A-12/SR-71 astro-inertial navigation system

Y

yahya

ACCESS: Confidential
Joined
2 April 2020
Messages
177
Reaction score
95
The A-12 and the later SR-71 used the NAS-14v2 astro-inertial navigation system (ANS) from the 1960s, which was particularly interesting in the pre-GPS times. The set apparently looked like this:

https://airandspace.si.edu/collecti...rtial-navigation-system-ans/nasm_A20090019000

theaviationist.com

Up Close And Personal With The Blackbird: An Exclusive Interview with Former Director of Worldwide SR-71 Operations

Former Director of The Fastest, Most Secretive Reconnaissance Program Lt. Col. Szczepanik Talks to The Aviationist. Only 86 pilots and 86 Reconnaissance
theaviationist.com theaviationist.com

https://theaviationgeekclub.com/the...2-was-crucial-in-blackbird-mission-heres-why/

https://theaviationgeekclub.com/her...n-system-was-essential-for-blackbird-mission/

https://www.thedrive.com/the-war-zo...ning-future-fights-in-gps-denied-environments

I wonder what was the real life accuracy of such a system compared to the regular INS given the very high speeds and on-board vibration of the A-12/SR-71. Did the astro-navigation part simply correct the drift of the INS along with the signals from other platforms like TACAN to improve accuracy?
 
Coincidentally, I've just learned that Nortronics (Northrop Electronics) had "recycled" Skybolt astro-inertial system, for the SR-71.
They had a Convair C-131B flying out of Hawthorne: testing the system in 1962-63: making the transition from Skybolt to SR-71.

https://alopez128.tumblr.com%2Fpost%2F15406861921
View: https://www.tumblr.com/alopez128/15406861921/sr-71-blackbird-astroinertial-navigation-system




I find fascinating that some Skybolt survived into SR-71.
 
Last edited:
baigar said:
As fan of the Ferranti navigation systems (Video published 2013) I am pretty sure, they never made a star tracking device, although it probably would have been possible by "abusing" the LRMTS as that one offers most of the required components already:
- Gyro stabilized platform
- Optical mirror
- High accuracy readout.
But I am curious: Was there any UK made (or Eurpean) device like the NAS-12V2 for aircraft?
Click to expand...

After some search I found, that at least there where investigations in the UK on a possible star tracking device for the BLACK KNIGHT project with a report issued in 1959 (see below). According to different reports there have been investigations on how to make windows suitable for use in hypersonic flight regimes. These reports survived in the National Archives of the UK.

1695221959414.png
 
Archibald said:
Coincidentally, I've just learned that Nortronics (Northrop Electronics) had "recycled" Skybolt astro-inertial system, for the SR-71.
They had a Convair C-131B flying out of Hawthorne: testing the system in 1962-63: making the transition from Skybolt to SR-71.

https://alopez128.tumblr.com%2Fpost%2F15406861921
View: https://www.tumblr.com/alopez128/15406861921/sr-71-blackbird-astroinertial-navigation-system




I find fascinating that some Skybolt survived into SR-71.
Click to expand...
I think that same star tracker system migrated into the early SSBNs, they were doing automatic star tracking for navigation updates through the periscope!

I don't know if that system is still physically present in today's SSBNs, the Ship's Inertial Navigation System (SINS) has very little drift, still measures up well against most Ring Laser Gyro systems.
 
Before 1960, the UK obviously was thinking on development of a star tracking device as the Technical Memorandum No G.W.366 shows. In this (once secret, now unclassified) report, an estimation of requirements is given together with recommendations, which light sensor to use. Given the technology of this time available photomultiplier tubes and a 2" optics seemed to be suitable for development of a star tracking device based on following magnitude 5.3 stars (but I do not know of any real work into development):

G.W.366.gif
 
The UK started early with star trackers, what were then termed ‘Auto-Astro’ systems.

Theoretical development work at the Telecommunications Research Establishment on Blue Sapphire which would be capable of tracking stars at night at the Telecommunications Research Establishment began around mid-1945 and flight trials of a prototype unit in a Vickers Viking began in December 1949. Blue Sapphire was intended for the NBS system of the future V-Bombers. Orange Tartan was a successor system using infra-red homing guidance technology to operate in daylight and be suitable for installation in either V-Bombers or guided-missiles. Technical problems with both systems saw work being abandoned in the mid-1950s. There is some evidence that Avro researched similar equipment during 1956-61, probably for missiles.
It's likely that any star tracker for Black Knight built upon the Orange Tartan work.
 
