Big Test Lady FWE arrives in Aero/Combustion Laboratory

Moderator: Mike Everman

milisavljevic
Posts: 855
Joined: Sat Oct 04, 2003 9:36 am
Antipspambot question: 125

The plot thickens

Post by milisavljevic » Thu Apr 30, 2009 12:43 pm

.
Having gone through so much grief modeling and remodeling the instrumented FWE,
why not tempt more pain by posting some of its predicted performance data? Yaaay!

Greg cannot measure thrust in the test cell; however, I will provide predicted thrust.

Modeling thrust is complicated in the FWE type by its characteristic combustor cone
and how the intake transition is simply plonked down on top of the combustor. That
means I get a range based on the effective combustion zone length: minimum rated
thrust is found near where the leading edge first breaches the combustor wall; most
probable maximum thrust is found part way across the opening (deeper in the cone).

[1] minimum rated thrust . : 50 N (5.1 kg; 11.2 lbf) (my mula is on/near this value)
[2] probable maximum .... : 66 N (6.7 kg; 14.8 lbf)
[3] absolute maximum .... : 97 N (9.9 kg; 21.8 lbf) (don't expect to see this value!)

With respect to the acoustic data, I will provide the closed pipe fundamental and its
equivalent acoustic temperature (not that it matters!). As the test cell will certainly
be warmer than the model default of 15 C (59 F), expect any measured frequencies
to be shifted slightly higher (I will give the range, below). Just for fun, I include the
the open pipe resonance: this may be captured in the data (hint: look near 157 Hz).*

[4] closed pipe fundamental : 130-137 Hz (noted above: expect mid-130s to 141 Hz)
[5] acoustic temperature ... : 634-644 K

Note: the acoustic data assumes an 82.6 mm flare and 1595 mm overall duct length.

Good luck to Greg today and tomorrow. Let's all hope data collection goes smoothly.

Cheers,
M.

*range: 156-157 Hz (expect 158-161 Hz; also look for a surprise near 1030-1040 Hz)
no safe haven for merchant scum


for ye merchants who do the prop'r t'ing only if
ye be haul'd-up on charges b'fore ye ship-mates
an' threat'nd wit' forfeiture of all ye precious loot
hear this - so-called stand-up guys YE BE NOT

avast!
Cap'n M.

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

Re: The double-null

Post by larry cottrill » Thu Apr 30, 2009 1:51 pm

milisavljevic wrote:.
Dang911 wrote: Do we even have to call it a flare? Why not decorative trim.
"Decorative trim", it is.
Actually, it would be better to call it by its classic name: the "slight tailpipe flare". The slight tailpipe flare is a wonderful feature. It solves the Goldilocks Problem: It is "just right". Small enough that its acoustic impact is trivial; yet, large enough that it eases, and thus enhances, the formation of the cold air piston in the tailpipe. This has been verified by Dynajet Jerry who testified that in the early days of Dynajet development, they were greatly surprised at the positive performance effect it provided. Bill Pardue has clarified that this positive effect only occurs in static running; the slight tailpipe flare becomes a negative performance element at moderately high speeds. But no matter, if we suddenly achieve moderately high speeds we will have a lot bigger problem to deal with than whether we keep the flare or not.

Greg, please see the attached, and at some point try to verify all "as built" dimensions, one last time. Once you let me know of any corrections, I will edit them in and change the red-boxed legend to indicate that this is the plan of what actually ran. Thanks!

L Cottrill
Attachments
Big_test_lady_basic_as_built_dims.png
As Built Dimensions drawing (NOT FINAL at this time) - Drawing Copyright 2009 Larry Cottrill

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

Re: The plot thickens

Post by larry cottrill » Thu Apr 30, 2009 4:46 pm

milisavljevic wrote:.Greg cannot measure thrust in the test cell; however, I will provide predicted thrust.

Modeling thrust is complicated in the FWE type by its characteristic combustor cone
and how the intake transition is simply plonked down on top of the combustor. That
means I get a range based on the effective combustion zone length: minimum rated
thrust is found near where the leading edge first breaches the combustor wall; most
probable maximum thrust is found part way across the opening (deeper in the cone).

