me that the critical ratio could be as low as 1.8 in some cases. A month or two later they issued a white paper letter that the anti-surge valve was not needed. I wish I had keep a copy of that letter.

Thanks, Ron

[turboron]

Chris, I just finished reading this thread completely. It is great for the DIY community that you are taking the time to publish this information. I will look forward to many posts in the future.

On a personal note, I got a big kick out of your teenage discovery of the world of rotating machinery and gas turbines. I think many of us have had the happy experience of finding we love gas dynamics at an early age. Beats golf.

Thanks, Ron

[CH3NO2]

^+1 What Ron said.

Thank you for a very useful thread Chris. Because of this thread I have the Lefebver book and with Chris's guidance have been able to plug many of the equations into a spread sheet to make a tool for optimizing many of the essential/critical parameters of a turbine combustion system.

This thread is an excellent kick starter, a resource, and should be "Stickied".

Tony

[madpatty]

I agree ☝️ with Tony too.
The thread is awesome for anyone looking a crash course on gas turbine combustion.

Cheers.

[finiteparts]

I finally was able to find the correct ignitor cable with a T5 style (button) and a T3 style (pin) end so that I could test out a few of the used ignitor boxes that I have.

The first is a Simmonds High Energy Pulsed Arc  Igntor box lighting up a Champion CH31815...man it is energetic!  That is a really nice plasma plume!  It is a nice test unit for stuff around the garage, since it is a 115 VAC powered box, I can just plug it in and pow!!!, it is kicking like a mule!



Another shot of the same setup...



The second box is also a 115 VAC...there is a noticable difference in the amount of energy being deposited across the ignitor gap.



I thought everyone would enjoy seeing these photos.

Chris

[finiteparts]

John,

In reference to this (I thought it better to move this discussion to here)...

jetandturbineowners.proboards.com/post/28276

There is nothing fundamentally "wrong" about using a tangential entry to the combustor casing. I think you are confusing direction of entry with entry velocity.

Usually the problem is that the entry tube is too small with a high entry velocity and thus the change in dynamic head due to the free vortex flow causes a reduction in the local available driving pressure through the liner holes. This reduction in local driving pressure reduces the kinetic energy of the jet which can translate into lower mixing energy in the PZ due to the jets reduced penetration angle (ideally we would want 90 degrees to liner surface).

I have actually been working through a more tangential entry can style combustor design. With the tangential entry velocity down below 120 ft/s, I can keep the variation in dynamic head from the outer case to liner streamlines down below 0.15 psia and thus the change in PZ Cd is only from 0.59 to 0.577. Which is not bad considering the Cd only comes up to peak of around 0.62 for the mass ratio I have in the PZ.

The idea that there is not that much aerodynamic diffusion left to do at combustor entry velocities (200-250 ft/s for the turbos that I have evaluated at upper design point) might be a point of argument. If we didn't try to recover the velocity any further than what is flowing out of the compressor scroll, we will consume an additional 1.2 to 1.9% of the combustor system pressure loss as compared to a design with a near 100 ft/s flow. Even at these "low" flows, if you try to decelerate it too quickly you are going to get large flow separations and no pressure recovery...i.e., essentially no benefit for adding the "diffuser".

- Chris

[racket]

Hi Chris

Yep , I agree , if done correctly with the air slowed down and suitably spread out, but guys never do that, therefore its better to try and discourage their use , its less work to construct a straight in flow with a bit of a funnel to distribute air than attempt to solve all the tangential variables

The main reason I don't like seeing tangential entry is because of the many hours I've spent at this keyboard trying to help guys get their tangentially equiped engines to run, I'm not concerned about the modest pressure losses etc etc, they are "insignificant" when compared with the major problems being confronted.

Ideally we need those nice "flowing" delivery tubes from the Allison 250 engines , but a tad hard for us backyard constructors to produce .

Cheers
John

[spartablades]

hi i m new here ..i m start to build that project ..a have a turbo with intake ( inducer diameter 7,4 cm)
i want start to fix tha flame tube ... is there somthing like calculator to have all dimensions about flame tube ... about combustion camber ..the holes around the flame tube? i found a calculate to internet but i want to be sure about that project thank you
my facebook: www.facebook.com/sakelaropoulos
my youtube channel: www.youtube.com/channel/UCDxCdaTpb_FUvczoCGi5QMg?view_as=subscriber

please i want directly to your advice .. thaankkkkk youuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu

[britishrocket]

I have spent the last few days reading through this thread. I have found it highly informative and interesting, despite the fact that it goes into waffle mode on a number of occasions.

It was worth reading for the scholarly treatment of this subject by the main contributor. I also feel that it is worth going through just for nuggets of information like the Cd value of a vane type swirler, or the high quality references that are cited.