From Paul Crickmore 'Beyond the Secret Missions':

"The Nortronics Division Astro Inertial Navigation System, used in the Skybolt air-to-surface missile, was modified for use in the SR-71 and designated as the NAS-14V2.
Producing a terminal error accuracy of less than half a mile after covering distances the equivalent of more than halfway around the world, it was justifiably considered by some to be the most outstanding piece of equipment to come out of the entire Senior Crown programme.
The ANS combined data from an inertial platform with a time datum supplied from a chronometer accurate to within 5ms. Position updating was achieved automatically by astrotracking six of the 52 most prominently visible stars at any time by day or night. The stars were computer catalogued in an ephemeris memory that could be continuously crosschecked for track position referencing. Scanning the celestial bodies sequentially, through a preprogrammed tracker mechanism mounted on a gimballed platform on top of the nav unit, the astro-tracker provided passive, refined location information.
When the autopilot was coupled to the ANS through the auto nav function switch, the aircraft could be flown automatically and precisely on a predetermined flight path. The preplanned route (worked out by highly experienced navigators) was electronically loaded via a Milar tape into the ANS’s computer memory a few hours prior to take off. Inflight modifications could be made by the RSO using his control and display panel. During flight, the computer-sequenced plan directed the aircraft from one destination point (DP) to the next.
Two further features of the ANS included ground reference position updating through the forward-looking view sight, and sensor monitoring by reference to control point (CP) actions. CPs were pre-determined track points, programmed into the ANS, that would activate or deactivate reconnaissance sensors and served to alert the RSO that the system was about to turn on, that the system had functioned correctly, and that programmed activities were being automatically carried out.
The ANS was backed up by a Singer-Kearfott SKN-2417 INS with a circular error probable (CEP) of 1nm per hour. This replaced the SR-71 gyro flight reference system (FRS) in June 1982. The earlier system had no navigation capability and provided heading and attitude information only.
At the heart of the new unit was an inertially-stabilized platform (four gimbals for all-attitude operation, vertical and azimuth gyros and three sub-miniature pendulous linear accelerometers). Located in the so-called R-bay, the gyro-platform provided navigation data to a control panel located in the RSO's cockpit."
 
Hood said:
The UK started early with star trackers, what were then termed ‘Auto-Astro’ systems.

Theoretical development work at the Telecommunications Research Establishment on Blue Sapphire which would be capable of tracking stars at night at the Telecommunications Research Establishment began around mid-1945 and flight trials of a prototype unit in a Vickers Viking began in December 1949. Blue Sapphire was intended for the NBS system of the future V-Bombers. Orange Tartan was a successor system using infra-red homing guidance technology to operate in daylight and be suitable for installation in either V-Bombers or guided-missiles. Technical problems with both systems saw work being abandoned in the mid-1950s. There is some evidence that Avro researched similar equipment during 1956-61, probably for missiles.
It's likely that any star tracker for Black Knight built upon the Orange Tartan work.
Click to expand...
The TRE-developed Blue Sapphire auto-astro system was never flown. An experimental model was prepared for trials in a Tudor but the project was cancelled in 1952 before it could be flown (due to accuracy and latitude limitations, and priority of other work). The RAE carried out supporting work which may account for the Viking trials. (Would be interested to learn more about those). Blue Sapphire was originally going to be day & night capable but ended up being night only. The day capability was split off as Orange Tartan. The Blue Sapphire/Orange Tartan work on tracking a bright point of light against a light background was picked-up by the IR homing folks and applied in the Green Thistle IR homing programme.
 
Meanwhile in 1949 the US Air Force flew a B29 with the first celestial-aided inertial navigation system called FEDE.

Dont ask me what FEBE stands for haven't been able to find it.

With it being one of the final test devices datung back to the 1930s. So eyeah the US also started early.

History Timeline | Draper

www.draper.com www.draper.com
 
yellowaster said:
The TRE-developed Blue Sapphire auto-astro system was never flown. An experimental model was prepared for trials in a Tudor but the project was cancelled in 1952 before it could be flown (due to accuracy and latitude limitations, and priority of other work). The RAE carried out supporting work which may account for the Viking trials. (Would be interested to learn more about those).
Click to expand...
Frustratingly I have lost the photos from Kew - bloody backup hard drive crash of a few years ago!

The reference was AVIA 6/16432 Blue Sapphire: development and preliminary flight trials of a latitude measuring system.
This file is: Royal Aircraft Establishment Report No. IAP.1458 Blue Sapphire: The Development and Preliminary Flight trials of a Latitude Measuring System, by G. E. Roberts, F. J. Twiney and R. L. B. Wall, August 1950

The notes made on the file from the photos I took are:
Servo system to direct the axis of a telescope at a star. Two systems examined; accelerometers on a star-stabilised platform made to monitor an automatic D.R. estimate of position; or star-stabilised system with integrated accelerometer control of an 84 minute period.

First system preferred for a manned aircraft as there is less restrictions on tactical freedom.

Experimental system ‘locked’ onto Polaris. Worst error 10’, probable error +/- 2.5°. Aircraft lost.

RAE Report IAP.985 looked at vertical determination for Blue Sapphire and suggested a Polaris vehicle.

Three Anschutz Type 1.0 gyroscopes for azimuth, N/S and E/W axis, required for rotational stabilisation, rigid mounting to platform. RAE telescope 1.1in diameter, 7.2in focal length, 1° look angle, mounted in centre of platform, using a periscope optical system.

