enginehistory.org Forum Index enginehistory.org
Aircraft Engine Historical Society Members' Bulletin Board
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

J58 installation in A-12/SR-71 nacelle

 
Post new topic   Reply to topic    enginehistory.org Forum Index -> Technical Discussion
View previous topic :: View next topic  
Author Message
pkent



Joined: 26 Feb 2013
Posts: 5

PostPosted: Tue Nov 26, 2013 17:23    Post subject: J58 installation in A-12/SR-71 nacelle Reply with quote

Hi,
I have just read J Thomas Anderson's new addition to the site 'How supersonic inlets work'.
On page 25 he mentions
Quote:
"suck-in doors open at low speeds to maintain small pressure loads on the nacelle skin".

Question 1. Can anyone point me to a photo showing these doors?


Quote:
to maintain small pressure loads on the nacelle skin

Question 2. Why was it necessary for small pressure loads?

I'd be grateful if anyone can help with answers.
Thanks.
Pete
Back to top
View user's profile Send private message
kmccutcheon



Joined: 13 Jul 2003
Posts: 191
Location: Huntsville, Alabama USA

PostPosted: Thu Nov 28, 2013 08:11    Post subject: Reply with quote

Good questoin, Pete. I'll take a stab at it...

Question 1. Can anyone point me to a photo showing these doors?
It's like trying to find a picture of a Unicorn, isn't it? SR-71s in museums are usually roped off, making it hard to see features underneath. The suck-in doors conform to the skin lines so closely that they are almost impossible to see from a distance. Most of the images you can find (please see
http://i47.photobucket.com/albums/f195/Neptune48/SR-71/_SCN0132_Adj.jpg) are close-ups with little context. To further confuse the issue, some diagrams (please see
http://www.meriweather.com/flightdeck/cutaways/sr71a.html item 59
and http://www.enginehistory.org/BBimages/J58Airflow.jpg)
imply that there may be suck-in doors topside, but I have never seen any.

There are, however a few images in the AEHS web site Members' Section that may be helpful:
http://www.enginehistory.org/BBimages/Suck-in01.jpg Warning, 2MB Image!
http://www.enginehistory.org/BBimages/Suck-in02.jpg Warning, 2MB Image!
http://www.enginehistory.org/BBimages/SR-71Dwg.jpg

Of the second image, Tom Anderson writes,
"This photo is a view looking aft at the open nacelle; the aft set of doors are the blow in doors that augment the airflow through the nozzle throughout the transonic region. I would guess that the forward two sets of holes are the suck in doors to relieve negative nacelle pressure during ground and low speed operation. The forward louvers, visible on the nacelle forward of the open engine bay, cover the exits of the inlet struts which flow centerbody bleed air."

Although unlabeled, the SR-71 drawing is also helpful, showing no topside features that could be the suck-in doors, but also showing these two rows of features on the underside.

Question 2. Why was it necessary for small pressure loads?
Please refer to the first drawing in Tom Anderson's Figure 24. During ground operations, before there is any pressure recovery from the inlet, the J58 compressor is consuming more air than the small inlet area can supply. The ejector also consumes some air, and the combination would result in a net negative pressure inside the nacelle. You can see in the diagram that air is flowing in via the centerbody bleed and the forward bypass doors. (Some diagrams, such as the J58 diagram above, actually show airflow through the suck-in doors moving both forward to the compressor and aft to the ejector.) The nacelle was designed to be pressurized from within, resulting in a pressure vessel with most of the force being resolved via hoop stress in the thin skin, and a relatively lightweight structure. Reversing this process puts a crushing force on the nacelle, which according to Tom Anderson, amounts to about 0.5 ATM or about 7 psi. This would be enough to crush the structure if the suck-in doors did not relieve the pressure. It is analogous to a toy balloon, which is only relatively rigid when inflated. Once the inlet system starts moving through the air and recovering pressure, the suck-in doors close and all air consumed by and surrounding the engine arrives via the inlet.
_________________
Kimble D. McCutcheon
Back to top
View user's profile Send private message Visit poster's website
pkent



Joined: 26 Feb 2013
Posts: 5

PostPosted: Thu Nov 28, 2013 13:50    Post subject: Reply with quote

Thankyou both for all those details. Now I know what to look for I can see closed doors on a Duxford photo I took.

I have been thrown by these doors ever since reading Paul Crickmore's book 'Lockheed Blackbird' in which he basically says the requirement for the doors first became apparent with the first start attempts on the J58 in the full nacelle. He says the starts were unsuccessful until the temporary expedient of removing an access door...this led to the addition of 2 sets of suck in doors as well as the engine start bleed being vented to the nacelle.

Perhaps you can help me with this one Kimble?
Ref Peter Law's presentation:
Question 3. Is the nozzle underexpanded at cruise and if so what were the ramifications?
or is my following reasoning wrong.
The ejector flaps are fully open by M2.4 (P.Law). I am assuming the primary nozzle is wide open with full AB. Further increase in speed to M3.2 produces a progressively higher expansion ratio up to 31.2 (P.Law) but no increase in the area ratio.. primary nozzle throat/ejector flap exit.
Since the ejector flaps are pressure actuated they follow the ambient pressure until they run out of travel at M2.4. From then on the jet is underexpanded.
Pete
Back to top
View user's profile Send private message
pkent



Joined: 26 Feb 2013
Posts: 5

PostPosted: Sat Nov 30, 2013 15:27    Post subject: Reply with quote

Quote:
Question 3. Is the nozzle underexpanded at cruise


I should have mentioned I have Paul Herrick's 'J58/YF-12 Ejector Nozzle Performance'. I have just re-read it and it says a shorter nozzle which was tested suffered performance losses due to being underexpanded which intimates that the actual nozzle was not.

