Lady Anne variation.

[Graham C. Williams]

Hi Graham,

hope you all got back ok after your break, pity about the weather but i hope you all enjoyed it nonetheless.

I've forwarded James's email from the GTBA

all the best

Nick



Thank you all so much for participating and helping on the GTBA stand at Bristol last weekend. For
those who were not able to participate for the full weekend, we had a really great time all
weekend. As I think most of you will have noted, the weather was variable but despite this and a
few technical problems we still managed to demonstrate engines throughout the full weekend. In fact
those present on Sunday will know we certainly made ourselves heard with a very impressive display
by Graham and his very powerful pulsejet. We managed two runs in quick succession before the
organisers approached and, with due apology, ask us kindly not to run it again due to concerns over
the noise - We therefore ran up Martyn's AMT Olympus to full power ( 50 lbs thrust ) as the
"quieter" option!

Thank you all again for your valuable input over the weekend. Unfortunately we were pipped at the
post this year on the "Best Club Stand Trophy" but let's see if we can win it back next year!
(Actually I think my wife was rather pleased, as it only means more silverware to polish thought
the year).

Just as a reminder, we have also been invited to participate as the Model Engineering Exhibition at
Ascot next month - It is another three day weekend - The dates are Friday 19th to Sunday 21st
September - Set-up will be on Thursday 18th Sept.

A number of you indicated to me that you would be interested to participate, so I would be grateful
of your response confirming you can support the GTBA stand and which days you can come along.
Again, any and all components and turbine related exhibits would be very much appreciated.

We have been asked to advise participation and model details and values as soon as possible, so it
would really assist me greatly if you could get back to me as soon as possible.

Looking forward to hearing from you.

Very best regards
James

[Graham C. Williams]

Dear M.
At some stage in this prolonged discussion you mention replacing the Divergent/Convergent section in the middle of the pipe with an ordinary taper. My laziness precludes me from finding your exact words. I know I ran the tests in the early modelling development of this motor but cannot find them now. Anyway NUDiS is now a lot better so I did some modelling over the Bank Holiday.
Replacing the (Type07 derived) section with a tapered cone of the same: Length, Start and End Dia. The motor runs but Thrust drops by about 2.22N (O.5 Lb or 0.23Kg). A little over a 10% drop.
I think this is due to the increased CC stagnation time this section creates within the design. I do not reject your earlier idea that this section of pipe may also be taking an active part in later combustion; there is certainly a pressure rise in this region but the local gas velocities appear too high in the model. I must also consider the limitations of 1D modelling.

Regards
Graham.

[milisavljevic]

Dear M.
At some stage in this prolonged discussion you mention replacing the Divergent/Convergent section in
the middle of the pipe with an ordinary taper. My laziness precludes me from finding your exact words.

Hello Graham --

Hah! My lazy-ness did not preclude from hunting that snippet down; however, I still owe you some definitions
and so will provide these first, and then respond to your, umm..."provocative" posting. Definitions to follow.

Cheers!
M.

[milisavljevic]

Hello Graham --

Okay! Here are some (rather late! ) definitions for what I call "step ratios" and "sigma scores"; just the basics.

Enjoy,
M.

I use three (3) step ratios when modeling pulsejets, two of which are simply measures of combustor confinement
and one that is not (and is misnamed?), but to paraphrase a song "it's my party and I'll name things (if I want to)".

S0 is used only for Ecrevisse, QD and RD type ducts, as characterised by a double step (aft of the combustor):

[1] S0 = [6 x sum of areas (of ports opening onto the combustor) / mean cross sectional area (of combustor)]

S1 is used in ALL duct layouts, ignoring the first step that characterises the Ecrevisse, QD and RD type ducts:

[2] S1 = [6 x sum of areas (of ports opening onto the combustor) / mean cross sectional area (of combustor)]

Multiplying the ratio by 6 is a historical artifact and is used only for my convenience. Obviously, a step ratio of 3
implies a confinement ratio of 50% (3 / 6 = 0.5). By definition step 0 values are always larger than step 1 values.