It would be nice to see the OP continue and detail the final design of his projected combustion chamber. I for one would find this edifying.

[CH3NO2]

^+1

That's putting it mildly and it's not just this thread.

Go to Chris's profile and follow the links to his other threads and posts. There you will find a mother load of gas turbine engineering education.

The best is when John and Chris have a technical disagreement. With each from very different experience bases the informational exchange rate goes up exponentially. X2

And everybody reading wins.

[CH3NO2]

Chris and John,

What are your thoughts on Trapped Vortex Combustors(TVC)?
Pros/Cons?
Other than running lots of CFD, is there a methodology for their design?

Thanks,
Tony


For those who are not already familiar them, the TVC is a combustor design where cavities are used to trap a vortex flow structure.
www.google.com/search?q=trapped+vortex+combustor&tbm=isch&ved=2ahUKEwi2wdyI8IXoAhUnQ1MKHQR0C2UQ2-cCegQIABAA&oq=trapped+vortex+combustor&gs_l=img.3..0.114543631.114550486..114550994...0.0..2.142.2873.33j3......0....1..gws-wiz-img.....0..0i67j0i131j0i5i30j0i8i30j0i24.WpCAfSfoD_k&ei=gUViXvb8LaeGzQKE6K2oBg

They also go by the name of Ultra Compact Combustor.
www.google.com/search?q=ultra+compact+combustor&tbm=isch&ved=2ahUKEwiU5P7nmonoAhWWKlMKHVzXDjAQ2-cCegQIABAA&oq=ultra+compact+combustor&gs_l=img.3...1961506.1965213..1966308...0.0..0.91.1000.13......0....1..gws-wiz-img.kcY0Mi3sLLU&ei=-wRkXtT6ApbVzALcrruAAw#imgrc=tepc41xjl9IjuM

It was invented in the 1980’s and is generally considered an advanced combustor design for turbines, ramjets and scramjets.
pure.tue.nl/ws/portalfiles/portal/94575839/1_s2.0_S0360128517300898_main.pdf
www.sciencedirect.com/science/article/pii/S0360128517300898
netl.doe.gov/sites/default/files/gas-turbine-handbook/3-2-1-4-1.pdf

TVC’s reportedly can operate over a wide A/F range without flame out because of the high degree of recirculation of reacting combustion products. Its not that the recirculation zone needs to be large, it needs to have a high degree of recirculation. The combustor primary zone runs very hot with a large amount of recirculating flames, radicals and disassociated combustion products. Lean or rich, the fuel is forced to react even if it is outside its normal A/F limit.
Many publications discuss the advantages of compactness, a stable vortex locked into its cavity regardless of inlet flow velocity, they can work with very little pressure drop and they can operate over a wide A/F range without blowout.

[racket]

Hi Tony

LOL.......I have enough problems with a conventional flametube , it appears they're like our dump afterburners where combustion is conducted in the vortex produced by the dump step , I guess the similarity is there because the dump A/B was a "child" from ramjet combustors , the TVC might be a tad more difficult to construct and fit into our applications .

Cheers
John

[finiteparts]

It is good to hear that the information that I have provided is useful. I will try to find some time to work on expanding this post further.

As for the trapped vortex combustor, I can say that I have had a fair bit of exposure to this concept and also had the great fortune to have discussed these with some of the early designers that worked on these concepts at the AFRL and other places. Unfortunately, this was all related to proprietary work and thus any knowledge that I have due to this is not to be disclosed to the public.

So, I will give you my personal opinion. There is a potential reward of reduced size, but the risk to this reward is that all the design knowledge of standard combustors is out the door. You will be on your own in developing the correct sizing, air spits, local FAR distributions, etc. and like any other combustor, when you get it wrong, you will be fraught with ignition, stability and operability problems. The design seems relatively simple in the papers, but be warned, the devil is definitely in the details.

Secondly, the thermal design of our can or annular combustors are not too challenging, but when you have a high surface velocity scrubbing the walls with a reacting flows, the liner cooling issues will be more challenging, likely to the point of creating high nickel emissions out the tailpipe.

I left out the stability of the combustor because I feel that a well designed swirl stabilized combustor is just as stable for our purposes.

Now, I try not to ever be the guy that tells anyone that it can't be done. So what I will say is that it will be challenging and the likelihood of success for the homebuilder is greatly reduced compared to a more standard can or annular arrangement. For "coolness factor" or bragging rights, it would be a neat thing to see running, but I am not sure how big the rewards would be for our style engines.

And of course the final point is that I will have to refrain from commenting due to my work related experience being completely proprietary knowledge.

Good luck,

Chris