[1] minimum rated thrust . : 50 N (5.1 kg; 11.2 lbf) (my mula is on/near this value)
[2] probable maximum .... : 66 N (6.7 kg; 14.8 lbf)
[3] absolute maximum .... : 97 N (9.9 kg; 21.8 lbf) (don't expect to see this value!)
Well, that's not bad, considering it's just a straight tailpipe engine. Greg might remember that I gave him an initial guess of "10 to 12 lb, maybe 15".

L Cottrill

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

Notes on Sensor Tap Loacations

Post by larry cottrill » Thu Apr 30, 2009 5:03 pm

I originally set up the sensor taps at round-number percentages of the chamber cone length and intake pipe physical length: 0%, 10%, 20% and so on. Obviously, there was nothing magical about these stations; they were just chosen as positions that were easy to comprehend when we're talking about the data taken there. Of course, it wasn't possible to weld them on at EXACT locations; each could be off by as much as 1mm either way. The chamber hasn't changed, so for the chamber taps, the percentages still stand at those 'round' values. The intake length has been radically reduced though, such that now the approximate locations will be:
- 0 % of intake pipe length
- 14.2 %
- 28.5 %
- 42.7 %
- 57.0 %
- 71.2 %
- 85.5 % (if this one wasn't sacrificed to shorten the pipe)

The points on the intake are at 24.5 (NOT 25.4!) mm intervals, based on the original design length. This information may be utterly meaningless, of course. But, the occasion might arise to talk about "the 14 percent intake point" or some such when we start talking about the data. What I hoped in originally locating them is that it would make it easy to visualize the exact region being discussed.

L Cottrill

Dang911
Posts: 395
Joined: Sat Jun 11, 2005 6:03 pm
Antipspambot question: 0
Location: USA-Florida, Georgia, Ohio
Contact:

Re: Big Test Lady FWE arrives in Aero/Combustion Laboratory

Post by Dang911 » Thu Apr 30, 2009 11:23 pm

Really short video of the Big Test Lady running. I have began data collection but I have more to do tomorrow.
Attachments

[The extension avi has been deactivated and can no longer be displayed.]

Louder is ALWAYS Better!!!

milisavljevic
Posts: 855
Joined: Sat Oct 04, 2003 9:36 am
Antipspambot question: 125

Re: The plot thickens

Post by milisavljevic » Fri May 01, 2009 1:50 am

.
milisavljevic wrote: Having gone through so much grief modeling and remodeling the instrumented FWE,
why not tempt more pain by posting some of its predicted performance data? Yaaay!
Greg, this motor is, to quote myself, "running like a piece of crap". You may ignore my request
to either (1) finish pounding out the flare to ID 96 mm or (2) hack it off completely and replace
the missing piece to restore an overall duct length of 1600 mm, but either improves this motor.

As is, this motor's spectrum has only two (2) peaks: the fundamental (115.7 Hz) and its second
harmonic (228.8 Hz). All other harmonics are rather suppressed; many are "buried in the grass".

It is not behaving at all as a closed pipe resonator; it is behaving as a quasi-synchronised duct.
milisavljevic wrote: [4] closed pipe fundamental : 130-137 Hz (noted above: expect mid-130s to 141 Hz)
[5] acoustic temperature ... : 634-644 K

Note: the acoustic data assumes an 82.6 mm flare and 1595 mm overall duct length.
The "final changes" I sent to you and Larry had specified the inlet flare at 58 mm, exit flare of
96 mm and overall length of 1600 mm. The model synchronised at a lock up of 7:1 (416 K) and
with an upper Helmholtz frequency of 131.3 Hz. As it is, the exit flare is 82.6 mm, duct length
was estimated as 1590-1595 mm, and the inlet flare is 57 mm. This model synchronised at 7:1
(406 K) and with an upper Helmholtz frequency of 116.7 Hz; can you see where I am going now?