Vertical system: vertical misalignment disregarded due to ½’ star alignment from rigid direction at star.
Monitor DR system: probably accuracy 4° with Mk.9 autopilot, available 6 minutes after locking. Tactical freedom to manoeuvre.
Barnes accelerometer.
Integrated accelerometer 84 minute system: set to local vertical and updated by aircraft accelerations, correction required for earth’s rotation. Only precise vertical is on the ground. Need a visible star on the ground and relies on star lock throughout the flight. Only 10-20% of nights in Europe meet this requirement and restrictions on turning flight. 1.5' accuracy requires position to be known to 1nm and 5kt speed. If star lock broken the system is only accurate for 2 minutes (10kt error).

17in diameter window gives 30° bank allowance. System fits in 14in diameter sphere weighing 241lb, requires 400W power.
Gee to be used to fix position every 5 minutes, auto observer used at 15 second intervals.

Vickers Viking VL228 flown by Flt. Lt. E. A. Clerk, first flight 8/12/49. Misting of window prevented lock. 28/2/50 flight, Polaris locked for 1hr 15min. 2/5/50 & 3/5/50 a double filter introduced, Mk.9 autopilot control gave better accuracy than manual flight. Probable accuracy under autopilot control +/- 2.5°.

Still need to settle method of wind compensation.
Click to expand...

So clearly some parts of the system - were test flown in 1949-50. It does feel like this was early days of development.
 
Firefinder said:
Meanwhile in 1949 the US Air Force flew a B29 with the first celestial-aided inertial navigation system called FEDE.

Dont ask me what FEBE stands for haven't been able to find it.

With it being one of the final test devices datung back to the 1930s. So eyeah the US also started early.

History Timeline | Draper

www.draper.com www.draper.com
Click to expand...

FEBE stands for Phoebus at least it's described so in numerous papers.
 
Hood said:
The reference was AVIA 6/16432 Blue Sapphire: development and preliminary flight trials of a latitude measuring system.
This file is: Royal Aircraft Establishment Report No. IAP.1458 Blue Sapphire: The Development and Preliminary Flight trials of a Latitude Measuring System, by G. E. Roberts, F. J. Twiney and R. L. B. Wall, August 1950
Click to expand...
Thanks for that. It's possible that the TRE and RAE co-operated on automatic astro. What is mystifying though is the lack of mention of the RAE work by the TRE - the only reference I can find is in October 1948: "RAE are developing a two-dimensional sighting head for flight experiments with the vertical seeking system. TRE are assisting in this work". It's almost as if there were two separate programmes.
 
The only TRE file I have on the subject is DSIR 23/17519 which is TRE Report T.2083 by S. Jones and L. Manns, dated December 1947.

By that time they had worked out a system but its clear from the report that several elements were still to be finalised - for example the sighting head, cameras and the intention to look at integrating it with a DR navigation system for when fixes were lost. It's clear the sighting head at not been built at that time (but they hoped to soon contract a company to do that), although two sets of the tracking equipment existed - one at the TRE and the other in use at Greenwich Observatory.
There is no mention of the RAE.

yellowaster said:
"RAE are developing a two-dimensional sighting head for flight experiments with the vertical seeking system. TRE are assisting in this work".
Click to expand...
This sounds like the system described in IAP.1458.

Which makes me wonder if "Experimental system ‘locked’ onto Polaris. Worst error 10’, probable error +/- 2.5°." mentioned in IAP.1458 was the original TRE system described in T.2083?

Allowing for the fact no complete system had been built by January 1948, and the RAE report being written in August 1950 it leaves a good 30-month window for the TRE and RAE to have built and tested their designs in sequence. Of course TRE might have handed over its work to RAE due to other pressures of work. If both TRE and RAE are claiming Blue Sapphire funding from the MoS then it signals that both were the same project and that the MoS was prepared to fund at least two streams of R&D. Collaboration seems more likely than competition (my suspicion is had they been separate programmes another Rainbow code would have been assigned).

@CJGibson has done far more rootling around at Kew on this topic that me and might know more.
 
I did, about 15 years ago. Interesting stuff, but I couldn't grasp the maths (it's a long time since I did such heid nippin' sums). I intended to explain how it all worked in Vulcan's Hammer but became bamboozled, so left it out.

Chris
 
CJGibson said:
I did, about 15 years ago. Interesting stuff, but I couldn't grasp the maths (it's a long time since I did such heid nippin' sums). I intended to explain how it all worked in Vulcan's Hammer but became bamboozled, so left it out.

Chris
Click to expand...
Piece'o'cake:

"The missile constantly knows its position, determining it by subtracting where it is from where it isn't or vice versa. This yields a deviation used by the guidance system to generate corrective commands. By driving the missile from its current position to where it wasn't, it arrives at where it was, making it the new position that it isn't. If the current position differs from where it wasn't, a variation occurs. This variation, the difference between the actual and non-existent positions, can be corrected if deemed significant. To calculate the deviation and its variation (error), the missile subtracts where it should be from where it wasn't or vice versa."
 

Similar threads

Back
Top Bottom