However, I'm still curious as to the 'why' since the area ratio seems just too low for M3+ expansion. (The fully open primary is about 11 sqft, 45"dia I measured, and I believe the open final nozzle is 26.7 sqft, 70"dia)

Pete
Back to top
View user's profile Send private message
kmccutcheon



Joined: 13 Jul 2003
Posts: 191
Location: Huntsville, Alabama USA

PostPosted: Mon Dec 02, 2013 14:15    Post subject: Reply with quote

Question 3. Is the nozzle underexpanded at cruise and if so what were the ramifications?

I have conferred with both Pete Law and Tom Anderson, who point out that the propulsion system, with its design point of M3.2 in a standard atmosphere at 85,000 ft, would have the ejector flaps at maximum opening. This opening size was designed to provide a nominal expansion ratio for those conditions. The M2.4 in the presentation was a typo, which has been corrected.
However, during our discussion it became clear that the ejector behavior depends on a number of factors, and these may affect the point at which it reaches maximum opening. For example, in very cold conditions the flaps achieve maximum opening below M3.2 and are thus slightly underexpanded at M3.2. Similar sub-optimal behavior occurs when any off-design condition is encountered.
_________________
Kimble D. McCutcheon
Back to top
View user's profile Send private message Visit poster's website
pkent



Joined: 26 Feb 2013
Posts: 5

PostPosted: Wed Dec 04, 2013 20:53    Post subject: Reply with quote

Thank you for pointing out the effects of off-design conditions. I think I see this now in a plot of flight data which shows the ejector exit diameter can vary anywhere between 90 and 100% open above M2.4 (fig 5 David Campbell's "F-12 Series Aircraft Proplulsion System Performance and Development".
http://arc.aiaa.org/doi/abs/10.2514/3.60402?journalCode=ja

Ref Tom Anderson's Fig26 which shows a progressively diminishing thrust contribution from the engine (17% at cruise) David Campbell in the above paper (same fig5 para) says that, from this max afterburning cruise condition, if the AB is reduced to min AB the engine would actually be dragging on the engine mounts at high Mach numbers.
Question 1. How does this negative thrust contribution from the engine arise?

Ref Tom Anderson's fig25, at the design point it shows the engine flow is about 88% and secondary/leakage flow is 12%.
Question 2. What proportion of the 12% is ejector nozzle secondary flow?

Ref fig26 again, and the intake thrust contribution.
Bob Abernethy on page 4 "More Never Told Tales of Pratt and Whitney"
http://www.docstoc.com/docs/141090918/Never-Told-Tales-of-P_W3
says the 22% airflow increase through the engine (due to comp bypass bleed) gave his expected 19% net thrust increase but he did not anticipate the 47% installed thrust increase, and he says "the increased airflow really helped Kelly's inlet performance".
Question 3. How does 22% engine flow increase produce a disproportionate benefit in the inlet?

Ref Pete Law's ECS description.
Question 4. What is the design point compressor delivery mass flow from which the ECS 40lb/min is taken?

Pete
Back to top
View user's profile Send private message
kmccutcheon



Joined: 13 Jul 2003
Posts: 191
Location: Huntsville, Alabama USA

PostPosted: Tue Dec 10, 2013 06:17    Post subject: Reply with quote

Regarding your 4 Dec 2013 Post:
Quote:
Question 1. How does this negative thrust contribution from the engine arise?

As I understand it, engine thrust was never negative. This phenomenon must have been a transient condition where pressure in the ejector acting upon the closing engine nozzle created a force that momentarily exceeded engine thrust, until the airplane slowed down.
Tom Anderson - These calculations are of the momentums of the flows in the nacelle at various points. It's fun to say that the inlet is what is really pulling the airplane around at cruise, but the reality of it is that nobody would be going anywhere without that wonderful engine doing the work.

Quote:
Question 2. What proportion of the 12% is ejector nozzle secondary flow?

Tom Anderson - The nacelle secondary flow at cruise is inlet shock trap, cowl, bleed air flowing to the ejector nozzle. This flow is about 7% of the 12% bleed and leakage flow shown on Figure 25.

Quote:
Question 3. How does 22% engine flow increase produce a disproportionate benefit in the inlet?

Tom Anderson - A supersonic inlet has a certain amount of flow. See the top line in Figure 25 for the SR inlet. Typical engines fall off in airflow at high supersonic speeds and the resulting flow mismatch with the inlet must be bypassed at the cost of a large amount of drag. The J58 flow at cruise does not fall off, which allows a good inlet match at both high speed and transonic speed. This is a major benefit to installed thrust.

Quote:
Question 4. What is the design point compressor delivery mass flow from which the ECS 40 lb/min is taken?

Pete Law - We maintained the ECS engine bleed flow between 40 and 43 pounds per minute at all conditions, from idle on the ground to max speed at altitude. The engine airflow per engine is very high, at over 300 pounds per second during takeoff and around 170 pounds per second at cruise (i.e., over 10,200 pounds per minute at cruise). Therefore, the 40 pounds per minute of engine bleed for the ECS is insignificant to engine performance.
_________________
Kimble D. McCutcheon
Back to top
View user's profile Send private message Visit poster's website
pkent



Joined: 26 Feb 2013
Posts: 5

PostPosted: Sun Jan 05, 2014 13:53    Post subject: design point thrust setting Reply with quote

At the design point what was the AB primary nozzle area, presumably partially closed at less than 100%?
There must have been some excess thrust to get there and no longer needed for cruise at M3.2 on a standard day?

Thanks
Pete
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    enginehistory.org Forum Index -> Technical Discussion All times are GMT - 6 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group