Past experience with FWE type ducts teaches that these typically behave as closed pipes. This
sent me back into the model, to look for a thermoacoustic mode more favourable to this state;
that is where this morning's acoustic data came from. It seems this duct prefers to be in-sync.

Full synchronisation is precluded by the under-sized exit flare. Despite Larry's assertion that a
"slight tailpipe flare" has but a trivial acoustic impact, the evidence is unequivocal: flares rule!

Destructive interference inside the duct, between the out-of-sync open and closed pipe modes,
is why the acoustic spectrum is basically thrashed. As pulsejets are thermoacoustic devices, a
reasonable (and correct) conclusion is that the thermodynamic qualities of this duct are f---ed.

The best way to fix this situation is to complete the flare. You and Larry should decide on this.

M.
no safe haven for merchant scum


for ye merchants who do the prop'r t'ing only if
ye be haul'd-up on charges b'fore ye ship-mates
an' threat'nd wit' forfeiture of all ye precious loot
hear this - so-called stand-up guys YE BE NOT

avast!
Cap'n M.

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

Goldilocks Found Poisoned!

Post by larry cottrill » Fri May 01, 2009 2:21 am

Well, NUTS!

M, the decision would be easy if Greg had just one thing -- time. But tomorrow's his last day on this work site. After that, he'll be out of there and this lab will soon be back to having real work to do, for real manufacturers. Not only would he need time to do the mod, he would undoubtedly need to re-run every test setup he tried today (I haven't heard from him privately, so I don't know how much effort this entails). Otherwise, how could the existing data be meaningfully related to data from the optimized motor?

Greg, it's up to you. Only you know what you've done, what would need to be re-done, and whether you have time for the mods described. Obviously (I think), the quicker modification would be to saw off the little flare and weld on some pipe, which you should have plenty of. What do you think?

L Cottrill

Dang911
Posts: 395
Joined: Sat Jun 11, 2005 6:03 pm
Antipspambot question: 0
Location: USA-Florida, Georgia, Ohio
Contact:

Re: Big Test Lady FWE arrives in Aero/Combustion Laboratory

Post by Dang911 » Fri May 01, 2009 3:15 am

It hurts to say, especially when what’s "right" may not be what’s "best", that I think I am going to leave it alone. I will continue to get test data that will be valid with today's data. Time is a killer and can't be slowed down or gotten back. With that being said, my regular pressure transducers with a slow refresh rate don't seem like they are going to be very useful besides acting as redundant barometers. They aren't reading pressure more than a few fractions of a PSI higher than atmosphere. I have yet to try the dynamic but that will be my first test tomorrow set to read at 1000 Hz, I have very high confidence in these. The TC's are working although I haven't taken a close look at the data. Temps seem to be around 2200 F; I know who would have thought just by looking at the metal (DUHHH...) Anyway that's all I have for you now. I will disclose all of the data once I get some more testing, hopefully it can be found useful.
Louder is ALWAYS Better!!!

milisavljevic
Posts: 855
Joined: Sat Oct 04, 2003 9:36 am
Antipspambot question: 125

Re: The plot thickens

Post by milisavljevic » Fri May 01, 2009 3:50 am

.
Dang911 wrote: It hurts to say, especially when what’s "right" may not be what’s "best", that I think I am going to leave it alone.
I spoke with Greg this evening, and sad to say, I have to agree with him here. Time has past us by.
At least this motor self-sustains, and as Greg says, "...(bad) data is better than no data". We tried.

Looking for the bright side in all of this, Greg may get another shot at testing a pulsejet, this fall...

M.
no safe haven for merchant scum


for ye merchants who do the prop'r t'ing only if
ye be haul'd-up on charges b'fore ye ship-mates
an' threat'nd wit' forfeiture of all ye precious loot
hear this - so-called stand-up guys YE BE NOT

avast!
Cap'n M.