S2 is also referred to (by me) as the "forward pressure ratio" or alternatively, "forward bias". If we represent the
thrust developed across a port as a pressure (dimensions of force per unit area), then the following definition of

[3] S2 = [effective pressure rise (summed across all inlet planes) / effective pressure rise (across exit plane)]

makes sense. I usually calculate S2 for maximum thrust as there are constraints on the minimum and maximum
allowed values of this pressure ratio. S2 can be calculated for any thrust setting, and it is always a maximum at
maximum thrust. Without expanation, the upper and lower limits on the S2 ratio are approximately 2.5 and 1.0.


NEW! Definitions for the sigma scores.

As per S2, the sigma scores relate thrust to cross sectional area, so the following definitions are straightforward:

[4] σn = [sum of thrust developed (across all inlet planes) / sum of cross sectional areas (for all inlet planes)]

[5] σx = [sum of thrust developed (across exhaust planes) / sum of cross sectional areas (for exhaust planes)]

[6] σ∞ = [sum total of all thrust developed (in the motor) / sum of cross sectional areas (for all inlet planes)]

As you can see, the measurements described in the above definitions are easy to obtain. What they do not show
are the relationships between these metrics, especially σ∞, and real world jet engines. It appears that a ceiling
value of σ∞ exists in turbojets (and remembering that all turbofans use turbojet cores) of about 83 kilopascals.

What should totally boggle your mind, however, is the fact that this exact value of σ∞ also applies to pulsejets!

This means no self-sustaining, naturally aspirated pulsejet operating in sea level static conditions, can develop
a total thrust per unit of inlet area (when expressed in units of pressure) that exceeds approx. 83 kilopascals.

To put this in forum friendly units, we're talking about a ceiling of 12.1 psi. Turbojets, pulsejets, just the same.

Chew on that for a while!

Cheers!
M.

[Graham C. Williams]

Dear M.
"provocative" Yes but I hope not offensive!

Thank you.
How do S1 and S1 'prime' differ in your report?
I recall that 'QD and RD type ducts' refer to two of your earlier designs is that correct?

Regards
Graham

[milisavljevic]

.
Hello Graham --

"provocative" Yes but I hope not offensive!

Ha! Not offensive! But I have to wonder about my forthcoming reply (haha!). I will try and contain myself.

How do S1 and S1 'prime' differ in your report?

Only in how I calculated the mean combustor cross section, what with it being a conic section throughout.

I recall that 'QD and RD type ducts' refer to two of your earlier designs is that correct?

Correct. The QD and RD types were derived from the Ecrevisse; Bill Hinote built the first QD (QD-80/88)
way back in Spring 2004. Despite the gaps in my pulsejet knowledge (compared to now), it turns out the
QD-80/88 was almost perfect. I called it a "hole in one" back then -- and it was -- in more ways than one!

The most recent in the series is "mini-thor" (μT.05.1; not presented on forum), a QD built in Sweden by
("Fricke") in August 2007. The "hybrid" superchines I hope to enter in the forum challenge
will, if successful, combine the superior thrust performance of the QD/RD with the linear chinese layout.

That being said, wrt. thrust performance AND fuel efficiency, no design has surpassed the QD/RD types.
So, while these may be older designs (~5 years), they are by no means obsolete. You will see more soon.

(Now I must get those definitions done...)

Cheers,
M.

Edit: Graham, please check back with my "definitions" posting, above; it's now complete. Enjoy!

[Graham C. Williams]

Hmmm nice!
Please see your PM.

Regards
Graham.

[milisavljevic]

Dear Graham,

Thank you for the PMs. I think that I can solve this minor mystery for you, but first, please accept my apologies
for wasting so much of your time crunching numbers. At least you were well "lubricated" (note to the "merkuns",
to read "lubricated" as "fortified by lager" )! In future postings I will include full units, and avoid ambiguities.

I work with metric units, but unfortunately, many of the people I communicate with (eg., Nasa engineers -- and
of course so many forum members) are comfortable only with (useless) merkun units. When I posted my report
I converted most (if not all) the results into merkun units (for example, the sigma scores were reported in psi).

From the report:

Sigma scores (propulsive efficacy, %)

• sigma ∞ = 4.8 (39.6%)
  sigma n = 1.4 (36.7%)
  sigma x = 3.2 (44.2%)

First of all, the entry for "sigma x" is in error; this should read "sigma x = 0.7 (44.2%)".
The exact computed values for all entries are (in order presented) 4.84, 1.35 and 0.65.

If you note that your FWE VIII takes two (2) ID 20.5 mm intakes, then the value of "sigma ∞" should check out.

Hope this helps!

Cheers,
M.

[Mike Everman]

Hi M. and Graham,
Hey, M., what were the final performance numbers for QD80/88? I'm not sure thrust or SFC were ever reported. I can certainly go back and look...

[milisavljevic]

Hey, M., what were the final performance numbers for QD80/88?

Oops! We're sorry, but your query has failed to meet standards of topical relevance.
A non-refundable service charge of USD 5.50 has been deducted from your account.

Error code: 049 (reply sent off-forum)

Cheers,
M.

[Graham C. Williams]

Dear M.
The Constraints imposed by this challenge are infuriating. It is a relatively simple matter to increase trust by increasing volume but this is a trade off that offers no improvement to this measure of efficiency. It is of no help seeing other designs that by the alternative measures of efficiency are better and yet worse in terms of N/L!

Those things to one side for a moment. 5Lbs (22+N) of Thrust was a personal design point for this motor ( A usable thrust from a modestly sized 'straight' package). I'm sure that with a combination of: even higher construction standards and minor dimension changes, 23 N will be exceeded without a significant change in volume. even so this design will be hovering around 22-24N/L values.
What can I see from your report? Well-designed motors with 'Bustle' style exhausts can be made to perform significantly to the Sigma measures of efficiency but to my current knowledge these motors (I think, because of the large volume of the 'Bustle') are relatively poor in terms of N/L. By far the best performance (N/L), I know of, uses a straight exhaust pipe, a significant contributor to low volume, in a 'U' bent linear format. I recall Mike made a larger version of this motor. What was the N/L value for that, do we know?

Graham

[Mike Everman]

N/L for my pocket jet scale up was 18.6
I've determined that my scaling was silly, and there was a build error as well. I want to rebuild that one, but would have to go from scratch. It was the easiest starter of all, though.

[Graham C. Williams]

Dear Mike.
18.6 N/L though not close to the 33N/L of the smaller motor it's still no slouch.
Regards
Graham.

[Mike Everman]

Yeah, it wasn't too bad. 2.65" dia, 45.7" long, 6.5 lb thrust max. I still haven't had time to work on my contest motor, but it is between a triple of the pocket jet, or a scale up of it to reach the volume.
I just feel the triple is going to be too much hassle what with synching and unknowns. If it all worked 3x the pocket jet, though, it would be 10.5 lb max, and hoping for about 15 with augmentation. That would be something. Most likely poor SFC on this in any case, un-augmented.

[milisavljevic]

Dear M.
The Constraints imposed by this challenge are infuriating.

Ha! That's what makes it a "challenge". Otherwise, it would be a mere dog and pony show.

I prefer to answer with my two forthcoming entries, both of which sport "bustle" tailpipes,
and if these run to spec, will yield volume specific thrust densities (N/l) in the upper 30s.

[hint]...I think you need to look into testing an alternative combustor geometry...[/hint]

Cheers!
M.

PS: Yes, Mike (and Fredrik!), I will get the spec to you! No need to mention it here. What?