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

Re: Big Test Lady FWE arrives in Aero/Combustion Laboratory

Post by larry cottrill » Fri May 01, 2009 6:56 pm

Dang911 wrote:With that being said, my regular pressure transducers with a slow refresh rate don't seem like they are going to be very useful besides acting as redundant barometers. They aren't reading pressure more than a few fractions of a PSI higher than atmosphere.
That's what I was predicting back in our earliest discussions (before you explained the dynamic sensors). Anything that reads only static pressure (no velocity component) and then averages over significant time frames is not going to show much. Back when we started burning up stainless steel tubing in the back of the Dynajet, what I was trying to show is that there's someplace in the engine where an ordinary gauge would show the thrust "pressure" -- and there was! Not at the exit end of the tailpipe (where the gauge read 0) but at the end of the nozzle zone, the "throat" of the engine, where the gauge pressure showed at about 4.5 PSI just before the SS tube started to be destroyed. That, of course, was a combination of static and velocity pressure, averaged by the inherent sluggishness of the gauge (i.e. mechanical inertia).
I have yet to try the dynamic but that will be my first test tomorrow set to read at 1000 Hz, I have very high confidence in these.
Yes, I'll bet the output from these will be wonderful to see! Your engineer friends might even be amazed at the amplitudes of the pressure swings.
Temps seem to be around 2200 F; I know who would have thought just by looking at the metal (DUHHH...)
The video gives the impression that the front dome is running MUCH hotter than the chamber cone. Does naked-eye observation bear this out? What can you say about the heating pattern of the chamber? How do you think the chamber material is holding up overall?

L Cottrill

Dang911
Posts: 395
Joined: Sat Jun 11, 2005 6:03 pm
Antipspambot question: 0
Location: USA-Florida, Georgia, Ohio
Contact:

Re: Big Test Lady FWE arrives in Aero/Combustion Laboratory

Post by Dang911 » Sat May 02, 2009 12:33 am

SO I thought I was going to have more time to run today but as it turns out I had a lot of loose ends which needed to be tied. The engine will remain in the lab and another engineer said he will most likely get me some more data in the future. I will be back to work in 3-4 months where I could pick up things. Believe me I am as bummed as much as everyone who was anticipating the data and I feel real bad for not being able to provide any at this moment. I will get some very limited data up in the next few days but I am now in the process of moving.

I can answer some of Larry's questions...
larry cottrill wrote:The video gives the impression that the front dome is running MUCH hotter than the chamber cone. Does naked-eye observation bear this out? What can you say about the heating pattern of the chamber? How do you think the chamber material is holding up overall?
I would have to say that's a camera trick I noticed them heating pretty much at the same rate. What another engineer said he noticed was the left side of the chamber heated faster than the right side, I can't confirm or deny this. As far as the material its pretty flaky, maybe even worse than carbon steel, but this is a big engine.

I do have to say that I have never seen such a beastly flame shooting from the intake, and its HOT, hotter than the exhaust (at least that's what my hand told me). Regardless its hotter than say my intake flame on my lockwood.
Louder is ALWAYS Better!!!

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

Re: Big Test Lady FWE arrives in Aero/Combustion Laboratory

Post by larry cottrill » Sat May 02, 2009 2:14 pm

Dang911 wrote:SO I thought I was going to have more time to run today but as it turns out I had a lot of loose ends which needed to be tied. The engine will remain in the lab and another engineer said he will most likely get me some more data in the future. I will be back to work in 3-4 months where I could pick up things. Believe me I am as bummed as much as everyone who was anticipating the data and I feel real bad for not being able to provide any at this moment. I will get some very limited data up in the next few days but I am now in the process of moving.
Greg -

The fault is mine, not yours. My initial rushed and faulty design cost at least a full day of effort for you, and probably a good day for M, too. It would have been so much better to start with a scale-up of a proven design -- that could have easily been adjusted for the tubing sizes we had to work with. It looks to me like the somewhat "larger than scale" intake ID made this design too "open" at the front end (recall my comment about the difficulty of getting a reasonable impedance from the intake) -- adjustments to intake diameter create HUGE changes in acoustics. That intake size would have been more appropriate for something like a scaled-up Lady Anne. I will modify the drawings I've posted here to try to not lead future readers down a blind alley.

Again, how can we thank M enough for his effort? He completely salvaged the project, saving you many hours of futility and giving you a running engine. If you'd had time to modify it completely to his specs, it would have been a darn good one, at least for a straight tailpipe motor. I can never re-pay what he was willing to do. But M, if you'll PM me the name and address of your favorite pub, I'll buy you a pint of whatever's your favorite. And that's not a bad offer from a Christian boy; you might as well take me up on it ;-)

Now, it's time to make sense of where we go from here. Here's what I think we should do: Go ahead and do whatever you want with the data you've got, but DON'T use it as a reference for the future. Instead, get the engine optimized and be ready to "bite the bullet" and start over. You're not going to gain anything more from the static pressure sensors, so in the future concentrate on the dynamics alone. Apparently only two are available, which means you'll have to do a lot of short runs with frequent changes to the setup (i.e. moving the sensor tubes around between ports). In the meantime, if M will give us the tail flare dimensions cast AS A SHORT CONE, I can create the cone out of the chrome-moly sheet (I have LOTS left!) and get it perfectly round so your friend at the lab can make the final cut and simply weld on the cone. You will now have a near-optimal design that will deliver outstanding data via the dynamic sensors (and, of course, you can still use the thermocouples, too). While he's in a welding mood, he could add a few taps to the tailpipe, at whatever we think are strategic spots. Only after that, he (or you, later on) can pick up testing and get the kind of data we all want.

What do you think? Yeah, it is Plan C, but I think it will make the project "pay out" in terms of the effort expended.

L Cottrill

milisavljevic
Posts: 855
Joined: Sat Oct 04, 2003 9:36 am
Antipspambot question: 125

Re: The plot thickens

Post by milisavljevic » Sun May 03, 2009 5:30 am

Larry and Greg,
larry cottrill wrote: My initial rushed and faulty design cost at least a full day of effort for you, and probably a good day for M, too.
Delivered with mock-wrath and virtual shaken-fist: A good day that I will never get back. Not ever. :evil:
larry cottrill wrote: Again, how can we thank M enough for his effort?
As per the above line: You don't want to go there... Mwahahahahahahahahahahahahahahahaha!!! :twisted:
larry cottrill wrote: M, if you'll PM me the name and address of your favorite pub, I'll buy you a pint of whatever's your favorite.
Hmm. Me thinks it better if you send the money to either the ACLU, EFF, or PFAW, in your name. :lol:
larry cottrill wrote: If M will give us the tail flare dimensions cast AS A SHORT CONE...
Define: "short cone". To answer the follow-up question: Yes, I will do it. Tell me when you need it. :P

With Villainy,
Evil M.
no safe haven for merchant scum


for ye merchants who do the prop'r t'ing only if
ye be haul'd-up on charges b'fore ye ship-mates
an' threat'nd wit' forfeiture of all ye precious loot
hear this - so-called stand-up guys YE BE NOT

avast!
Cap'n M.

tufty
Posts: 846
Joined: Wed Dec 24, 2003 12:12 pm
Antipspambot question: 0
Location: France
Contact:

Re: The plot thickens

Post by tufty » Mon May 04, 2009 7:05 pm

milisavljevic wrote:Define: "short cone".
It lies between a "very short cone" and a "medium cone". And sometimes comes with a Cadbury's Flake in it.

Image

Mike Everman
Posts: 4933
Joined: Fri Oct 31, 2003 7:25 am
Antipspambot question: 0
Location: santa barbara, CA
Contact:

Re: Big Test Lady FWE arrives in Aero/Combustion Laboratory

Post by Mike Everman » Mon May 04, 2009 7:14 pm

There is a serious problem with that design. I am no expert, but I think it simply will not stand the heat!
Mike
__________________________
Follow my technical science blog at: http://mikeeverman.com/
Get alerts for the above on twitter at: http://twitter.com/mikeeverman

Post Reply